Information pushing method and device

Abstract

The invention discloses an information pushing method. The method comprises the steps as follows: obtaining scoring parameters corresponding to user historical operations in a preset duration for the same type of pushed information, and meanwhile, obtaining similarity parameters between to-be-pushed information and information corresponding to the user historical operations; according to a preset diversity regulation factor and a preset accuracy regulation factor, calculating pushing scores corresponding to the to-be-pushed information according to a preset algorithm by using the obtained scoring parameters and the similarity parameters; pushing the corresponding to-be-pushed information according to the obtained pushing scores. The invention further discloses an information pushing device. The information pushing method and device have the benefit that the information is pushed according to the pushing scores due to the fact that the pushing scores of the to-be-pushed information are calculated on the premise that the accuracy and the diversity of the to-be-pushed information are balanced, and both the accuracy and the diversity of the information pushing are considered.

Description

technical field [0001] The invention relates to data processing technology, in particular to an information push method and device. Background technique [0002] At present, music (such as songs) and application programs (such as games) waiting for push information to make personalized recommendations for different users has become one of the focuses of widespread attention in the industry. Before pushing the above information to be pushed, it is usually necessary to treat the push information such as songs , game applications, etc., and then push the above information to be pushed according to the scores. [0003] Before pushing the above-mentioned information to be pushed, the commonly used method of scoring the above-mentioned information to be pushed is: obtain the operation information corresponding to the user's historical operation records and similar information similar to the above-mentioned operation information, and obtain the above-mentioned similar information a...

AI Technical Summary

Problems solved by technology

Since this scoring method is simply calculated based on the user's historical operation records and similarity, it cannot balance the accuracy and diversity of the information to be pushed, so it is easy to make the information with a lower accuracy rate be scored higher instead, resulting in the information being pushed to the user. The information accuracy rate is low, for example, the push information is not what the user is interested in; or, too much pursuit of the accuracy of the scoring, ignoring the diversity of the push information, resulting in the problem of single push information and insufficient diversity

Images