Social network user behavior relation deduction system and method based on indefinite semantics

Abstract

The invention discloses a social network user behavior relation deduction system and method based on indefinite semantics and belongs to the field of data mining and service information recommendation. According to the social network user behavior relation deduction system and method based on indefinite semantics, whether a user is in a behavior relation or not is deducted according to user geographic position similarity and user activity similarity; users in behavior relation with the user are inferred according to the geographic position similarity and the activity similarity, so an unknown user behavior can be predicted accurately, and the quality of information recommendation service can be greatly improved. Experiments prove that the social network user behavior relation deduction method is superior to an existing behavior prediction method in terms of unknown user behavior prediction accuracy and has very high practical application value, so if the method can be popularized, the accuracy of user behavior prediction will be greatly improved, and the method has prominent significance for an enterprise to clarify a target customer group and make a correct decision.

Description

technical field [0001] The invention belongs to the fields of data mining and service information recommendation, and in particular relates to a system and method for deducing social network user behavior relationships based on uncertain semantics. Background technique [0002] Online social networks have become an important platform for users to connect with each other and share various information, such as Twitter, Facebook, Instagram and other well-known social networks. Among them, Twitter has become the mainstream social media abroad because of its concise content and powerful real-time user interaction. [0003] The "" character in Tweet (Twitter) means "mention", which means "to tell someone". It is not difficult to see that the "" operation is one of the effective ways to reveal the close relationship between users. However, after reviewing the information, I learned that the current research on social relations in academia and industry, on the one hand, is limited ...

AI Technical Summary

Problems solved by technology

However, after reviewing the information, I learned that the current research on social relations in academia and industry, on the one hand, is limited to the "follow / be followed" relationship, or the "Retweet relationship", that is, by calculating user similarity and analyzing network topology. , Measuring user interactivity and other means to analyze static social network user relationships and their strengths, lack of in-depth discussion on the "" relationship, but the fact is that in terms of reflecting the closeness of users, the "" relationship is more important than the "following / being followed" relationship Stronger, because you can only "follow" a certain user, so the current social network relationship research misses the mining of closer similar semantic behavior relationships

On the other hand, although some research works have proposed location-related user relationship inference, they only studied the simple explicit connection relationship of "follow" and "push", and have more complex implicit semantic connections between users. , such as catering, travel, shopping and other semantic relationships reflected by "activities" have not been discussed. In fact, similar users can be found more accurately by using the semantic information embodied by location, but this point has been ignored.

To sum up, the research on social network user behavior relationship is not deep enough and perfect, and the behavior relationship mining technology is not comprehensive. These deficiencies eventually lead to inaccurate information recommendation service objects.

Images