ºÚÁÏ´«ËÍÃÅ

Big Data Analytics & (Gen)AI:

Nowadays, companies and organisations have access to vast and ever-growing volumes of data – for example, via social media and the internet (e.g. online social networks, wikis, rating and review communities, discussion forums), as well as in traditional databases (e.g. data warehouses, customer databases), through direct customer contact or in human resources management. In other areas too, such as professional sport, large volumes of data are increasingly being generated, for example through match and player tracking data.

The vast majority of the available data is stored in unstructured form (e.g. images, videos or text). To unlock this treasure trove of data, methods for automated analysis are required. In the research area of Big Data Analytics & (Gen)AI, we investigate specific applications and the resulting benefits of artificial intelligence (AI) methods – including generative AI – in the analysis of (un)structured data. The interpretability of these methods is also examined in the research area of Explainable Artificial Intelligence (XAI).

The focus is on the following areas:

• Analysis of unstructured data using (Gen)AI
• People Analytics and data-driven decision support in business
• Sports analytics and data-based performance analysis in professional sport
• AI-supported human-machine interaction

#BigData #AI #GenAI #PeopleAnalytics #SportsAnalytics #XAI

Projects

• Automation of incoming mail in the insurance industry
• Evaluation of player actions in football
• Use of humanoid robots in customer service
• Future skills and competencies
• GenAI in knowledge work
• Information extraction
• (Partial) automation of online customer service

¶Ù²¹³Ù±ð²Ô±ç³Ü²¹±ô¾±³Ùä³Ù:

In the wake of digitalisation, organisations today have access to vast and ever-growing volumes of data (keyword: ‘Big Data’). However, empirical evidence shows that the data analysed and utilised is often characterised by poor data quality – even in internal corporate customer databases, on average around 30% of the stored data values are incorrect. This results, for example, in additional annual costs of 15 million dollars for the average American company. However, poor data quality is not only a major problem for businesses – in politics and society too, the need for reliable information is growing in the age of ‘fake news’. Quantitative methods are therefore needed to measure, control and improve data quality.

This issue has particularly critical implications in the age of Artificial Intelligence (AI), where poor data quality leads directly to uncertainty. If these underlying uncertainties in AI forecasts are ignored – for example, by simply replacing missing values with averages – AI systems often issue overly confident and sometimes incorrect recommendations. For human decision-makers in human-AI teams, this poses enormous risks: A deceptive illusion of certainty arises, which can lead to overreliance and potentially serious misjudgements. To harness the transformative potential of AI in a responsible and trustworthy manner, it is therefore essential to quantitatively capture uncertainties arising from data quality defects, map them transparently and disclose them to users.

#DataQuality #FakeNews

Projects

• Data Quality in the Automotive Industry
• Data Quality in User-Generated Content (DQNGI)
• Data Quality in User-Generated Content (DQUGC)
• Data Quality Measurement and Measures for Wikis and Knowledge Graphs (DQMM@Wiki)
• Measuring and Improving Data Quality in Unstructured Data (DQMM)
• Making uncertainties visible: Uncertainty-aware AI

Explainable AI (XAI)

Artificial intelligence (AI) is playing an increasingly significant role in our lives. From chatbots, spam filters and shopping recommendations for individuals to fraud detection for financial service providers and marketing initiatives for businesses – AI is already behind much of this. The growing prevalence of AI systems is opening up undreamt-of possibilities. However, areas of application where critical decisions are made – such as in corporate financial control or when it comes to the creditworthiness of private individuals – are under close scrutiny. The reason is the opacity of AI. In fact, studies show that many Europeans view decisions made by AI systems with unease. And this is the case even when the algorithms demonstrably deliver better results than human experts. In critical areas of application, it is important to understand how AI results are arrived at. This is where the field of Explainable AI (XAI) comes in: the research focuses on methods that make the results and functioning of AI systems comprehensible to human users.

The focus is on the following areas:

• Methods for ensuring the traceability of AI results
• User-centred design of XAI applications

#XAI #Trust #Explainability #Responsibility #Design Science #AI Made in Germany

Projects

• Explainable AI in Controlling
• Guess the City
• Personal inflation calculator
• Skill Compass
• Review-based explanations for recommendations in e-commerce
• XAI Demonstrator
• XAI Studio
• XAI-as-a-Service (XAIaaS)
• XAI in Continuing Education (XPERT)
• XAI and Human-in-the-Loop (X-Loop)

Social Impact of Information Systems

Modern information systems can not only deliver economic value but also make a significant contribution to addressing social and societal problems and challenges. Our research into the social impact of information systems focuses on the pressing social issues of our time, such as unemployment, skills shortages, integration and the strengthening of democracy. For example, our studies have shown that digital applications enhance the efforts of young jobseekers and that online peer groups (i.e. digital self-help groups) in particular prove to be valuable in numerous social contexts, such as unemployment under difficult conditions, unemployment among older people, career guidance for young people, or the integration of refugees. The digital nature of modern information systems brings particular advantages, such as flexibility in terms of time and location, as well as the possibility of anonymity and thus a protected exchange. Artificial intelligence can act as an additional catalyst for social innovation, for example by enabling counselling to be scaled up and personalised.

Our research focuses on the following areas:

• Design and evaluation of new information systems to address societal challenges
• Social media and AI for counselling and mentoring
• Digital disinformation

#Social Impact #AI for Good #Digital in the Public Sector

Projects

• Bias and fairness in AI
• Digital Peer Group Counselling (DIGIPEG)
• Explainable AI for Learning about Digital Disinformation (EKILED)
• Feasibility study: ‘App-based peer group counselling’
• Smart Integration

Mathias Klier is active as a reviewer, associate editor and track chair of international conferences such as the International Conference on Information Systems (ICIS) and the European Conference on Information Systems (ECIS). He is the author of numerous articles in books and journals such as ACM Journal of Data and Information Quality, Decision Support Systems, Electronic Markets, Journal of Information Science, Journal of Management Information Systems and Management Information Systems Quarterly. Furthermore, he has presented the results of his work at international scientific conferences such as European Conference on Information Systems (ECIS), International Conference on Information Systems (ICIS) or International Conference on Management Information Systems (WI) .

List of publications