Cognitive distortions are negatively biased thinking patterns and erroneous self-statements resulting from and leading to logical errors in one’s own internal reasoning. Cognitive distortions have an adverse effect on mental health and can lead to mental health disorders in extreme cases. This paper belongs to a bigger project which aims to provide an application for detecting and classifying cognitive distortions in texts. As no public data sets were available for the task, the first contribution of the proposed work lies in providing an open-source labeled dataset of 14 cognitive distortions consisting of 34370 entries collected via crowd-sourcing, user questionnaires, and re-purposing emotions dataset from social media. The dataset is collected in cooperation with a licensed psychologist. We implemented a baseline model using Naïve Bayes and Count Vectorizer and different CNN, LSTM, and DNN classifiers to classify cognitive distortions based on the dataset. We investigated the usage of different word embeddings with the best-performing models. The best-performing model relied on a CNN with pre-trained Sentence-BERT embedding with an F1-score of 84 % for classifying cognitive distortions. The best-performing model was built into C-Journal, a free journaling and mood-tracking mobile application that pinpoints potential thinking distortions to the users.

Users have different expectations while interacting with chatbots. However, current chatbots do not consider the users’ preferences and needs, creating unpleasant biases towards certain user groups. User characteristics play a vital role in predicting preferences and thus eliciting better responses as advocated by Character Computing. A framework was proposed in the literature to build character-based chatbots with adjustable traits. In this paper, we customize the framework modules to build two chatbots for different domains using the same set of chatbot characters. First, we present a character-based chatbot for academic support. It provides students with information regarding the content, dates, timings, locations, and deadlines of their courses. Second, we present a character-based chatbot for COVID-19. It provides users with answers to frequently asked questions, advice and precautions, cases per country, and symptoms checking. Users can select their preferable chatbot character to interact with, and receive personalized responses accordingly. Moreover, we investigated the prediction of a user’s favorite chatbot character based on their personality traits to create adaptable chatbots in the future. Finally, we investigated the effect of a chatbot’s character on its likability and trustworthiness.

Ontologies represent the backbone of the semantic web since they represent knowledge in a machine-understandable way. The more complex the ontologies are, the greater is the need for visualization tools to ease the understanding of the large amount of knowledge represented by these ontologies. Visualizing large ontologies is a still under-investigated interesting challenge. In this paper, we present VisCOnto, a visualization tool of complex ontologies by extracting sub-ontologies that correspond to excerpts of a large ontology focusing on some concepts of interest. VisCOnto combined with WEBVOWL can create sub-ontologies to create expressive information visualizations by harnessing the semantic reasoning capabilities provided by ontologies. A proof of concept use case for applying VisCOnto to visualize an existing complex ontology of human behavior and character is presented.

Now that most daily endeavors have migrated to computers and the Internet, having systems that adapt to users is a rising trend. By detecting and adapting to the user's traits and character instead of only the current states, there is great potential for novel user experiences. Character Computing which was introduced in [3] is now extended to the field of experimental psychology and psychological theorizing. Our approach integrates psychological and computational aspects of advanced sensing and processing technologies to detect the building blocks of a user's character traits and states which will allow computer systems to adapt to the user's behavior and his/her feelings more reliably and mimic them according to situational demands. Accordingly, in this workshop the different aspects of Character Computing are to be discussed from an interdisciplinary perspective taking computational and psychological accounts into consideration in order to strengthen already investigated and potential applications as well as the theoretical foundation and challenges of Character Computing.

The emerging field of Character Computing aims at combining computer science and Psychology to achieve humanaware computing that considers all aspects of human behavior and human experience, including the human user's states, traits, and surrounding situations. We propose developing an computational and psychologically-driven architecture for Character Computing based on existing accepted psychological models of cognition, personality and emotions. This would help towards further defining Character Computing and setting strong theoretical foundations for it. The computational architecture is at the heart of this PHD thesis. Developing the architecture requires input from Psychology and Computer Science which is currently already ensured by continuous input from the supervisors Prof. Cornelia Herbert (Ulm University) and Prof. Slim Abdennadher (GUC). Beta-versions of the architecture are then tested in experimental psychological experiments and use-cases that would most benefit from considering the character of the user as a component. The situations most telling and affected by that are investigated through machine learning techniques, experimental psychological methods and computational modeling. The use-cases range from IoT applications, smart assistants, autonomous cars to health care and education.

With the expansion of multicultural communities, raising awareness to diminish cultural misconceptions became a must. We investigate different research probes using interactive Virtual Reality (VR) applications to increase such awareness. VR environments provide a high sense of immersion, which enhance the user's experience. In this work, we present three VR interactive research probes that allow users to understand cultural differences and to reduce cultural misconceptions.