A Personalized Location Recommendation System Related to Geographic Features

Abstract

The invention discloses an individual position recommending system related to geographical features, and mainly aims to overcome the defects that the conventional collaborative filtering recommendation algorithm is poor in performance in a data sparseness scene. According to the technical scheme, different functional modules are coordinated with one another to achieve recommendation: a database acquiring module is adopted to acquire target information; a user database module is adopted to save target information in a user database; a user preference digging module is adopted to acquire a candidate position recommending list which is ordered according to user preference from a user database; a geographical database module is adopted to store the target information in a geographical database; a geographical feature digging module is adopted to acquire the candidate position recommending list which is ordered according to geographical features from the geographical database; a recommending module is adopted to acquire recommendation results by using the recommending list ordered according to user preference and the recommending list ordered according to geographical feature influence. As geographical features are taken into account, the problem of data sparseness is alleviated, and the individual position recommending system can be used in site push service based on a position social network.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to a personalized recommendation technology related to geographic features, which can be used for location push services in location-based social networks. Background technique [0002] With the rapid development of database technology and the wide application of database management systems, more and more data are accumulated. Faced with the surge of data, people hope to dig out many important information hidden behind, so that these data can be better used to serve people. Service providers have a large amount of personal information and historical records related to users. Using these data to actively recommend relevant services to users can not only facilitate users, but also benefit service providers. [0003] Traditional recommendation system algorithms mainly use recommendation algorithms based on collaborative filtering and content-based recommendation al...

AI Technical Summary

Problems solved by technology

[0006] In the above methods, since the influence of geographical features on location recommendation for users is not considered, the performance is very poor in data sparse scenarios, such as location recommendation in different places, and it is also unable to deal with the cold start problem of new users and new items. effective solution

Images