Font Size: a A A

Recommendation Model Of Mathematical Expressions Based On Collaborative Filtering

Posted on:2019-04-15Degree:MasterType:Thesis
Country:ChinaCandidate:C Y WangFull Text:PDF
GTID:2428330569479255Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
In view of the urgent need for users to obtain scientific information in the environment of rapid information growth,and aiming at the shortage of present mathematical expression retrieval technology in providing personalized recommendation services for users,a mathematical expression recommendation method based on collaborative filtering is proposed in order to improve the performance of the mathematical expression retrieval system,On the basis of the introduction and analysis of mathematical expression retrieval,personalized recommendation,fuzzy set theory and the development of technology,a collaborative filtering method is used to study and design the model of mathematical expression.First,based on the analysis of mathematical expression characteristics and according to the structural level characteristics of mathematical expressions,a mathematical expression subtype position selection mode is added to the input end of mathematical expression to make the retrieval results close to the user preferences.Secondly,the fuzzy set method is used to model the user 's preference by multi-feature index,and the rating matrix of the user to mathematical expressions is formed according to the membership function of the user's behavior.Last,according to the idea of collaborative filtering algorithm,we measure users' similarity by using close degree to generate the nearest neighbor set.Then,the recommended list of mathematical expressions is generated by setting thresholds so that it can achieve collaborative filtering recommendation of mathematical expressions.The experimental result shows that the method proposed in this thesis is feasible and effective.
Keywords/Search Tags:Mathematical expressions, Recommendation, Collaborative filtering, User modeling, Multi-feature fuzzy sets, Close degree, Normalization
PDF Full Text Request
Related items