Page: 1 | 2 | 3
A chain is only as strong as the weakest link. If one part of something is weak, it jeopardizes the integrity, quality, or effectiveness of the whole. The idiom is applicable to almost all cases composed of more than one part. Let’s take a car, for example. Its main objective is to ensure that people or objects reach from point A to point B in a short and safe manner. It consists mainly of an engine, transmission, battery, alternator, radiator, front axle, front steering and suspension, brakes, muffler, tailpipe, fuel tank, rear axle, and rear suspension. No matter how powerful its engine is, an incorrect setting or malfunction of one of these parts will cause the car to run incorrectly or not start at all.
The main economic objectives of a business are profit earning, creating customers, acquiring and increasing market share, innovation, utilization of resources, leveraging productivity, etc. To achieve these objectives, a business needs a strong sales operation and a strong total experience (TX) structure that supports this operation. Total-experience is a business strategy that aims to create a better, holistic experience for everyone who engages with a brand (customers, employees, users, partners, etc.). No matter how good and powerful the product is, a problem that occurs in one of the links that make up the TX chain, for example, in the employee experience (EX), will disrupt the entire customer journey and will lead to a decrease in the company’s sales ratios. The statistics below confirm this:
- According to a recent report, disengaged employees cost UK companies around £340 billion each year.
- Organizations that score in the top 25 percent on employee experience report double the return on sales compared to organizations in the bottom quartile.
- Companies with highly engaged workforces outperform their peers by 147% in earnings per share.
If you want higher sales figures, faster revenue growth, and profitability you have to provide an employee experience that makes your sales team feel valued, driven, and engaged.
To Be Successful in CX, First Consider Improving Your EX
The leading actors of the customer experience scene are the employees of a business. Therefore, a business whose goal is better customer experience should also focus primarily on better employee experience. Employee experience can be defined as the sum of what people encounter and observe during their tenure in an organization.
There is no doubt that today, a significant part of employee-customer interactions takes place over contact centers. The metrics used for evaluation of a contact center such as average call abandonment rate, percentage of calls blocked, the average waiting time, the average speed of answer, average response time, first call resolution, occupancy rate, service level, customer satisfaction, etc. have a direct effect on the customer journey and are mostly related to employee experience.
Here clearly shows that for a business determined to strengthen the TX chain and in return boost sales, it is vital to monitor contact center activities wholly, to evaluate and analyze the monitored data accurately and to intervene and provide coaching timely. On the other hand, as we put in our previous posts, making it fulfilled with manual evaluation tools is impossible. Growing volume and the increasing importance of customer-agent interactions compels the implementation of conversational AI and Analytics solutions as the most efficient tool for monitoring, evaluating, analyzing, and coaching.
As a global supplier of various conversational solutions, Sestek offers a robust and AI-powered tool for analyzing agent performance as well. Sestek Agent Performance Analytics provides an automated solution to boost agent performance. This automated reporting interface generates evaluations on all recorded customer interactions, providing supervisors with crucial data to improve customer and agent experience. Using intelligent and customizable evaluation forms, advanced call filtering systems, and a user-friendly interface, the Sestek Agent Performance Analytics system will help reduce operational costs while improving call center performance.
The most recent success story of Sestek Agent Performance Analytics was with ING Turkey. ING Turkey, a part of ING Group, one of the largest financial institutions globally, was targeting to increase sales performance effectively to manage its call center operations with more than 200 agents. ING Turkey was searching for a solution to evaluate 100% of all interactions and effectively analyze them for actionable results to improve agent performance and increase sales revenue.
Using Agent Performance Analytics, ING Turkey monitored, evaluated, and analyzed 100% of all customer conversations and gained valuable insights to improve both customer experience and agent performance. The results were quite satisfactory:
- 9% increase in sales conversations and 25% increase in profit per agent:
Agents’ script adherence was monitored and analyzed. Detailed feedback and guidance were provided to improve the sales performance of agents.
- 20% decrease in customer complaint calls:
Possible complaint calls are detected by analysis with speech analytics, and the majority were resolved by the proactive complaint management system.
Sestek Agent Performance Analytics helped ING to effectively train agents, improve their performances and increase the sales revenue significantly. Click to check out our success story here.
Please visit this link for more details and case studies about Sestek Agent Performance Analytics.
Author: Çağrı Doğan, Sestek Marketing Team, Sestek
Publish Date: May 16, 2022
The customer experience (CX) refers to the interactions a business has with customers at every channel related to marketing, sales, and customer service. In general, customer experience is the sum of all interactions between a customer and a brand.
The customer experience is not just a series of transactions; It is also about emotions. What do current or potential customers feel or think about the brand? At each touchpoint, customers’ thoughts will be positively or negatively affected. So, there are important decisions to be made at each touchpoint, and these decisions will affect the experience and ultimately the success of businesses.
Why is Customer Experience So Important?
As products and services become similar and competition heats up, customer preferences are now based on experiences with the company rather than on specific product features and functions.
As customer experience becomes the main competitive edge, businesses must ensure that their customer experience strategy delivers personalized and satisfying interactions at every customer touchpoint. These interactions have an increasing impact on customers’ overall perception and impression of brands. This makes the customer experience critical to success.
The statistical insights, we’ve read from the article Top 40 Customer Experience Statistics To Know in 2021 on customer experience, say a lot about the significance of CX.
You can find a few of them below:
- 86% of professionals, engaged in or leading CX, expect to compete based on CX
- $35 billion is lost every year by US businesses in customer churn caused by avoidable CX issues.
- 83% of executives feel that unimproved CX presents them with considerable revenue and market share risks.
- 74% of consumers are at least somewhat likely to buy based on experiences alone.
In the light of all the figures, we can say that effectively listening to customers and acting fast to close the loop on their issues are becoming more and more vital for businesses. However, the high volume of customer interactions makes it impossible to manually review and analyze them accurately. The manual reviewing process can only focus on a fraction of interactions and is far from providing holistic results.
How to Overcome the Challenge: Speech Analytics
With the help of AI (artificial intelligence), businesses can apply in-depth analysis to customer interactions across multiple channels. These analyses include not only textual and statistical details but also emotional data. With advanced features like emotion detection and sentiment analysis, businesses can gain valuable insights to make wiser decisions.
Call centers are still at the forefront of customers’ preferred channels for interaction. Especially during the pandemic, the call traffic at this channel increased significantly. This shift calls for greater care in choosing the right tools to monitor, evaluate and analyze customer experiences.
Choosing the right speech analytics solution is very important for the correct evaluation of the vast data obtained through customer interactions. The success of the solution for speech analytics has a direct impact on the call center activities such as:
- Call center monitoring and reporting
- Agent performance analysis and training
- Customer satisfaction and service quality analysis
- Automation of quality monitoring processes
- Crisis management
- Competition, market, and campaign feedback analysis
- Identification of cross-sales opportunities
As a global technology company helping organizations with Conversational Solutions be data-driven, increase efficiency, and deliver better customer experiences; we provide a very competent and proficient tool for speech and text analysis. Recently, we deployed our speech analytics product for Halkbank, the third-largest bank in Turkey.
The analytics solution implemented for Halkbank’s call center monitors all (100%) customer-agent conversations and evaluates in these following parameters:
- Customer/Agent Anger Ratio
- Agent Speech Duration & Speed
- Agent Interruption/Overlap Rate
- Agent Silence Rate
- Agent Waiting/Hold Time During Call
This solution offers actionable insights to quality management teams to coach and guide agents accordingly to improve the service level. The implementation resulted from the inputs below:
- 25 seconds decrease in ticketed call duration
- 10% decrease in silent rates
- 5% decrease in hold time
- 75% improvement in customer anger ratio
The solution implemented for Halkbank also stands out with its high accuracy rate of 94% for speech-to-text conversion. This high accuracy rate enables quality management teams to discover customer pain points and frictions more accurately and it also provides deep insight on how to support agents for improved performance. Click to check out our success story here.
Please visit this link for more details and case studies about our solution for speech analytics.
