Short text clustering method for book titles in book market

Abstract

The invention provides a short text clustering method for book titles in a book market. The method comprises the steps that word vectorization is carried out on text data and set clustering key words, the distances from text data word vectors to a clustering key word vector are calculated, the clustering type of the text data is determined according to the distance from each text data word vector to the clustering key word vector, and the text data is divided into the corresponding cluster set according to the clustering type of the text data; the term frequency-inverse document frequency TF-IDF values of feature words in all the text data word vectors in each clustering set are calculated, and the feature words with the TF-IDF values meeting set conditions are determined as updated clustering key words of the clustering set; according to the updated clustering key words, the clustering type of the text data is determined. By the adoption of the short text clustering method, the clustering type of the text data can be more accurately determined, and the clustering result can better meet practical requirements of users.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a short text clustering method for titles in the book market. Background technique [0002] With the in-depth transformation of traditional industries on the Internet, Internet e-commerce websites have developed rapidly, and online shopping has become a trend. E-commerce websites have a huge amount of products. Because the product information on the Internet is intricate, the classification is cumbersome, and the update rate is fast, manual labeling of products often consumes a lot of manpower. Therefore, automatic product classification has become a basic requirement of e-commerce. For the classification of commodities, data mining methods are often used at home and abroad to operate. [0003] In the prior art, a K-means clustering algorithm is usually used to classify commodities. The K-means algorithm is an unsupervised clustering algorithm. It is based on a certain dista...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a short text clustering method for titles in the book market, which is used to solve the problem that the clustering methods in the prior art cannot accurately cluster titles in the book market

Images