Adoption of Farm Technologies across the Terai Region of West Bengal: A Comparative Topic Modelling Study of Empirical Literature Review
DOI:
https://doi.org/10.23910/1.2026.7025Keywords:
Technology adoption, literature review, topic modellingAbstract
This study was executed from June, 2025 to January, 2026 in the Terai region of West Bengal mainly in Jalpaiguri and Cooch Behar district. The objectives were to identify technologies introduced in KVK villages over the past years, compare adoption gaps and performance between intervened and non-intervened farmers, analyse socio-technical and economic determinants, and design a feasible extension roadmap. To achieve these objectives, a systematic qualitative comparative analysis using literature review and advanced topic modelling was undertaken. A curated corpus of forty-five high-quality researches was constructed through rigorous inclusion criteria and quality assessment, followed by preprocessing and implementation of five computational approaches: Non-negative Matrix Factorization (NMF), Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), and optimized variants. Multi-metric evaluation integrated reconstruction error, perplexity, topic diversity, interpretability, and computational efficiency. Results revealed consistent thematic clusters across models, including the central role of KVK as institutional hubs, gendered pathways of adoption through SHGs and training, and regional specificity, with Jalpaiguri and Cooch Behar emerging as technology adoption hotspots. Technology-specific themes such as solar irrigation under PM-KUSUM, mechanization via power tillers, direct-seeded rice, and integrated pest management appeared across findings. Divergence patterns highlighted methodological trade-offs like NMF with count data produced granular differentiation, while probabilistic LDA yielded more coherent narrative themes. The findings underscored both robust thematic structures and methodology-dependent artifacts, positioning topic modelling as a valuable but interpretive tool. This study provided a data-driven foundation for inclusive agricultural strategies tailored to smallholder realities in South Asia.
Downloads
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Debraj Saha, Soumyadeep Thakur, Kalyan Kanti Das

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Authors retain copyright. Articles published are made available as open access articles, distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited. 
This journal permits and encourages authors to share their submitted versions (preprints), accepted versions (postprints) and/or published versions (publisher versions) freely under the CC BY-NC-SA 4.0 license while providing bibliographic details that credit, if applicable.