Author: Çağrı Doğan, Sestek Marketing Team, Sestek
Publish Date: April 13, 2022
As in our previous post (From Single-Use Bots to Intelligent One-for-All Bots), the scope and scale of bots is expanding day by day. Furthermore, bots, whose capabilities have increased and diversified through artificial intelligence and machine learning, also facilitate the provision of more inclusive and accessible services for people of different ages, cultural backgrounds, disabilities, gender, temporary and permanent impairments, race and socioeconomic status.
There is no doubt that algorithms are one of the main actors shaping our present and future experiences. It has effects on our past experiences as well. An artificial intelligence application that successfully deciphers destroyed or missing Ancient Greek inscriptions and seems much more successful than human beings in this task (DeepMind AI beats humans at deciphering damaged ancient Greek tablets) is just one example among many others. It was also an artificial intelligence application that revealed that the work “Samson and Delilah”, which has been on display at the London National Gallery for 40 years, allegedly belonging to Peter Paul Rubens, is fake (Was famed Samson and Delilah really painted by Rubens? No, says AI).
As can be seen from these examples, algorithms are already more adept at distinguishing fake from real than the human beings who created them. On the other hand, bots equipped with superhuman abilities gain the ability to pretend to be human, and this enriches the possibilities of people to pretend and to show oneself different from what they really are. For example, machine learning and artificial intelligence technologies make it possible to clone a person’s voice and imitate their speech by making use of a person’s speech recordings. In this way, it is possible for people who have lost their voice or already passed away to continue to speak via their own voice, albeit through a digital interface. Similarly, an actor can voice different projects at the same time by cloning his own voice and increase his income (dubbing artist and actor Tim Heller explains that by cloning his voice, he can do many things simultaneously, such as animating cartoon characters, voicing books and documentaries, speaking in video games, and voicing in movie trailers). Similarly, it is possible for a singer to have his/her cloned voice sing in languages he/she does not speak.
What if behind the familiar and reliable voice on the phone there is a hand you wouldn’t want to shake?
Every new technique we develop makes us go beyond what we achieved previously, using much less effort and fewer resources. However, the impact of developments is not limited to our good and harmless abilities. The criminal world also (FBI says profits from cybercrime hit $3.5 billion in 2019) develops and diversifies its methods by making use of new techniques. The news about people being deceived or defrauded via phone or e-mail by strangers no longer surprises any of us. However, deepfake technology, which can make the cloned image or voice do the desired movement and say what is wanted, also increases the possibility of encountering a bot with the appearance and voice of someone we already know, and this sounds quite frightening.
If he/she is not nearby, hearing his/her voice over the phone can be considered the surest way to understand that we are communicating with someone we know. The incident of a bank manager is a case in point. In early 2020, a bank manager in Hong Kong received a call from a man whose voice he knew (he was the manager of a company he had spoken to before). The manager had good news: his company was about to make an acquisition, so the bank had to allow some money transfers of $35 million. A lawyer named Martin Zelner was hired to coordinate the procedures, and the bank manager could see emails from the manager and Zelner in his inbox and verify which amount of money was to be sent where. The bank manager started making the transfers, thinking everything looked legitimate. What he didn’t know was that he had been tricked as part of an elaborate scenario where the scammers used “deep fake” technology to clone the company executive’s speech.
The first recorded case, which is thought to have used the voice cloning technique, took place in March 2019. A fraudulent transfer of 220,000 EUR (Related article from The Wall Street Journal) was requested in an incident where scammers used AI-based software to imitate the voice of a CEO, and what cybercriminal experts described as an unusual case of AI used in hacking.
It may be frightening that criminal world actors come up with more sophisticated and complicated scenarios by taking advantage of new technologies, but it is not a healthy and logical reaction to refrain from using these technologies. It is highly probable that we would still be living in caves if our ancestors had only focused on how unsafe the life outside was and had not thought of developing appropriate protective methods and tools. On the other hand, it is impossible for us to protect personal computers against external attacks by examining the files one by one; nor does it seem possible for us to protect ourselves against increasingly sophisticated fraudulent methods by being more vigilant or by continuing to use outdated security methods. For instance, if a multi-factor authentication system or a voice biometrics solution equipped with passive authentication and effective fraud prevention capabilities, capable of operating independently of language and accent, and of distinguishing whether the voice is reproduced by digital means, were used in the examples above, it would be highly likely that the cases would have been prevented from producing undesirable results.
In fact, voice, like fingerprints or iris, is a uniquely human trait. This paves the way for voice to be used as a powerful authentication tool. Unlike PINs, passwords and answers to challenge questions, voice biometrics can’t be compromised without the knowledge of the voice’s owner. This is one of the factors that makes voice verification much more secure. However, it is not possible to do such an analysis by manual listening. The conversational biometrics solutions we developed as Sestek analyze the voice based on over 100 parameters. The solutions in question have playback manipulation detection functionality. This means that the solution will detect whether the party on the other end of the phone is actually speaking or is playing a voice sample. When a recorded voice sample is played, the technology can detect and report the situation by using the synthetic voice detection feature. The system, which can detect even known fraudsters with its biometric blacklist detection feature, uses a voice change detection algorithm to determine if a user’s voice has changed during a conversation. ING Turkey, one of our customers using this technology, shortened the average call time by 19 seconds for calls requiring identity verification. Thanks to this saving, they reduced operational costs and increased customer and representative satisfaction.
For details on our conversational solutions, please visit
Publish Date: December 20, 2021
The software which performs repetitive, automated and predefined tasks is named as “bot”. When we hear the word “bot”, it is highly likely that most of us first think of chatbots. Probably the first thing that comes to mind when we say ‘chat’ is the correspondence through internet applications. However, we all know that every healthy individual belonging to the human family can be an actor of chat action by signing or speaking and later, on writing.
Yes, we, as a member of the human family, have been doing this for so long that we can say chat is one of our unique and natural features. As it is one of the common features of the entire human family, we don’t need something like “chatman” to name any group in this family. Undoubtedly, in more and more situations in our daily life, we come across conversational algorithms or chatbots more often; for example, when buying plane tickets or clothes from an online store. Although chatbots are now appearing in almost every aspect of our daily lives, the history of bots in this field is not very long. It was generally implemented with the aim of making users feel that they are chatting with a real person; a computer, program, algorithm or artificial intelligence that enters into a dialogue with a person is called a chatbot. The surge in the use of chatbots, whether simple or more advanced artificial intelligence applied, has been driven by the enormous expansion of the Internet and especially social networking sites.
One of the oldest and best-known chatbots is a program called Eliza created by the Artificial Intelligence Lab at MIT between 1964-1966. After that date, we see many chatbot projects implemented as a result of research and development studies. However, they can be used publicly and globally in the 2000s with the widespread use of the Internet. Today, a significant part of customer service is carried out through chatbots. Business Insider estimates that the global chatbot market, with a volume of $2.6 billion in 2019, is expected to reach $9.4 billion in 2024, with a compound annual growth rate of 29.7%. The report also suggests that the highest growth in the chatbot implementation will be in the retail and e-commerce sectors, driven by the increasing demand for seamless omnichannel experiences.
Today, simple chatbots, whose capabilities are limited to answering FAQs in a certain field, are being replaced by chatbots powered by artificial intelligence, machine learning, natural language processing, speech recognition and analysis, speech synthesis and voice biometrics technologies. In this way, multifunctional/capable chatbots can be developed that can express themselves more humanely, understand various expressions implied by different styles and terms, and adapt the language they use to the style of the people they are communicating with.
We are now in a world where chatbots are the main actors in the field of corporate communication. Their use not only helps organizations improve their own potential, but also means a more enhanced user experience. With bots powered by technologies such as Speech Recognition, Speech Synthesis, Voice Biometrics, Natural Language Processing, Artificial Intelligence, voice and text analysis, Sestek, where I used to work on different projects for many years, can easily design customizable chatbots for the needs of organizations. Based on Sestek’s 21 years of conversational technologies development experience; with the solutions currently being used by 275 institutions in 20 countries; self-service rates are increasing significantly and the workload and stress on the agents are relieved. For example, according to September 2021 data, 250,000 customer interactions have been carried out via the Whatsapp bot utilized by Hepsiburada, Turkey’s leading e-commerce platform. With this application, 14% increase in self-service rates and 10% reduction in contact time with live representatives can be achieved. You can watch TR AI Week session titled “The Magnificent Duo: Artificial Intelligence and Self Service” by Taner Timirci (Hepsiburada, Chief Operations Officer) and Selin Özbalmumcu (Sestek, Sales Manager for Turkey) for details.
