Research On Collaborative Filtering Recommendation Algorithm Based On Artificial Fish Swarm Algorithm

The Internet is closely related to everyday life of each person. As the dataaccumulated unceasingly, the size of the amount of the Internet data is becomingincreasingly overload. It is particularly difficult for people to find their neededinformation from the vast amounts of information. Recommendation technology isapplied to solve this problem. As the most widely used recommendationtechnology which has achieved greater success currently, the collaborativefiltering recommendation technology select users(or items) which have highersimilarity with the target user(or item) as its nearest neighbors according to thetarget user’s(or item’s) accessed data and evaluated information. Then the targetuser’s(or item’s) predicted score can be obtained according to the scores of theneighbors and then items can be recommended to the target user.In practical applications, however, collaborative filtering recommendationis facing many problems such as data sparse, the accuracy of recommendat-ionloss and soon on.In order to respond these problems, this paper is divided into the followingareas:(1) On the basic of former research, to improve the fault of typical AFSA, animproved algorithm was introduced. An improved AFSA was presented bydynamically adjusting the step of artificial fish, which solve execution speedslowly. The improved AFSA of introducing devouring and jumping behavior cansolve easy to fall into local optimum. The improved AFSA has some advantagessuch as faster execution speed, higher precision of solution.(2) In view of the recommendation accuracy and speed problem, theimproved AFSA is applied in CF, which is named an IAFSA CF. First, the IAFSACF fills the vacancy of rating data. Then the IAFSA CF clusters users and selectsthe nearest neighbor of target user. So more accurate neighbor and more preciserecommendation results can be gotten.(3) This paper used the Movie Lens datasets to test and verify the IAFSA CF.The experimental results show that the improved AFSA has faster execution speedand higher precision of solution than AFSA. The improved AFSA is an effectivemethod. The IAFSA CF has faster selecting the nearest neighbor of target userthan CF. The IAFSA CF has higher precision than other recommendationalgorithms.