This study evaluates the applicability of text summarization algorithms to articles in the field of agriculture. The abstract and conclusion sections of articles on the topics of "agriculture" and "organic agriculture" were analyzed using extractive text summarization algorithms: TextRank, LexRank, Luhn, and LSA. The summaries generated by each algorithm were compared using the cosine similarity measure. These similarities were then visualized on a 2-dimensional plane using a Venn diagram. The findings indicate that there are similar tendencies in the algorithms' selection of content-focused sentences for both agriculture and organic agriculture articles. Notably, it was observed that among the text summarization algorithms, LexRank and LSA produced more consistent results across both datasets. In conclusion, it has been demonstrated that summarization methods can be effectively applied in agricultural research to reduce information density.
Temizhan et al. (Fri,) studied this question.