Research on Validity of Low Frequency Phrases for Online Comment Data
Online comments play an extremely important role in the era of e-commerce,but with the rapid increase in the number of comments,it is difficult for people to quickly find the useful data they need.Therefore,how to find valuable information from a large number of comments has become the focus and difficulty of our research.To address this issue,this paper first conducts a comment usefulness voting analysis on the dataset.Secondly,sort the low-frequency phrases and output them in ascending order to analyze their importance.Finally,through comparative experiments,the average use-fulness of the top 100,top 200,and top 300 phrases is calculated separately to verify the better performance of the pro-posed method,and to clarify that ignoring low-frequency phrases will result in the loss of many important information.
online reviewphrase average validityvoting analysis
万方数据
维普
