Hot Spot Discovery and Evolution Situation Analysis Method in Network Forum

Abstract

The invention provides a method for hotspot discovery and evolution situation analysis in a network forum, using a crawler program to extract the HTML text of all posts in a forum; extracting pageviews and replies from each post text to form a binary group, assuming the first The binary group of the i post is (xi, yi); use the formula to calculate the score of the i-th post; obtain a popular post, and count the number of replies that occur every day from the topic posting to the capture time according to the reply record to get the binary group . The fitting effect and evaluation index are listed below: SSE=1.548e+07. SSE is the sum of squares of the error term, which reflects the discrete state of each observation value of each sample, and is also called the sum of squares within the group or the sum of squares of the residual error. R-square = 0.8339. R-square is the fitting coefficient, the larger the value, the better the fitting degree. RMSE = 525.7. RMSE is the root mean square error, which can be used as a numerical indicator to measure the measurement accuracy. After analyzing these indicators, it can be seen that the fitting effect is relatively ideal. Find the largest extreme point as x0=14; take x1=15>x0, y'(x0)<0, so the popularity of this post is declining.

Description

technical field [0001] The invention relates to a method for hotspot discovery and evolution situation analysis in network forums, and belongs to the technical field of network evolution analysis methods. Background technique [0002] Existing solutions to the problem of finding hot topics in online forums are mainly to simply use the number of views and replies as the evaluation criteria, that is, if the views and replies are large, the topic of the post is more popular. Such a simple judgment often ignores the inherent mathematical meaning of these values; using manual tracking methods in situation analysis, managers of online forums often monitor the content of posts to obtain their evolution laws, which is time-consuming and expensive. strenuous. Contents of the invention [0003] The purpose of the present invention is to be able to find popular and long-duration posts in the current network forum, and the obtained results are as consistent as possible with the actua...

AI Technical Summary

Problems solved by technology

Such a simple judgment often ignores the inherent mathematical meaning of these values; using manual tracking methods in situation analysis, managers of online forums often monitor the content of posts to obtain their evolution laws, which is time-consuming and expensive. strenuous

Images