Micro-blog hot topic discovery based on string frequency

Abstract

This paper puts forward a method to find the hot topic in the micro-blog text data set: select the most frequent strings and then find the hot topic from these frequent strings, the length of the string is set to 1, 2 to 8 words respectively, the most frequent top 50 strings were selected from each group, altogether 400 strings, by browsing these 400 strings, the current hot topics among the micro-blog data set can be found. Experiment demonstrates that these 400 frequent strings can be got in short time, and the hot topics of micro-blog data sets can be found from these 400 frequent strings.