Intelligent Identification of Consumers’ Implicit Needs and Quality Adaptation Strategies Based on Text Mining

In the digital consumption era, consumer feedback texts contain substantial implicit needs that traditional surveys fail to capture due to sample bias and expression barriers. This study, based on 62,716 e-commerce reviews, constructs a five-layer framework—data collection, text mining, implicit need identification, need-quality mapping, and quality adaptation strategy—using jieba segmentation, LDA topic modeling, SnowNLP sentiment analysis, and keyword-pattern matching. Results show that: (1) LDA extracts 8 structured topics from all reviews and 6 pain-point topics from negative reviews; (2) 36,486 implicit need records are identified across 8 dimensions; (3) The SECI model’s externalization stage is operationalized through text mining; (4) Kano classification yields three priority levels (P0, P1, P2) with differentiated strategies. This research offers a methodological reference for enterprises to identify consumer needs and optimize quality management in digital transformation.

Keywords: Text Mining; Implicit Need Identification; SECI Model; Kano Model; Knowledge Management.