A5012928144- Ethics and Social Impacts of AI
- Imbalanced Data Classification Techniques
- Semantic Web and Ontologies
- E-commerce and Technology Innovations
- Sentiment Analysis and Opinion Mining
- Data Stream Mining Techniques
- Explainable Artificial Intelligence (XAI)
- Topic Modeling
- Privacy-Preserving Technologies in Data
- Data Quality and Management
- Adversarial Robustness in Machine Learning
- Machine Learning and Data Classification
- Spam and Phishing Detection
- Web Data Mining and Analysis
- Advanced Database Systems and Queries
- Risk and Safety Analysis
- Blockchain Technology Applications and Security
- Domain Adaptation and Few-Shot Learning
- Natural Language Processing Techniques
- Image Retrieval and Classification Techniques
- Psychology of Moral and Emotional Judgment
- Complex Network Analysis Techniques
- Stock Market Forecasting Methods
- IoT and Edge/Fog Computing
- Artificial Intelligence in Healthcare and Education
L3S Research Center
2017-2022
Leibniz University Hannover
2017-2022
University of Stavanger
2021
University of Patras
2015-2016
Abstract Artificial Intelligence (AI)‐based systems are widely employed nowadays to make decisions that have far‐reaching impact on individuals and society. Their might affect everyone, everywhere, anytime, entailing concerns about potential human rights issues. Therefore, it is necessary move beyond traditional AI algorithms optimized for predictive performance embed ethical legal principles in their design, training, deployment ensure social good while still benefiting from the huge of...
Abstract As decision‐making increasingly relies on machine learning (ML) and (big) data, the issue of fairness in data‐driven artificial intelligence systems is receiving increasing attention from both research industry. A large variety fairness‐aware ML solutions have been proposed which involve fairness‐related interventions algorithms, and/or model outputs. However, a vital part proposing new approaches evaluating them empirically benchmark datasets that represent realistic diverse...
The widespread use of ML-based decision making in domains with high societal impact such as recidivism, job hiring and loan credit has raised a lot concerns regarding potential discrimination. In particular, certain cases it been observed that ML algorithms can provide different decisions based on sensitive attributes gender or race therefore lead to Although, several fairness-aware approaches have proposed, their focus largely preserving the overall classification accuracy while improving...
Automated decision making based on big data and machine learning (ML) algorithms can result in discriminatory decisions against certain protected groups defined upon personal like gender, race, sexual orientation etc. Such designed to discover patterns might not only pick up any encoded societal biases the training data, but even worse, they reinforce such resulting more severe discrimination. The majority of thus far proposed fairness-aware approaches focus solely pre-, in- or...
Sentiment analysis is an important task in order to gain insights over the huge amounts of opinions that are generated social media on a daily basis. Although there lot work sentiment analysis, no many datasets available which one can use for developing new methods and evaluation. To best our knowledge, largest dataset TSentiment [8], 1.6 millions machine-annotated tweets covering period about 3 months 2009. This however too short therefore insufficient study heterogeneous, fast evolving...
Abstract Class imbalance poses a major challenge for machine learning as most supervised models might exhibit bias towards the majority class and under-perform in minority class. Cost-sensitive tackles this problem by treating classes differently, formulated typically via user-defined fixed misclassification cost matrix provided input to learner. Such parameter tuning is challenging task that requires domain knowledge moreover, wrong adjustments lead overall predictive performance...
AI-based systems are widely employed nowadays to make decisions that have far-reaching impacts on individuals and society. Their might affect everyone, everywhere anytime, entailing concerns about potential human rights issues. Therefore, it is necessary move beyond traditional AI algorithms optimized for predictive performance embed ethical legal principles in their design, training deployment ensure social good while still benefiting from the huge of technology. The goal this survey...
Breast cancer is the second leading cause of cancer-related death after lung in women. Early detection breast X-ray mammography believed to have effectively reduced mortality rate since 1989. However, a relatively high false positive and low specificity technology still exist. In this work, computer-aided automatic mammogram analysis system proposed process images automatically discriminate them as either normal or cancerous, consisting three consecutive image processing, feature selection,...
Recent studies in big data analytics and natural language processing develop automatic techniques analyzing sentiment the social media information. In addition, growing user base of high volume posts also provide valuable information to predict price fluctuation cryptocurrency. This research is directed predicting volatile movement cryptocurrency by finding correlation between them. While previous work has been developed analyze English posts, we propose a method identify Chinese from most...
Breast cancer is the second leading cause of cancer-related death after lung in women. Early detection breast X-ray mammography believed to have effectively reduced mortality rate. However, a relatively high false positive rate and low specificity technology still exist. In this work, computer-aided automatic mammogram analysis system proposed process images automatically discriminate them as either normal or cancerous, consisting three consecutive image processing, feature selection,...
An ideal safe workplace is described as a place where staffs fulfill responsibilities in well-organized order, potential hazardous events are being monitored real-time, well the number of accidents and relevant damages minimized. However, occupational-related death injury still increasing have been highly attended last decades due to lack comprehensive safety management. A smart management system therefore urgently needed, which instructed automating risk evaluations alerting departments...
Lexical approaches for sentiment analysis like SentiWordNet rely upon a fixed dictionary of words with sentiment, i.e., that does not change. With the rise Web 2.0 however, what we observe more and often is are sentimental per se, associated positive/negative feelings, example, "refugees", "Trump", "iphone". Typically, those feelings temporary as responses to external events; "iphone" latest iphone version release or "Trump" after USA withdraw from Paris climate agreement.