Another milestone project is the virtual assistant named Selim, which developed by Sestek for KUVEYT TURK and used by their 2.4 million customers in 12 months, has reached a success worth mentioning in recognizing what is said (intent recognition rate) with an accuracy rate of 97%. Selim was able to answer 6.2 million questions from 2.4 million users in 12 months, and the rate of connecting KUVEYT TURK customers to live representatives decreases by 29%. For details, you can review the case study titled (ACCELERATING DIGITAL TRANSFORMATION WITH VIRTUAL ASSISTANTS).
At the end of the day, the digital world continues to flourish, with the growing data flow rates and the increasing amount of data we collect in the digital network. With the advancement in artificial intelligence and machine learning, bots are perfecting their ability to process the data in question. Considering that bot assistants can make a reservation for a hairdresser or in a restaurant without making the person they are talking feel they are communicating with a bot, (Google Assistant calling a restaurant for reservation), we can say that we are confidently moving towards a future where a bot, which is initially intended to be used in a certain area, can be easily adapted to work in different areas as its processing power, knowledge and capabilities improve. In a nutshell, bots are expanding the boundaries we set for them day by day. One can surmise that in near future we will be able to define a bot as “a tool that is able to chat as one of its unique and natural features”, and there will no longer be a need for a special name like “chatbot”. Moreover, in the not-too-distant future, thanks to the “unsupervised learning” capability, which we have already started talking about a lot, for example, a bot initially programmed for customer relations will be able to train itself with the knowledge it receives in that field and can serve beyond its original purpose such as a sales assistant bot or HR bot when needed.
Author: Çağrı Doğan, Accessible Products Consultant, Sestek
Introduction to Chatbot Design
Building The Right Chatbot To Ensure Better Customer Experiences.
Today, chatbots are a significant part of customer services. From reserving a ticket to completing a banking transaction, they can offer any service that a live agent can. But being able to do these tasks might not be enough for enhanced customer service. Because customers not only expect high-quality services but also personalized experiences. According to recent research, more than 63% of customers expect personalization as a standard of service. So, chatbots need to answer this need to ensure an enhanced customer experience. Why do chatbots need personalization? Today’s customers seek more than functionality. They expect authentic connections with businesses. And the only way to build a strong and emotional bond between customers and chatbots is personalization.
Publish Date: November 11, 2021
Conversational technologies have been on our agenda for a long time, and these technologies are expressed with various concepts such as bots, virtual assistants, digital assistants, chatbots, etc. Is it really necessary to have so many terms that we witness new ones being added every day? Are these increasingly confusing concepts that important?
As someone who has spent 20 years in the technology field and believes that communication is a multi-channel experience, I have never liked the term “chatbot” since I first heard it. If we look at the history of this technology, which has become increasingly popular since 2016 and presented as if it were a new discovery; we can see that its foundations were laid in 1950 with the discovery of the “Turing Test” by a world-renowned mathematics professor named Alan Turing. The Turing Test determines whether a machine can demonstrate human intelligence. If a machine can engage in a conversation with a human without being detected as a machine, this means it demonstrates human intelligence.
The Evolution of Customer Service Automation
In fact, in the process triggered by the Industrial Revolution, the basic expectation of human beings from technology was the more effective use of machines in many fields that require manpower. The increasing use of machinery in the customer service area, which we, as Sestek, also focus on, has often resulted in customer dissatisfaction, especially in the first examples in the past.
Primitive versions of voice recognition, bots, and similar technological solutions started to be used in the 1990s. Although they provided cost advantages to companies, they were insufficient in terms of customer experience and therefore led to customer dissatisfaction. In recent years, rapid development in AI technologies such as deep learning, machine learning, etc., increased the success of such technologies positioned in the customer service automation field.
The Real Matter is the Right Experience Design
Bot, chatbot, virtual assistant, digital assistant… Whatever the name given to technology, the basis of a successful self-service solution lies in the correct design of the customer experience. Instead of only cost-oriented designs, bots that offer human-like conversations -except reflecting their emotions as humans do- and provide 24/7 service have entered our lives.
I’m sure that some people reading this piece will say, “When we call or text banks, we don’t want technologies that don’t make us talk to real people.”. But believe me, those who deal with a properly designed system integrated with the right technologies, within six months at the latest, report a better satisfaction result than their previous experience.
In this regard, I would like to explain with an example why technology has gained an unstoppable speed. You may think it would be more practical for a 70-80-year-old user to tell customer service about his problems with a live representative. But, let’s imagine that the virtual assistant provides personalized service to that person as soon as he calls customer service, understands everything he says, and performs his transactions five times faster than in the past. In such an experience, both the service recipient and the service provider will be happy.
An Undeniable Need: “Omnichannel”
We have seen in the pandemic that our service demands have almost completely moved to the digital environment, and our communication experience has changed irreversibly. One of the most important issues when positioning bots here is to offer an “omnichannel,” that is, a “single experience” independent of the channel. Namely, you get service from an e-commerce company, and when you contact customer service, you expect it to give you the same experience from every channel. When you call the contact center, submit a support ticket on a website, or communicate with WhatsApp or a voice assistant, you have only one expectation: You want your request or problem to be understood and answered accurately as soon as possible. At this point, there are many solutions that look like “chatbots” at first glance but can’t actually offer this experience. These solutions, which work independently of other customer channels, remain unaware of the dialog initiated by the customer in a different channel. This means that the customer has to express herself again and again in every channel, thus causing loss of time and customer dissatisfaction.
Whether it is called “chatbot,” “virtual assistant,” or a different term, these technologies, which are at the center of customer service automation today, are based on the principle that users receive service by communicating with systems. When designed correctly, these technologies enable transactions in a much shorter time for users. With the automation they provide, they offer cost advantages and efficiency to businesses. Regardless of the purpose of use, these technologies need to be developed based on customer needs and introduced to customers in a way that best meets their expectations. In other words, the customer is at the center of the business, and only the work carried out based on this fact can be successful.
Publish Date: September 21, 2021
Once a scene in science fiction movies, artificial intelligence is a natural part of our daily lives today. When we open a film from Netflix or follow a profile on Instagram and get movie or profile suggestions in just as we want, not to mention Amazon’s real-time pricing or Google search’s semantic ability.
Companies invest in their talents and technologies more and more every day to get the best performance, precision, and accuracy from their algorithms to provide the information that users are hoping to find.
According to Ritu Jyoti, the Vice President for AI Research at International Data Corporation (IDC), AI has become the top topic of all industries for its versatile application areas and resilience - and pandemic only magnified the effect. The same report mentions that the expected growth in revenues for AI solutions will reach $500 billion by 2024.
AI is known for its “black-box” nature, lack of transparency while offering almost limitless possibilities for developers and scientists who train AI decision systems based on a specific domain with providing no visibility or rationale behind it. This can be negligible when we talk about getting movie suggestions from Netflix, but if this system is used for disease diagnostic, or sentence a criminal subject, the crucial need for explaining AI is inevitable. Moreover, today’s industries and governments require these technologies to make sure that users or customers trust the AI-based systems when making decisions; they have the right choice with the help of AI. Therefore, finding an approach and developing algorithms that transform black-box systems to glass-box systems are significant. It will be possible for people to trust machines on their decisions by understanding how they think and why they choose what they choose. This is where explainable AI comes into play.
What is Explainable AI?
Explainable AI (XAI) is a suite of machine learning techniques that produce more explainable models. The main goal of XAI is to explain how algorithms come up with a decision and which factors affected their decision points and eventually ended up with that solution. It is an emerging field that sits at the intersection of different focuses: transparency, causality, bias, fairness, and safety.
XAI progress has affected by three accelerating factors:
- Increasing ethical concerns. The growing need for transparency, required by laws like GDPR about how personal data is used.
- As explained with examples in the introduction, before putting trust in machines’ decisions, humans need to be convinced. And that would be only possible with the explainability of AI systems.
