Knowledge Technologies for the Social Sciences (KTS)
The department KTS improves the findability and reusability of our research data. This is done with the help of an integrated search, automated processing methods for digital behavioral data as well as their integration into innovative research data infrastructures. To this end, KTS researches AI-based computer science methods for the interpretation, integration and use of heterogeneous data.
GESIS-Search
Find information about social science research data, publications on research data as well as open access publications.
GESIS Knowledge Graph
The GESIS Knowledge Graph (GESIS KG) represents metadata of scientific resources available in the GESIS Search and its semantic relationships in an integrated and consistent form and makes them accessible for reuse.
Methods Hub
The Methods Hub is a collaborative online platform dedicated to advancing social science research by sharing computational methods and related tutorials for using, managing, and enriching digital behavioral data (DBD).
GESIS Web Data
The service GESIS Web Data acts as an umbrella for different activities around collecting digital behavioral data from the Web, especially from online platforms, including social media.
Our Teams
Information & Data Retrieval
The Information & Data Retrieval team focusses on service and research activities concerned with the efficient search for research data and related information resources in the social sciences. The goal of the team is to provide suitable state-of-the-art search technologies for GESIS and to offer innovative search solutions for the research community. The team leads the technical development of the GESIS-Search and conducts information retrieval research as a foundation to further improve the existing search infrastructure. To do so, the team conducts research in various IR areas, e.g. interactive information retrieval, semantic search, conversational search, recommendation and adjacent fields, such as Natural Language Processing and Scientometrics. The team integrates research-based innovations into the GESIS Search to enable a comprehensive search across all information resources, a good user and search experience and links between the information resources based on the GESIS Knowledge Graph. The team collaborates closely with HII, IE&L and FAIR on various aspects of information and data retrieval and scholarly document processing.
Information Extraction & Linking
The team Information Extraction & Linking develops methods for the integration and linking of digital behavioral data (DBD) and research data. The team conducts applied research in the fields of natural language processing (NLP) and Information Retrieval (IR). Examples include methods for the identification of research methods and datasets mentions in scientific publications, online discourse analysis, and micro-post retrieval for social science concepts from social media archives (e.g., X/Twitter and Telegram). Furthermore, the team advances the linking and homogenization of established vocabularies and annotation frameworks that support a holistic view of research data. Our research contributes to the development of our two core services, i.e., the Web Data for the Social Sciences and the GESIS Knowledge Graph (KG).
FAIR Data
The FAIR Data team improves the visibility and usability of research data along the FAIR principles. A particular focus is on improving the findability and interoperability of GESIS data. To this end, cooperation with national and international research data infrastructure initiatives (in particular, NFDI and EOSC) is a central task, ensuring transfer to the community and connectivity to standards and innovations from the community (e.g. FAIR Assessment). The team is involved in four NFDI consortia (KonsortSWD, NFDI4DataSience, BERD@NFDI and Base4NFDI). The team conducts research at the interface of information extraction, knowledge graphs, data integration and information retrieval to continuously contribute to the improvement of research data and services.
Big Data Analytics
The team Big Data Analytics works on research and services related to large scale computing. The goal is to make the (scalable) analytics FAIR in terms of responsible data science alongside being (Findable, Accessible, Interoperable and Reproducible). Our activities range from the design of scalable infrastructure for large scale data processing to the development of analytics methods for the social sciences. The aim is to make these methods publicly available, reproducible and explainable wherever possible. The team works at the intersection of the knowledge graphs, distributed analytics and big data.
Team Big Data AnalyticsHuman Information Interaction
The Human Information Interaction team works at the intersection of Human Computer Interaction and Machine Learning. By applying methods of user-centered design, the team actively involves users in the development process and thus contributes to the development of digital services with convincing usability and user experience. One example is the GESIS Search, our integrated search system for different types of information. Another goal is to use machine learning to generate insights about cognitive processes and personality traits from digital behavioural data and to incorporate them in user models. In research, the team is looking at online discourse behavior and how this can be improved, what eye tracking and physiological data (e.g. EEG) reveal about cognitive processes in reading and comprehending text (e.g. Web content or survey items), and how users with more vague information needs can be better supported in their information search process.
Team Human Information Interaction