The University-Industry Interaction: A Systematic Literature Review

Academic Journal of Interdisciplinary Studies. 2025. Vol. 14. No. 4. P. 42–62.

Harutyunyan H., Krakovich V.

The study provides a systematic review of university-industrial partnerships using BERTopic thematic modeling. The analysis includes 1055 articles from the Scopus database, selected according to the keywords university and business. Irrelevant topics are excluded (for example, family business, business ethics), as well as articles with missing metadata. Topics are identified by combining titles and annotations of articles, vectorization with BERT, dimensionality reduction through UMAP, and clustering using K-Means. The optimal number of clusters (10) is determined based on the Kalinski-Kharabas index. To interpret the themes, c-TF-IDF was used, highlighting key bigrams. After excluding common terms (business university, university industry), the five most significant bigrams were transmitted to ChatGPT to form cluster names. Each cluster was analyzed on the basis of the four most cited articles, which made it possible to identify the main research areas in the field of university-industrial relations. Additionally, a geographical analysis was carried out evaluating the contribution of countries based on weighted citations: if a publication had several authors from different countries, the contribution was calculated in proportion to the number of co-authors. The results provide a holistic view of the key themes, challenges, and trends of university-industry partnerships, which can be useful for researchers in the fields of academic entrepreneurship, technology transfer, and innovation policy.

Keywords: University-Industry Interaction, BERTopic thematic modeling