The Precise And Targeted Advertising Drive Research Of Recommendation System In Mobile Environment

The rapid development of Internet brings people into an age of information explosion. Due to the limitations of people’s attention capacity, it can be difficult to find information or project of interest in massive data information quickly. In addition, with the rapid development of mobile communication technology, the use of mobile terminals such as mobile phones and tablets become more convenient. A large number of web application users has changed from web to mobile client version. There is great potential in making personalized information recommendations to promote user experience for mobile clients.The use of personalized recommendation in the Internet has made certain achievements. The most successful application is collaborative filtering algorithm. However, there are some issues with that, such as data sparseness and cold start. At the same time, due to the mobility of mobile users, the user’s location and the surrounding environment changes in real-time. The traditional collaborative filtering algorithm could not deal with these changes and give precise recommendation well. In order to deal with the above issues, this paper come up with a user clustering method based on the user’s social network information: the user similarity computing considers the user’s interaction information to cluster users’ friends with similar interests. Friends’ recommendation method which is based on the users’ trust will improve the quality of recommendation. At the same time, this paper increased the marking time weight and project distance weighing to improve the precision of the recommended results. Meanwhile, the recommendation algorithm introduced distance selection mechanism and reduced the amount of calculation to improve the efficiency of the recommendation.The improving methods raised in this paper undergo matlab analogue simulation which is based on the dataset of the public comments, and verified the effectiveness of personalized recommendation technology integrated with social information and location information.