User Demand Research Based on Internet Information

With the rapid development of the Internet, network interactions become more frequent. How to use network big data to obtain user needs in a timely, fast, and accurate manner is the only way to improve the competitiveness of enterprises in the market. This article focuses on product-related online review texts, and proposes a new method for obtaining user needs based on Internet information, which is, using the TF-IDF function to extract product feature words, then using a method of similarity forest calculation based on path and depth to classify the review text, and using the sentiment dictionary to calculate the sentiment tendency of the review text under each type of feature set. So, a product feature evaluation model based on text mining can be constructed. Then, through the automatically generated survey questionnaire to simulate the shopping decision behavior of the user, a large amount of real user demand information is quickly obtained. This method also provides convenience for users' shopping decisions and has verified its feasibility and effectiveness by taking Jingdong Mall's top ten mobile phone brands as an example.