Short text clustering method based on deep learning

Abstract

The invention discloses a short text clustering method based on deep learning. The method includes the following steps that S101, the semantic similarity between short texts is obtained through calculation through a convolutional neural network; S102, the semantic similarity is applied to a clustering algorithm, and the short texts are clustered. The existing short text clustering accuracy is improved, then mass short text data can be more rapidly and accurately subjected to clustering analysis, the method can be widely applied to the fields such as short text clustering tasks, sentiment analysis and recommendation systems, short text similarity parts and short text clustering parts are calculated through the convolutional neural network without preprocessing input short text data, and the length of the input short texts can be increased.

Description

technical field [0001] The invention relates to the technical fields of deep learning and text mining, in particular to a short text clustering method based on deep learning. Background technique [0002] Text clustering is a major topic of cluster analysis algorithms in the fields of data mining and natural language processing. With the rapid popularization of the Internet and the rapid development of information technology, the total amount of data has become larger and larger, and the relationship between data has become more and more complex; at the same time, due to the development of social media, text data has grown rapidly, and usually In the form of short texts: such as Weibo, product reviews, and geographic location information, how to accurately and quickly extract valuable information from a large-scale short text data set has become a new challenge. [0003] The usual practice is to use text clustering and other methods to effectively organize short text inform...

AI Technical Summary

Problems solved by technology

But the disadvantage is that it does not take into account the semantic connection between short texts, making the effect of clustering unsatisfactory

Images