Research On Collaborative Filtering Recommendation Technology For Resisting Random Attacks

With the rapid development of the Internet,people have entered the era of data explosion.The amount of data can represent the development ability of an enterprise,and the quality of data can promote the updated iteration of enterprise products.Facing such massive data,how to quickly provide users with useful,key and interesting information becomes particularly important,so the recommendation system emerges at the historic moment.Collaborative filtering recommendation system is one of the most widely used technologies in the recommendation field.This dissertation studies the traditional Collaborative Filtering recommendation algorithm and the K nearest neighbours(KNN)attacks problems it faces.To gain a balance between the ability to resist attacks and the recommendation precision in collaborative filtering recommendation systems,we design a collaborative filtering recommendation algorithm for resisting random attacks,termed Resist the Random Attacks on Collaborative Filtering,or RRACF in short.Then we use Elasticsearch to design a new algorithm,Resist Random Attacks on Collaborative Filtering with Elastic Search,named as ES-RRACF.The RRACF algorithm resists random attacks through using the z-score data standardization technology,the user's interest drift weight and the Laplace noise mechanism;meanwhile it has good recommendation accuracy.Based on RRACF,the ES-RRACF algorithm uses the Elasticsearch technology instead of traditional methods to select the nearest neighbours set.This makes ES-RRACF save a lot of computing time.In order to verify the effectiveness and practicability of the algorithm,the MovieLens data set is used in this dissertation for experimental test.The ES-RRACF algorithm is deployed in a prototype film recommendation system for testing.Results show that ES-RRACF has practical value;it can make the recommendation system more robust.