Font Size: a A A

Intelligent Dating Site Design & Implementation Based On The Item-based Recommended/ Collaborative Filtering Recommendation

Posted on:2016-08-11Degree:MasterType:Thesis
Country:ChinaCandidate:S H NieFull Text:PDF
GTID:2308330464972735Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
With the development of network technology in the past, information on the net growth like a bomb for decades,what make kinds of problem like "Information Overload" and "Information Trek",In the era of information society and network,the Internet provide more and more information and services for human,it has broken the limits of space and time like traditional life and learning,shopping or chatting on the internet is so convenient for us now,but we have to face voluminous data and spam at the same time.The Intelligent recommendation systems arises at the historic moment for the issues above.Based on the study of the theory of intelligent recommendation algorithm a intelligent dating site was be designed and implement,the main work as follows:Firstly,introduce the basic model of intelligent recommendation system and the five facts what will affect the performance,analysis the item-based recommendation and the collaborative filtering recommendation algorithm,and their respective advantages and disadvantages.In addition,we discuss the basic theory about others and illustrates the significance about why we should use hybrid recommendation algorithm in a actual project.Secondly,Demand analysis for the project.In terms of functional,we provide a use case diagram and analysis all of the function modules;and discuss the concurrency,security,the data capture performance and some other performance.Thirdly,Design the On the database tables for the implementation of the hybrid algorithm and then provide a ER diagram of the system.A hybrid algorithm was be designed based on the tables.in this algorithm,based on the item-based recommendation algorithm as assistant to fill the data to solve date problems about data sparseness and mitigation system cold start;the collaborative filtering recommendation as the system core algorithm to calculation the similarity between user based on the cosine similarity algorithm.The hybrid algorithm is come to be simple mode,fast response and highly precision.especially suitable for the small and medium-sized system.Finally,we do the detailed design for the system,what about the system frame,the function structure and the chart for each function model;briefly introduce the development tools and some related technologies.Eventually,we implement a intelligentdating site base on the SSH frame.after testing,most of the functions can be run in a correct,stability and security way,the system is come to be a qualified intelligence dating site.
Keywords/Search Tags:Item-based recommend, Collaborative filtering recommendation, Dating site, SSH
PDF Full Text Request
Related items