- Better human-machine synergy. Machines are part of our daily lives more than ever and enhanced our life with their wide range of functionality and increased intelligence. So, it is important to create an environment where both humans and machines are working together.
XAI was first introduced in 2004 by Van Let et al. to explain a game simulation that they developed to train the US army. Although it is a game, it is especially designed to better train the soldiers without losing the effectiveness of the education material. Full Spectrum Command consists of users (soldiers) and non-player characters that AI controls. After a mission is done, users can click on subordinates and ask questions within the system or review key aspects presented by the case. This study also significantly influenced the gaming industry and its technologies.
Understanding Explainable AI
A simple illustration is shared below to understand the main concept of XAI. There are few highlights worth mentioning. First, today’s section of the picture represents how the classical mechanism of ML works. The system gets training data, applies ML processes, and concludes or makes a recommendation according to how the model is trained. It is also important to notice that there is one-way interaction with the system’s user; this shows no explainability in this system. The user only sees the final decision made by the system.
On the other hand, there are explainable models and explainable interface layers in the XAI concept instead of the learned function. Explainable models are responsible for taking the task and offering a recommendation or an action; the interface is responsible for justifying the cause of why the system has made that decision. Then the user makes a final decision based on the explanation, so there is a two-way interaction between the system and the user. Explainability interfaces can benefit from Human-Computer Interaction (HCI) techniques to generate effective explanations.
Figure 1: Comparison of AI and XAI concepts proposed by DARPA
Some Application Areas
Explainable AI has a wide variety of application areas, including healthcare, finance, insurance, law, etc. Here healthcare and law case studies are briefly shared to give a holistic overview of how XAI contributes to industries.
Explaining Medical Diagnosis
While describing XAI, it is mentioned that XAI is deployed to convert a black box into a white box. No matter how advanced the decision model is, there is always a need for a human in the loop for approving that model gives the right decision and handles unexpected scenarios. The medical domain is a suitable topic for describing such cases. Convolutional neural networks used to interpret medical diagnostics by using computer-aided diagnostic systems. Patient data collected through Magnetic resonance imaging (MRI) and computed tomography (CT) scans and existing diagnostics archives were processed, and the model trained to make accurate diagnostics by looking at the patient’s scans. In this way, the research team trained the model to identify disease patterns by looking at the existing atlas.
XAI in Legal
Explainable AI has great potential in legal applications. Courts can benefit from XAI for its pragmatic and transparent approach like judicial reasoning, bottom-up approach while making decisions, case-by-case considerations of delicate cases, and even stimulating the paperwork required for legal settings and audiences. It is also believed that XAI helps law to become more transparent and become independent from private actors by becoming more publicized. Judge is the main consumer of XAI algorithms recommendations and decisions. Taking reasonable explanations for sentences, the likelihood of that crime will occur, due processes, etc. require explanations from XAI models. As real-life cases, data collected, and model is trained by the collected data; common law of XAI will be created. This law can be used to compensate the explanation requirements by criminal, civil, administrative law settings, or it can be used by judges, juries, defendants, etc.
Explainable AI emerged as an answer to the increasing need to understand machine learning and AI better. Being able to comprehend AI will help us build better human-machine collaboration and contribute to a transparent approach where ethical concerns and trust issues can be resolved easily.
Author: Gülşah Keskin, Product Analyst, Sestek
Publish Date: July 13, 2021
Conversational AI has become the driving force behind digitalization projects. Businesses use this technology to automate customer-facing touchpoints on any channel. Conversational AI reduces costs and increases efficiency by automating repetitive tasks and allowing human agents to focus on more crucial tasks. Besides, enabling customers to engage with technology in a much more natural way ensures an enhanced experience.
One of the biggest challenges that companies face when building or buying a conversational solution is intelligence. An intelligent system can offer a human-like conversation and understand multiple ways in which the same information is being phrased. The intelligence of a conversational AI solution relies heavily on the design of the dialog flow. This flow is the brain of the solution. But also, as important as the brain, is the user experience. Here are four hints to design a smart conversational AI solution.
1. Know Your Audience
To understand your customers’ expectations, you need to know where they are coming from. Try to define your audience by considering their demographics, such as age, gender, profession, geography. Try to answer these questions: Who are they? Students or employees? Which age group(s) do they belong to? What are their habits? How about their language and tone? What kind of sentences would they use? Would they prefer short and direct communications, or would they enjoy longer conversations? Be familiar with the way they speak; the phrases and slang they use. Also, consider their preferences and habits. Knowing all these details helps you design conversations that your customers can easily engage with and be happy with the result.
2. Intent Recognition is Vital
Offering a system that answers simple FAQs is not enough for today’s customers. Customers want to say simply what they want and to be understood by the systems. They don’t want to lose time with detailed queries. So, businesses need to offer solutions that not only understand what customers say but also understand what they mean. And this is possible with the intent recognition technology. This feature understands the meaning behind customer queries with high accuracy. If a query is ambiguous, the AI will ask additional questions to make sure. This results in a human-like dialog between customers and machines.
3. The Heart of The Conversation: Dialog Flow
To ensure a natural and smooth dialog, you should build conversations that sound more human and less machine. While building your dialog flow, focus on language details. Consider your audience, the language, and the tone they use every day, and build your dialog flow accordingly.
Make sure that you keep the conversation short by only asking the necessary questions. Keep the prompts short, and don’t confuse your customers by offering multiple options at once. Be concise. Don’t reply with ten lines of information when two will do. Never forget that customers might change their minds during a dialog and ask for something totally different from what they had initially asked for. You should be ready to interpret these changes and instantly adapt to them.
4. Respect Your Customers
Customer satisfaction must be at the center of your dialog design. Customers always prefer to interact in their way. So, you shouldn’t force them to engage in your standard format. Let them be free to choose. When engaging they can use formal language or everyday language. And your AI solution should be able to adapt. This is possible with NLP-based conversational solutions.
Time is the most valuable asset. Respect your customers’ time. Offer solutions that smoothly integrate with different channels. By doing so, you can save them from repeating themselves whenever they change their engagement channel. Build a system that can pick up the dialog from the channel your customer left off. In short, put yourself in your customers’ shoes and design conversational AI solutions that you would enjoy interacting with.
To learn more about designing smarter conversational AI solutions, download our “Conversational AI E-book” by filling the form below.
Author: Çağrı Doğan, Accessible Products Consultant, Sestek
Today’s customers expect smooth journeys.
They want to interact with brands easily, at any time, at any channel; contact centers, chatbots, messaging apps, smart assistants. And while doing this, they expect to be understood fast. They want to be understood before they open their mouth. They want to be understood not only by humans (customer reps) but also by machines. The answer to this expectation is Conversational AI.
Publish Date: May 25, 2021
In the past few years, advances in artificial intelligence led to the widespread use of Conversational AI. The rise of the technology continues thanks to its successful use cases in both consumer and enterprise applications. According to Research & Markets, the Conversational AI market generated $3 billion and is predicted to reach $15 billion in 2024, advancing at a 30% CAGR.
The Rise of Conversational AI
The rising demand for AI-powered customer support services, positive return on investment (ROI) for companies deploying Conversational AI solutions, and an increasing number of solution providers in the market are effective in this growth. So, the adoption of AI in the enterprise sector is increasing. According to Gartner, 31% of CIOs have already deployed conversational platforms, representing a 48% year-over-year growth in interest. Conversational AI is implemented across various use cases, including customer service, sales support, human relations, employee engagement, customer engagement, retention, and more.
What does Conversational AI Offer?
Today’s customers expect smooth journeys. They want to interact with brands easily, at any time, at any channel; contact centers, chatbots, messaging apps, smart assistants. And while doing this, they expect to be understood fast. They want to be understood before they open their mouth. They want to be understood not only by humans (customer reps) but also by machines. The answer to this expectation is Conversational AI.
Natural Human-Machine Interaction
Combining technologies like natural language processing (NLP), speech recognition, and text-to-speech, Conversational AI enables smooth interaction between customers and machines. The technology allows customers to naturally interact with systems in their own words via speech or writing. Conversational AI provides a personalized and enhanced experience for customers. Customers can complete various tasks simply by speaking to systems as if they are speaking to a human.
