We constructed a set of programming language system for text comment data mining and quantifying score, and put forward a text comment weighted synthetic score decision model —— Fussy and comprehensive evaluation model based on analytic hierarchy process(AHP). It can help companies maximize profits and give consumers a better shopping experience. First, we cleaned and processed the data, which was a huge and critical step. We got a comprehensive score for each review text by word segmentation, keyword filtering, and generic word analysis. In this process, we built a TF-IDF model and the output of the model can easily determine whether a word is a keyword or not. Based on this, a product star keyword database was built, and word clouds were generated to link up the following generic word analysis and output the product text review score results. Then, we proposed a model of fussy and comprehensive evaluation based on analytic hierarchy process. Since the boundary between good and bad of three products is not clear, it is difficult to classify them into a certain category. So we first used analytic hierarchy process to calculate the weight of evaluation, evaluation star, product star and then made a comprehensive evaluation of all the factors. According to the subjection degree theory of the mold mathematics, the comprehensive evaluation method transforms qualitative evaluation into quantitative evaluation, gives three products the evaluation and judges which star level they belong to. Next, we analyzed the sensitivity of the model and tried to find the direction of future improvement and development. Finally, according to our analysis, baby pacifiers and hair dryers are rated five stars on a comprehensive review. The manager can adjust the sales ratio of each product appropriately and improve the shortcomings of each product.

Xinyue Zhang, Hong Guo. Text mining and decision-making analysis of E-commerce Review based on R language. Academic Journal of Humanities & Social Sciences (2020) Vol. 3, Issue 3: 52-63. https://doi.org/10.25236/AJHSS.2020.030307.

[1] Chao Song, Xiao-Kang Wang, Peng-fei Cheng, Jian-qiang Wang, Lin Li. SACPC: A framework based on probabilistic linguistic terms for short text sentiment analysis [J]. Knowledge-Based Systems, 2020.
[2] Bing Liu. Text sentiment analysis based on CBOW model and deep learning in big data environment [J]. Journal of Ambient Intelligence and Humanized Computing, 2020, 11(2).
[3] Shailendra Kumar Singh,Manoj Kumar Sachan. SentiVerb system: classification of social media text using sentiment analysis [J]. Multimedia Tools and Applications, 2019, 78(22).
[4] Kashfia Sailunaz,Reda Alhajj. Emotion and sentiment analysis from Twitter text [J]. Journal of Computational Science, 2019, 36.