The aim of the Text Mining of Political and Legal Texts (POLTEXT) Icubator Project is to apply cutting-edge Big Data methodologies to the study of political and legal documents. Traditional approaches to the analysis of qualitative data sources (such as text, pictures and videos) mostly relied on the hand-coding and manual labelling of data. While human insight is still crucial for understanding the meaning and content of qualitative data, the traditional approach also entails some risks when it comes to the validity and reliability of research results. Two independent human coders may assign a different label to the same text segment due to the inherently subjective nature of human decision making process. Furthermore, the large scale of the data sources (e.g. the complete text of all legislation in a country, which may span several hundred years) may render human data processing unfeasible altogether.
For these and similar reasons the approach of quantitative text analysis and text mining has emerged as the new industry methodological standard in Big Data projects in the social sciences dealing with textual sources. The aim of this project is to create an international network of like-minded scholars who apply text mining techniques and procedures to the distinctive data sources of their research projects. By fostering interdisciplinary co-operations (not just between project participants, but also with scholars in Data Science and Computational Linguistics), joint projects between multiple generations of scholars based in various European countries the incubator offers a pathway to much-needed knowledge transfer and to social science research projects of the highest quality.