Reducing Costs and Enhancing Experience
Keeping costs minimum while offering high-quality customer service is among the biggest challenges that businesses face. Conversational AI automates routine customer service tasks by allowing customers to self-serve. This helps companies reduce operational costs while increasing efficiency. Offering enhanced customer service also provides an effective differentiation tool for businesses. Conversational AI leads to higher customer satisfaction and greater customer loyalty. This means a sustainable competitive advantage and a positive brand perception from customers.
3 Steps of Conversational AI Deployment
Deploying Conversational AI for the sake of “everybody else is doing it” might be the worst thing you can do for your business. Boston Consulting Group’s latest study shows approximately 70% of organizations fail in their attempts for digital transformation. You will need a well-thought strategy before you take any action. Following the steps below might help you build and implement a result-oriented conversational AI strategy.
Step 1: Set your end goal
So, you are not implementing Conversational AI to jump on the bandwagon. Then, try to discuss within your company (within your team) the following questions:
⦁ What do we want to achieve with implementing AI? What is our end goal?
⦁ How can AI serve our business objectives?
⦁ What are the main pain points of our customers that we think AI can help solve?
⦁ How will this solution help them?
⦁ How can we set up KPIs to monitor progress?
Step 2: Select the right vendor.
Developing AI solutions within your company will take a serious amount of time and effort. When there are AI vendors working on these solutions for more than decades, it would be wise to get some outside help.
But choosing the right vendor is important. While deciding on the technology provider, make sure that they have the following capabilities:
⦁ Technology and industry-specific expertise
⦁ UX-oriented approach
⦁ Competence in professional services
Step 3: Phase the plan
⦁ Bringing together your team with your technology provider’s team to determine requirements.
⦁ Prepare checklists on specifications, installation requirements, and KPIs beforehand.
⦁ Testing technology specifications to see if specifications are implementable in practice.
⦁ Launch internally before offering it to your customers to complete user and security testing and apply necessary fixes on time.
⦁ Now your project is live, and your customers can start interacting with your solution.
⦁ Monitor customer behavior and get as much feedback as possible to detect improvement needs.
⦁ The success of any project depends on objective performance evaluation.
⦁ Continuously monitor and analyze your efforts to measure the effectiveness of the solution and define your next steps for improvement.
⦁ You can use Conversational Analytics tools such as Speech, IVR, and Bot Analytics for an in-depth evaluation.
To learn more about leveraging self-service automation and enhancing the customer experience with Conversational AI technologies, download our “The Conversational AI Playbook” by filling the form below.
Publish Date: April 1, 2021
From mobile devices to smart homes and websites to virtual assistants, conversational platforms are everywhere we touch. By using voice as the most natural form of interaction, conversational-AI transforms any platform into a helpful assistant.
The use of voice-activated digital assistants is increasingly becoming common in cars as well.
According to recent research by Market Insight Reports, AI in the automotive market is expected to be appraised at USD 12 billion by 2026.
The Transformation of Infotainment Systems
Before the proliferation of conversational systems, infotainment systems were popular in-car systems.
The in-vehicle infotainment system had its origin in the 1930s, but the first-ever car radio, named ‘Motorola’, was introduced in 1950. After several advancements in the automotive industry, during 1970−1977 automotive cassette tape player was introduced. The integrated GPS navigation system was introduced by Toyota in 1987, followed by other players in the following years.
In the late 1990s, remote diagnostics came into the picture and after 2003, vehicle health reports became an inclusive part of connected car services. In the late 1990s, smartphone technology also evolved and around 2004−2006 smartphone connectivity for in-vehicle infotainment was introduced. By the end of the decade, alternates for in-vehicle smartphone usage, such as large display screen that includes services like audio, visual, e-mail, vehicle diagnostics, navigation and compatibility of mobile apps came into the picture. After the 2010s, we started to see more voice-activated systems in cars. The rise of voice-based assistants also accelerated the adoption of conversational systems in vehicles.
What does In-Car Conversational AI offer?
The convenience of conversational systems in cars is undeniable. Letting the drivers operate all in-car systems by voice enhances the driving experience and increases security by minimizing distraction.
Enhancing Security with Conversational AI
Driver distraction is an important road safety issue. National Highway Traffic Safety Administration (NHTSA) estimates that in 25% of accidents in the US driver distraction is the main reason for an accident. This means just in the US, 1.2 million incidents each year happen because of driver distraction.
Modern cars with advanced infotainment systems often need more cognitive attention, causing more distraction. That is why researchers are searching for better ways to manage distraction by improving the interaction between the car and the driver. And conversational AI technology offers an effective solution. The technology allows hands-free interaction via natural speech. Advancements in speech recognition and natural language processing technologies enable an ongoing conversation between the driver and the user. This ensures an uninterrupted driving experience, which increases security.
Enhancing Driving Experience
In-car conversational AI applications enable users to interact with voice, the most natural interface. The hands-free nature of this technology provides a convenient experience for drivers. So, drivers can accomplish various tasks without taking their hands off the wheel. This makes voice-enabled assistants more of a must-to-have than a nice-to-have for cars.
Conversational AI transforms current in-car infotainment systems into easy-to-interact digital assistants. By speaking to these systems, drivers can accomplish various tasks and have a fun driving experience: Making a phone call, receiving navigation directions, sending a text or email, learning the weather forecast, and so on. Drivers can do all these things only by speaking to their in-car assistant.
Implementing the Technology
As the adoption of in-car conversational AI rises, more companies will be offering this technology as a given service for their customers. That is why offering these technologies alone will not be enough to differentiate from the competition. As more brands provide such technologies, they will need to find new ways to differentiate from their rivals. Here is a checklist for automotive brands to offer an effective in-car conversational system:
- Set Your End Goal
Deploying Conversational AI for the sake of “everybody else is doing it” might be the
worst thing you can do for your business. You will need a well-thought strategy before you take any action. To draw a roadmap, you need to set your goal first. Then, try to discuss within your team
the following questions:
- What do we want to achieve with implementing conversational AI? What is our end goal?
- How can conversational AI enhance driver experience?
- What are the main pain points of drivers that we think AI can help solve?
- How will this solution help them?
- How can we set up KPIs to monitor progress?
- Select the right vendor
Implementing conversational AI solutions is a serious decision that requires a serious amount of time and effort. Collaborating with an expert vendor would make this process easy and seamless. So, while deciding on the technology partner, you’ll work with look for the following capabilities:
- Expertise on NLP: The performance of a conversational AI system depends on the NLP engine. The language, the linguistics of phonetic spelling, dialects, cultural nuances, and domain-specific terminology determine the effectiveness of NLP engine. So, make sure that you’re working with a vendor who has experience in these areas.
- UX-oriented approach: UX is all about how a product or solution fits user expectations. In other words, the success of conversational AI depends on its ability to provide a great UX, which is directly related to dialog design skills. To ensure a natural and smooth dialog, you should build conversations that sound more human and less machine. This requires expertise not only in linguistics but also in contextual capabilities. So, your technology provider should have experience in designing smarter dialogs and eventually, smarter systems.
- Competence in professional services: Conversational AI projects require a rigorous approach. Continuous monitoring and improvement are necessary to ensure an enhanced driver experience. While selecting your technology provider, consider their capabilities in professional services, including customizations, training, implementation, and post-implementation support. Make sure that your technology provider understands your motivation and offers a project management approach accordingly.
- Phase the Plan
- Prepare: Determine the requirements by bringing together your team and your technology provider’s team. Prepare checklists on specifications, installation requirements and KPIs beforehand.
- Test: Test technology specifications to see if they are implementable in practice. Apply as many internal tests as possible before offering the technology to your customers. So, you can complete user and security testing and apply necessary fixes on time.
- Monitor: After your project goes live, monitor driver behavior and get as much feedback as possible. These will help you to determine what you need to do to improve UX.
- Evaluate: The success of any project depends on objective performance evaluation. This requires continuous monitoring and analysis. Conversational Analytics tools can help you measure the effectiveness of the solution and guide you through your next steps for improvement.
