Text relevance calculating method and device

Abstract

Embodiments of the invention provide a text relevance calculating method and a device thereof. The method comprises the following steps: receiving a first character string and a second character string; calculating a text relevance characteristic value of the first character string and the second character string, and calculating a semantic relevance characteristic value of the first character string and the second character string; fitting the text relevance characteristic value and the semantic relevance characteristic value into a relevance characteristic value of the first character string and the second character string based on the logistic regression model. The text relevance calculating method and the device thereof increase the precision of relevance judgment, save storage space and reduce the cost.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of Internet applications, and more specifically, to a text correlation calculation method and device. Background technique [0002] With the rapid development of computer technology and network technology, the Internet (Internet) plays an increasingly important role in people's daily life, study and work. Various applications emerge in endlessly on the Internet. [0003] Search advertising is a very important business in the Internet advertising ecosystem. It is attached to search engines and is essentially keyword-based sales matching. In the commercial promotion database, in addition to providing the title and description of the ad for display, the advertiser must also add some relevant keywords (that is, purchase words) to the ad, and specify the matching type, bid and orientation Match target traffic (that is, users who match the search intent). In a classic matching process,...

AI Technical Summary

Problems solved by technology

[0006] However, the traditional calculation method based on the word vector space model in the document faces the problem of sparse features on short texts

At the same time, since the word segmentation results of short texts depend on the language model, the consistency of different word segmentations cannot be guaranteed, and the sparseness of vectors will also be exacerbated to a certain extent.

Therefore, the traditional calculation method based on the word vector space model in the document has the disadvantage of low accuracy of correlation determination

[0007] Moreover, in the traditional calculation method based on the word vector space model in the document, a large amount of storage space is required to store the word vector, so the storage space is wasted and the cost is increased.

Images