Method and device for obtaining subject vocabulary entry

Abstract

The invention provides a method and a device for obtaining a subject vocabulary entry. The method comprises the following steps of: S1, obtaining search keywords from a user search log; S2, taking nouns or phrases which point to a certain entity as a central word set; S3, taking words or phases which are used for describing the property of one aspect of the entity as a marker word set; and S4, taking vocabulary entries which are formed by utilizing the vocabulary entries in the centre word set and the marker word set and are in the form of 'central words+marker words' as a candidate subject vocabulary entry set; and filtering the candidate subject vocabulary entry set to obtain a subject vocabulary entry set, wherein the step of filtering the candidate subject vocabulary entry set comprises the specific steps of: judging whether the vocabulary entries appear in the search keywords or not and removing the vocabulary entries which do not appear in the search keywords. Compared with the prior art, manual editing is not needed when the subject vocabulary entry is obtained, so that the result deviation is small and the accuracy is high.

Description

【Technical field】 [0001] The present invention relates to Internet application technology, in particular to a method and device for obtaining subject entries. 【Background technique】 [0002] Network interactive question-and-answer platforms, such as Baidu Zhizhi, Sina Aiwen, Google Q&A, Soso Ask, Yahoo Knowledge Hall, etc., provide a platform for netizens to interact and communicate. Users can organize related knowledge introductions or pictures around a certain theme, such as "World of Warcraft", "Moon Quest", etc., to form a complete theme knowledge. Users can also use knowledge search to find answers to questions they encounter in daily life, study or profession. Users can also communicate with other people who share common interests, ask questions, find answers, help each other, and share knowledge. [0003] The question page of the question-and-answer platform is the main source for users to obtain problem-solving information. In order to meet the needs of extended br...

AI Technical Summary

Problems solved by technology

[0005] However, because the first method is completely organized manually, it requires very high knowledge and ability of editors, and the pressure of artificial intelligence and manual labor is very high. Few people can participate in and edit knowledge masters

The results of topic organization depend on the different understandings of each person, so the edited results are very different; in addition, due to the limited ability of people, it is difficult to update knowledge in time

[0006] The second method is online topic automatic mining based on keywords. Since the topic description form is only keywords, keywords have insufficient ability to express topics and poor readability, which restricts users' desire to browse and click.

Moreover, the granularity of knowledge based on keywords is too large, and the knowledge of corresponding words is complicated, which cannot meet user needs

Images