Method and device for providing link recommendation

Abstract

The invention discloses a method and device for providing link recommendation. The method comprises the steps of: extracting link information contained in community share according to a predetermined time period; acquiring microblog messages containing links and microblog message attribute in the link information; scoring the links corresponding to the microblog messages according to a predetermined scoring policy on the basis of the microblog message attribute acquired; acquiring the links which are ranked top N, extracting webpages corresponding to uniform resource locator URL of the links to classify the contents, and acquiring abstract data of the URL, wherein N is a predetermined natural number; and generating link recommendation and providing the link recommendation to users. With application of the invention, the availability of link recommendation can be improved, and the cost of manual operation can be lowered.

Description

technical field [0001] The invention relates to computer communication technology, in particular to a method and device for providing link recommendation. Background technique [0002] With the development of computer communication technology, especially the development of 3G network and smart mobile terminals, users' network life is becoming more and more abundant. Due to the richness of network content, network users spend more and more time obtaining relevant information from it. For example, various community websites (including but not limited to Tencent Weibo, Qzone community) have accumulated a large amount of information, including personal mood portrayal, replies, forwarding fans' information, expressing personal opinions on certain things, etc. Content, because most of the content is highly personalized information, it is difficult to focus on forming hot content, so that the time spent by users on Weibo is not directly proportional to the information obtained. Ho...

AI Technical Summary

Problems solved by technology

[0012] It can be seen from the above that the existing method of providing link recommendation, link recommendation through editor's choice, is closely related to the editor's personal hobbies and life background, which is relatively subjective, and the selected link recommendation does not necessarily correspond to hot news. It is a hot news recognized by network users, the availability of link recommendations is low, and the cost of manual operation is high; and the selection and generation of link recommendations based on the number of repetitions of web pages generally come from various news websites, and each news website’s own hotspots The selection of news is determined by editors, and the number of people who participate in the determination of hot news is relatively small. Therefore, the hot news corresponding to the link recommendation obtained based on the number of repetitions of the web page is not necessarily the hot news that network users pay attention to, and the manual operation higher cost

Images