Method and device for screening text content

Abstract

The invention discloses a method and a device for screening text content. The method comprises the following steps: obtaining at least two pieces of candidate text content from different data sources, carrying out word segmentation on each piece of candidate text content, selecting a preset quantity of words with the highest weights to form a text characteristic vector, and representing the text characteristic vector as a first text characteristic vector; obtaining standard text content which is associated with the candidate text content, carrying out word segmentation on the standard text content, selecting a preset quantity of words with the highest weights to form a text characteristic vector, and representing the text characteristic vector as a second text characteristic vector; and calculating a distance between the first text characteristic vector and the second text characteristic vector, judging whether the distance value obtained by calculation is greater than a distance set value or not, and if the distance value obtained by calculation is greater than the distance set value, reserving the corresponding candidate text content as a recommendation source, otherwise, removing the corresponding candidate text content, and taking the removed candidate text content as the candidate source. The scheme can realize a purpose that the high-quality text content can be automatically selected.

Description

technical field [0001] The invention relates to text information processing technology, in particular to a method and device for screening text content. Background technique [0002] When querying the target text content through the network, there are often multiple candidate text content, and high-quality text content needs to be screened out. The text content is, for example, a novel text, an online text about a specified subject, etc.; the following takes online reading of a novel text as an example for specific description. [0003] Online literature first emerged on the Internet and is a popular form of text on the PC side. With the rapid development of Internet literature, more and more Internet novel websites have emerged. The same novel often appears on many websites, but the quality and degree of impurity in the text of the novel vary from website to website. In practical applications, it is often necessary to screen novel texts to find high-quality novel texts. ...

AI Technical Summary

Problems solved by technology

[0006] This method has the following disadvantages: due to the large number of third-party novel websites, and the uneven network speed, update speed, and content quality, readers need to spend a lot of time on manual search and screening to select high-quality online novels, and the reading efficiency is low.

[0009] This method has the following defects: this scheme reduces the search time for readers, but it still needs to be manually screened after the search, which still takes a lot of time; and, for many chapters of a novel, some chapters may still appear in the Inconsistent quality on different websites may require frequent switching between websites when reading a book, disrupting the continuity of the reading experience

[0010] To sum up, in the case of multiple candidate text content, human participation is required for verification to filter out high-quality text content, which is inconvenient to operate and low in efficiency

Images