Sestek and In-Car AI
At Sestek, we offer omnichannel conversational AI technology with a wide range of use across multiple channels, including voice IVRs, chatbots, virtual assistants, and intelligent platforms. Lately, we collaborated with TOFAŞ, the leading automotive manufacturer in Turkey. Together we will develop an in-car voice assistant, which will be a first in the industry. Our Conversational AI technology will enable a dialog between the driver and the virtual assistant. Drivers will be able to interact with the assistant by natural speech. The assistant will be able to make sense of what is said and return with the necessary answers, and when it needs additional information, it will be able to request detailed information by asking various questions to the driver.
The assistant will provide route and road status information, recommendations specific to the driving characteristics of the user, and support driving safety with instant verbal warnings. In this way, a more advanced driving experience will be possible in terms of safety, convenience, and comfort. The most remarkable difference of the application from known virtual assistants is that it can analyze many instant data to be taken from the car. This continuous feedback and analysis will be used to support the driver. To learn more about this project, please click here.
Author: Çağrı Doğan, Accessible Products Consultant, Sestek
Publish Date: March 9, 2021
Conversational technologies transform the customer journey. By allowing customers to use their own words to interact with systems, conversational technologies offer the most natural communication method. And the conversational journey starts with speech recognition technology.
Speech Recognition (SR), also known as automatic speech recognition (ASR), catches spoken words and phrases and converts them to a machine-readable format. This is the first step to let users control devices and systems by speaking instead of using conventional tools such as keystrokes or buttons.
Why is Speech Recognition important?
As the first step, the accuracy of speech recognition is key to a successful conversational journey. If you cannot accurately translate voice into text, you cannot understand what your customers are saying, and you will not be able to solve their problems. The accuracy of SR increases the efficiency of self-service applications and allows companies to deliver improved customer experiences. Since SR is the core technology that empowers conversational solutions, the success of a conversational system depends on the capabilities of its SR technology. In other words, to ensure a smooth conversation between machines and the customers, a comprehensive Speech Recognition solution is crucial.
To offer an effective conversational product, make sure that your SR solution ;
- has a high recognition accuracy
- offers advanced natural language support
- supports multiple languages and accents
- easily integrates with multiple technologies like AI, natural language processing (NLP), and machine learning (ML)
- has a flexible structure that supports omnichannel deployment
How Sestek SR stands out
20 Years of Know-How
Sestek SR is the product of Sestek’s 20 years of experience in building highly accurate speech solutions. Since day one, we have been working hard to make our technology more accurate and robust. Empowering Sestek Speech Recognition with the latest technologies like neural network (NN) improves its recognition accuracy and as an R&D company, we have been investing in this for a long time.
End-to-end Conversational Journey
Sestek SR is the core technology behind our main products such as voice IVRs, virtual assistants, and conversational analytics. Moreover, Sestek SR is a component of our omnichannel automation solutions. Meaning when you implement Sestek SR once, you can benefit from the technology at any channel you are willing to build conversational solutions for your customers.
Tailor-Made for Different Verticals
Each business has different priorities when it comes to offering the best customer service. Each business needs specific solutions rather than one-size-fits-all ones to build the right conversational journey.
Sestek Speech Recognition’s highly customizable structure enables us to build a tailor-made conversational solution for each company. The technology can be trained with specific language models according to industry and vertical needs.
Difficult to Build Difficult to Implement
Building highly accurate speech solutions in house might take significant time and effort. Collaborating with experienced vendors saves more than money, it can contribute to the awareness within your organization. But this requires a close relationship with your technology provider. Your technology provider needs to understand your needs fast and offer intelligent guidance with proven processes and advanced tools. Sestek offers end-to-end professional services, including strategy building, application design, deployment, testing, and optimization. Our team’s expertise relies on hands-on experience in speech tech, gained from 20 years of developing conversational solutions. This may be our most significant differentiator to our global competitors’ deploy and forget approach.
SR Accuracy Test
Sestek SR is the product of our continuous R&D efforts. We optimize our product with the latest technologies and methods in a way that increases recognition accuracy.
Lately, we developed a new model where we used a neural network as a technological leap. And to measure the success of this model, we tested the accuracy of our speech-to-text engine. We compared our engine with Google and IBM’s SR engines.
For manual testing, we used two sets of random data from call center recordings, two sets of recordings of medical articles. For automated testing, we used 3 YouTube videos.
In the manual test, recordings were listened to and labeled all the automatic transcribed words/phrases as correct/wrong and calculated final word-error rates within the data set. WER (word-error-rate) is a common metric for SR engines; it is the ratio of the total word of error (substitutions, deletions, and insertions) to the total number of words in the reference. The smaller the ratio, the more accurate the engine.
The first table shows the results of manual calculation, and the second one shows the result calculated automatically using the reference text. Here are the results:
As seen above, our NEW approach provides nearly 30% improvement for accuracy.
With these numbers, we are not suggesting that we are certainly better or the rest is certainly worse. The speech recognition process includes calculating and optimizing millions of parameters over a vast search space, and it is hugely stochastic (what we engineers call as the pattern that may be analyzed statistically but may not be predicted precisely). A vendor’s SR engine can perform better than others for a specific recording, but the same engine can perform worse for another one.
We are simply suggesting that our SR technology can easily compete with billion-dollar vendors such as Google and IBM.
Speech recognition is among the leading technologies used in conversational automation. The performance of this technology plays a crucial role in the success of conversational customer services. By offering an easy-to-use and advanced conversational system, businesses can improve customer experience. That is why choosing the right speech recognition technology is an important decision to make. Sestek offers an effective solution not only with its advanced technical features and high accuracy rates but also with 20 years of know-how and distinctive professional services. Click here to test our Speech Recognition technology for the following languages; Turkish, English, Flemish, French, Russian, and Turkish.
Publish Date: October 10, 2020
Customer satisfaction is the key factor behind the success of a business. The more satisfied a customer is, the higher the chances they become loyal customers. This means they will stay with your brand and spend more than others. Therefore, keeping customer satisfaction as high as possible is important for the sustainability of a business.
Improving customer satisfaction requires understanding customer expectations better. This is possible with continuous listening and monitoring. By doing so, businesses not only figure out what customers expect but also detect their pain points, which show up as complaints.
Call centers are the primary customer service points that handle customer complaints. Complaint management is a tough task for call center teams. Providing on-time feedback and reducing the number of complaints is important.
Speech Analytics offers an effective solution for complaint management. The technology analyzes 100% customer interactions and provides call center managers with insights into customer satisfaction, agent performance, and service quality.
The steps below can help call centers to reduce customer complaints and increase customer satisfaction with Speech Analytics:
- Detect the problem
With manual evaluation methods, only a small ratio of recorded calls can be evaluated. With such a limited evaluation, it is almost impossible to detect complaints. On the other hand, Speech Analytics analyzes 100% of the calls and allows supervisors to pinpoint the calls that include complaints.
- Find the root causes
Detection of the behaviors that cause customer complaints is the primary step. Speech Analytics allows call centers to take one step further by showing the real reasons for these complaints with root-cause analysis. This analysis lets managers compare dates, agents, agent groups, queries, and voice channels to identify and respond to common problems.
- Take action
Features like statistical comparison and automatic evaluation allow supervisors to generate in-depth reports about agent performance. They can transform evaluation results into agent feedback and training material to improve agent performance. So, they can guide agents through enhanced customer service.
Here is how one of our dear customer Credit Europe Bank Russia reduced customer complaints at its contact center by 35%
As one of the leading financial services providers in Russia, Credit Europe Bank is featured in Forbes TOP 10 Banks in Russia List. The bank was searching for solutions to increase the efficiency of its customer service operations.
CEB Russia was targeting to increase efficiency for its call center, collections, customer care, telemarketing activities. The bank needed to monitor and evaluate inbound/outbound customer calls to gain insights on how to increase call quality, agent performance, collections performance, sales revenue and to reduce customers` complaints executing preemptive actions. This required an automated quality management approach due to the vast amount of calls that cannot be fully evaluated with manual monitoring methods.
The Results After Speech Analytics
- 35% decrease in customer complaints
- 25% increase in customer satisfaction
- 2X Increase in sales at mobile banking channel
Publish Date: August 12, 2020
The world’s leading research and advisory company, Gartner includes Sestek in its Market Guide for Speech-to-Text Solutions, published in April 2020.
