Unlocking insights: integrated text mining and interpretive structural modeling for enhanced user review analysis

Effective keywords are extracted from the massive milk product user review data to construct thematic terms and explore the elemental influence relationships to assist manufacturers, and e-commerce platforms in understanding user behaviour and preferences and further optimise product design and marketing strategies. By fusing two different text mining methods, term frequency-inverse document frequency (TF-IDF) and Word2vec, we explore the semantic relationships, then visualise the relevance of user reviews by drawing knowledge graphs with Neo4j, and finally, be able to explore the relationship between the themes of the mined reviews, interpretative structural model (ISM) was used for a comprehensive evaluation, and the effectiveness of the method was verified on the Suning.com website dataset. The fusion of text mining and systematic analysis helps users to locate products quickly and precisely from the huge review information. The six elements of user reviews were categorized as freshness of taste, discounted prices, logistics, customer repurchase, product packaging, nutritional composition, and their element levels were divided into three layers. the first layer was discounted prices, customer repurchase, and logistics; the second layer was product packaging and nutritional composition; and the third layer was taste freshness.