CROSS-DISCIPLINARITY IN COGNITIVE SCIENCE: A DOCUMENT SIMILARITY ANALYSIS

Date

2024-1-22

Author

Alaşehir, Oğuzhan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

118
views

43
downloads

Cite This

Systematic quantification of cross-disciplinarity necessitates bibliometric and spatial analysis, socio-institutional aspects, or text-based techniques. Especially, with the advancement in bibliometric methods, a variety of measures have been developed. Yet, while these measures capture a snapshot of the concept of cross-disciplinarity, they overlook the content itself. With the rise of Data Science and Natural Language Processing (NLP) techniques, analyzing and evaluating vast volumes of documents has become technically possible. Our study introduces a methodology using text-based techniques, offering valuable insights into the relationship between publications and their specific research fields, showing potential as a robust measure of cross-disciplinarity. This approach utilizes Doc2Vec for vectorization and cosine similarity for measuring the similarity among the articles. We designed and developed models utilizing the Doc2Vec method for analyzing cognitive science and related fields. Cognitive science was chosen as a case study due to its inherent cross-disciplinarity. Cognitive science was established as a cross-disciplinary domain of research in the 1970s. Since then, the domain has flourished, despite disputes concerning its cross-disciplinarity. Our findings reveal that this methodology is applicable to quantify cross-disciplinarity. Furthermore, we observed that cognitive science collaborates closely with most constituent disciplines. For instance, we found a balanced engagement between several constituent fields—including psychology, philosophy, linguistics, and computer science—that contribute significantly to cognitive science. In our analysis, we find that the scholarly domain of cognitive science has been exhibiting overt cross-disciplinary collaboration for the past several decades.

Subject Keywords

Cross-disciplinarity, Interdisciplinarity, Text similarity analysis, Natural language processing, Doc2Vec modeling

URI

https://hdl.handle.net/11511/108305

Collections

Graduate School of Informatics, Thesis