Sestek was listed under Broader NLP Suites and Services of the Platform and Services category. This category covers the vendors who have the most well-developed value-added services and differentiate with speech features, the ability to deliver edge-based models, domain customization, and system integration support. This confirms Sestek’s leading position in the crowded conversational AI market.
Recognition accuracy is among the distinguishing features of Sestek’s Speech-to-Text technology. Offering high accuracy rates in more than 15 languages, including English, Spanish, French, Russian, Turkish, and Arabic, Sestek provides frictionless experiences both for end-users and for business units.
Sestek’s CEO, Professor Levent Arslan, says, “Speech-to-Text is the core technology that empowers our conversational solutions like Conversational IVR, Chatbot, and Speech Analytics. Our vast vertical market experience in financial services, retail, telecom, and healthcare helps us deliver tailor-made projects in a fast and highly accurate manner. We are proud to be recognized as a leading technology provider by Gartner.”
To see the summary of the report, please click here.
Publish Date: May 13, 2020
Download our free e-book to find out.
Can technology find a cure for COVID-19? It is still too early to give an answer to this question, although researchers are working on it.
If we change the question as “Can technology help us to handle these difficult times?” We don’t need time to find an answer. The answer is obviously, “Yes.” Because like a godsend, technology is helping us transform the way we work and the way we live.
Being prepared with a digital workforce and digital technologies paid off. Thanks to advanced digital technologies, millions of people easily adapted to the changes due to social isolation concerns. Schools managed to switch to online classes so that education was not interrupted. Millions of employees continue to do their job without leaving their homes. And many companies continued to serve their customers without needing physical contact points.
And AI was on the stage as always. Conversational AI technologies enabled us to reach brands easily whenever we need them. We continued to get the high-quality service as we used to do simply by interacting with a chatbot, a virtual assistant, or a speech-enabled IVR system.
A crisis means a shaky ground for your brand’s image. If you can’t provide your customers with what they need on time, this might damage your brand. On the other hand, offering your customers consistent self-service across any channel, they prefer would help you to turn the crisis into an opportunity for your business. And to achieve this, you can get help from Conversational AI.
As Sestek Marketing Team, we prepared a playbook to guide you through your Conversational AI journey. By downloading our free e-book, you will learn about the definition of Conversational AI, along with technologies supporting it. You will also see an industry snapshot that defines today and tomorrow of the technology. You will dive into the benefits of conversational AI, with a list of products that include this technology. And the final section of our e-book which was designed as a playbook aims to guide you through the implementation of the technology in your own business.
Publish Date: May 4, 2020
With any tool, technique, method or system they developed, humans lead to reorganization of the natural, spatial, temporal conditions which created and defined them. Let’s take AI, for example.
There is no field where AI does not interfere, interact, lead to change, or improve. Of course, one of the most important issues of our life is health. And the use of artificial intelligence in health has already started to transform this field.
Physicians have been performing analysis, diagnosis and treatment for hundreds of years. They accumulate and convey what they know and experience verbally and in writing. This is how medical science / art / profession has evolved and continues to evolve. Of course, medicine is not an isolated field, developments in the fields of biology, anatomy, physiology, etc. have led to the development of medicine. Moreover, the development of engineering disciplines, the development of many fields from genetics to imaging, from biomedical devices to hygienic issues have greatly contributed to the development of medicine and human health.
Especially, the amount of data growing day by day and the increase of analytical applications will contribute to the development of analysis, diagnosis and treatment methods. As the work done previously by human mind is done through the algorithm, error rates will decrease, sensitivity will increase and as a result, more lives can be saved, life expectancy will be longer, health quality will increase, health spending will decrease.
Especially if AI comes into play, human errors will decrease, sensitive diagnoses beyond the human mind will be made, the best treatments can be developed based on the data collected worldwide, even preventive measures can be taken based on the predictions, and recommendations and actions can be produced to eliminate diseases.
Medical Solutions Powered by AI
We can already talk about many medical solutions powered by AI. The first examples that come to mind are applications related to personal health assistance. One of them is ADA. Ada’s core system connects medical knowledge with intelligent technology to help all people actively manage their health and medical professionals to deliver effective care.
Another one is Apple’s iOS Health. This health app consolidates data from your iPhone, Apple Watch, and third-party apps you already use, so you can view all your progress in one convenient place. You can see your long-term trends, or dive into the daily details for a wide range of health metrics.
The use of artificial intelligence in medicine is no longer a myth. Now, the greatest assistant of doctors in every field are algorithms, machine learning systems and robots equipped with many abilities…
AI revolutionizes health as it does in every area of our lives. Health services worldwide are also significantly affected by this change. Machine learning and AI affect physicians, hospitals, and all other health-related areas.
According to Eric J. Topol’s article published in the journal Nature Medicine, everyone in the healthcare industry, from specialist doctors to first aiders, will use artificial intelligence technology in the near future.
According to GE’s projection, the artificial intelligence market for the health sector will exceed $ 6.5 billion by 2021. Considering that 39 percent of decision makers in the health sector plan to invest in machine learning and predictive analysis systems, this figure will increase further in the coming years.
How will AI Contribute to Our Health?
So, how will AI, ML and algorithms create changes in hospitals and contribute to our health?
We can say that the most benefited area is and will be the diagnosis of diseases. Accurate detection of diseases requires years of medical education. Diagnosing even after this training, is challenging and time- consuming. In many areas of medicine, the fact that the demand for specialists has exceeded the supply puts physicians in stress, and the diagnosis of diseases is further delayed.
Machine learning - especially deep learning - algorithms have made great progress in the automatic diagnosis of diseases recently, making the diagnostic process cheaper, easier, and more accessible.
Machine learning is useful in the following similar areas, where the diagnostic information examined by physicians is digitized:
– Lung cancer and stroke diagnosis by analyzing computed tomography scans
– Determination of the risk of sudden heart attack by analyzing electrocardiograms
– Classification of lesions by analyzing skin images
– Determination of diabetic retinopathy indicators by analyzing eye images
Thanks to the abundant data available in these areas, algorithms can be as successful as specialist physicians on the diagnosis. The only difference is that algorithms can diagnose in a very short time and can do this cost-effectively from anywhere in the world.
AI is especially popular in the field of Radiology. More than two billion chest X-rays are taken each year in the world. According to the researches, AI algorithms are more successful than people in evaluating these X-rays and diagnosing diseases. In addition to X-ray films, these algorithms are used in all kinds of medical imaging systems such as CT, MR, echocardiogram, and mammography, and results are obtained at speeds up to 150 times compared to humans.
According to studies, physicians spend much more time on data entry and desk work than they do actually talking to and engaging with patients. When processes like data entry and analysis of test results are automatedAI systems will alert and inform doctors about potential problems, enabling them to be more interested in patients and interpret signals more healthily. Considering that the world population is getting older and the need for a doctor is increasing, every second gained can lead to the survival and prolongation of many people.The question of whether AI or physicians are also on the popular side of the issue. In emerging countries such as China where there is an acute shortage of trained doctors, “Doctor vs. machine” competitions are very popular. This is illustrated by the Chinese TV broadcast of the brain tumour diagnosis and progression prediction competition between a team of 25 expert doctors against the Biomind artificial intelligence (AI) system. The 2:0 win of the AI over the humans in analyzing brain images gained high visibility in China.
AI-supported Surgery & Drug Development
Another area where artificial intelligence is used in medicine is surgery. AI systems can guide surgeons during the operation by analyzing patient data before surgery. Systems can also combine data on past surgeries and develop new and more effective surgical techniques. Researches show that complications are reduced by five times, and hospital stay is reduced by 21 percent in AI-supported operations.
Another field that uses artificial intelligence is drug development. Developing drugs is a very expensive process. The majority of analytical processes during drug development can be carried out much more effectively by machine learning. This will shorten years of work and reduce millions of dollars of investment.
AI is successfully used in all four basic stages of drug development:
– Determining the targets to be intervened
– Identifying potential drug candidates
– Acceleration of clinical trials
– Finding biomarkers for the diagnosis of the disease
AI-supported Personalized Treatment
The last area powered by AI on which I want to talk about is Personalized Treatment. Different patients react differently to medications and treatments. Therefore, personalized treatment is critical to prolonging patients’ lifespan. However, it is not easy to identify the factors used to determine which treatment method to choose.
In the article of Doctor Bertalan Meskó, who describes artificial intelligence as “the stethoscope of the 21st century,” it is stated that AI will make the “uniform” treatment history and suggest personalized treatments, therapies, and medications.
Machine learning can automate this complex statistical study and identify indicators that will be used to determine the patient’s response to a particular treatment. The system learns this by cross-evaluating similar patients by comparing the treatments and results applied to patients. The resulting predictions can make it easier for doctors to determine which treatment to apply.
For example, colorectal cancer patients in Brazil usually refuse the surgical removal of the colon because of cultural reasons. That’s why oncologists turn to methods such as radiotherapy and chemotherapy. However, only 20 percent of patients respond positively to these methods. So, how will it be determined which patient is in this 20 percent group? Here, deep learning algorithms come into play. Algorithms scan the data of patients and determine the appropriate treatment method in a short time and accurately.
AI and the Coronavirus
It is obvious that AI makes remarkable contributions to healthcare. And a question comes to mind since it is high on the agenda: What about coronavirus? Although the spread of the virus is a very recent development, AI-powered applications for virus diagnosis have already appeared. AI company Infervision launched a coronavirus AI solution that helps front-line healthcare workers detect and monitor the disease efficiently. Imaging departments in healthcare facilities are being taxed with the increased workload created by the virus. This solution improves CT diagnosis speed, they claim. Chinese e-commerce giant Alibaba also built an AI-powered diagnosis system. They claim it is 96% accurate at diagnosing the virus in seconds. Let’s hope that AI contributes to the development of an ultimate solution to stop the spread of the disease.
The global willingness to use artificial intelligence and robots is increasing.We can say that the main factor in this increase is the desire for faster, intuitive and low-cost health services. Trust in technology is critical for increased use and acceptance; however, ‘human relations’ remains a key component of the health care experience. So, it looks like we will be able to get the most effective results when we combine the power of AI with humans.
Publish Date: March 28, 2020
ProPublica’s survey had revealed that the risk assessment algorithm named COMPAS and AI behind the system tends to identify blacks as more risky than whites.
The famous trolley dilemma on ethical philosophy asks: “would you kill one person to save five?”. In this question, you are asked to imagine you are standing beside some tram tracks. In the distance, you spot a runaway trolley hurtling down the tracks towards five workers who cannot hear it coming. Even if they do spot it, they won’t be able to move out of the way in time.
As this disaster looms, you glance down and see a lever connected to the tracks. You realise that if you pull the lever, the tram will be diverted down a second set of tracks away from the five unsuspecting workers. However, down this side track is one lone worker, just as oblivious as his colleagues.
So, would you pull the lever, leading to one death but saving five?
This is the crux of the classic thought experiment known as the trolley dilemma, developed by philosopher Philippa Foot in 1967 and adapted by Judith Jarvis Thomson in 1985.
The trolley dilemma allows us to think through the consequences of an action and consider whether its moral value is determined solely by its outcome.
The trolley dilemma has since proven itself to be a remarkably flexible tool for probing our moral intuitions, and has been adapted to apply to various other scenarios, such as war, torture, drones, abortion and euthanasia.
Of course, there is not a single correct and moral answer to this question about how people think when deciding on an action. However, it is estimated that many people answered this question as “yes, I pull the lever, I can sacrifice one worker to save the lives of five workers”. Also, this answer can be found moral by many people.
Today, apart from philosophy, this dilemma is brought to the agenda by adapting to artificial intelligence. Although there are no AI implementations that can think like a human and make moral judgments, it is often expressed by scientists that we’re approaching this. Of course, how these dilemmas can be solved by AI is of utmost importance. Especially considering that driverless cars will come to traffic in the next ten years, Though not expected of it, AI is thought to have to make some decisions and achieve moral results. On the other hand, it is often mentioned that the possibility of artificial intelligence applications and robots equipped with AI can pose a greater danger than leaving people unemployed. The danger is racist and sexist bias and prejudices in decisions made by AI. Research on the results of AI algorithms used in a number of experiments and decision making processes gives an idea about the magnitude of this danger.
Recently, A research conducted by MIT is particularly remarkable. In this research, the application of artificial intelligence, which is expected to recognize and distinguish the thousand photos uploaded to it, differentiates whites in a perfect way, But, When it comes to blacks it starts to make a big mistake. When the person in the photo is a white man, the software is right 99 percent of the time.
But the darker the skin, the more errors arise — up to nearly 35 percent for images of darker skinned women, according to a new study that breaks fresh ground by measuring how the technology works on people of different races and gender.
Research shows that speech examples used to train the machine learning application is likely to lead to bias. Such problems with the technology have been evident in popular tools such as Google Translate. Recently, while translating Turkish to English Google Translate matched a number of jobs and situations with men and some with women (for instance the sentence “o bir aşçı” translated as “she is a cook”, the sentence “o bir mühendis” is translated as “he is an engineer”) and of course the sexist bias of these translational content has been the subject of debate.
As most of you may remember, a recent example of the biased AI is An AI application developed by Microsoft. In 2016, Microsoft launched the chat application called Tay, which learned human behavior using artificial intelligence algorithms and interacted with other users on Twitter with what she learned. Tay was designed to learn to communicate with people and tweet with data provided by other users on Twitter. In sixteen hours, The tweets she created with the data she collected from Twitter users became sexist and pro-Hitler. On March 25, 2016, Microsoft had to close Thai by apologizing to all users for these unwanted aggressive tweets.
In the text of the apology, Microsoft stated that “artificial intelligence has learned with both positive and negative interactions with people” and therefore “the problem is as social as it is technical”. In fact, this seems to be the highlight of an entire discussion. It can also be clearly seen that although Thai was taught very well to imitate human behavior, she was not taught to behave correctly or morally.
As all these examples clearly show, The racist, sexist, or in some cases status bias produced by artificial intelligence arise from the data sets used to train AI. The datasets used by artificial intelligence algorithms are of course collected from the internet, which is the biggest resource. For example, Microsoft’s Thai who tries to tweet and interact with people in this way, or Google Translate are trying to learn the words, how and with which other words they are used together, so they try both to capture the meaning and to produce answers using natural language against what they understood. Artificial intelligence establishes some relationality through its algorithm while it’s learning which words, how and with which other words these words are used statistically in the datasets provided from the internet. These can sometimes be relationalities whose cause is not understood by human. But in any case, these are not artificial intelligence produced by itself, but the relationalities that exist in the data set that it uses. Therefore, it can match feminen pronouns with cooking, cleaning or secretarial jobs and masculine pronouns with engineering. In other words, the issue appears not as the prejudices of artificial intelligence, but as data sets used in the learning processes of algorithms. That is; racist and sexist content of the internet where this data is collected makes AI produce biases.
As said in the Microsoft statement, social causes rather than technical reasons lie at the root of the problem. While AI learns with the data produced by real people, it can learn to behave like a human, it can analyze the data much faster than the human mind, but at last it cannot learn whether this behavior is right or wrong. But on the other hand, do people always act “good” and “right” in the real world? Maybe, As those who claim that artificial intelligence is not biased, AI produces the most realistic results, but expectation is to see the most suitable results for an ideal world. Considering that there are inequalities and prejudices in the world we live in and the historically produced data is biased, there is no surprise that AI applications also make biased decisions and have real world bias in their decisions. On the other hand, while answering the question “Would you kill one person to save five people?”, It is not unlikely that AI would take into account the race and sex or status of these people, that is, making the dilemma deeper.
Humans shouldn’t be a single source in AI Training
Maybe it is not a very good idea for artificial intelligence to learn merely from people. It is certain that alternative learning ways for artificial intelligence, data sets that are meticulously prepared, cleaned from prejudices and bias as much as possible or algorithms showing how the AI came to which result and how will allow us to progress on these problems. When these are possible, there may be some things that people can learn from AI. Then, It may also be possible for us to negotiate the trolley dilemma and its variations with AI.
Publish Date: February 7, 2020
Page: 1 | 2 | 3