A Graph Based Ranking Algorithm for Recommender Systems
سال انتشار: 1398
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 581
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
RSETCONF01_014
تاریخ نمایه سازی: 17 فروردین 1399
چکیده مقاله:
Recommender systems are an emerging technology that helps consumers to find interesting products. A recommender system makes personalized product suggestions by extracting knowledge from the previous users’ interactions and ratings on products. Such services are particularly useful in the modern electronic marketplace which offers an unprecedented range of products which makes it impossible for any user to explore all of them before making decisions and increases the importance of these systems as a forecasting device. However, a recommendation system normally requires comprehensive data relating users and products. Insufficiently comprehensive data creates difficulties for creating good recommendations. One approach to solve this problem is to use random walk with restart (RWR), which significantly reduces the quantity of data required and has been shown to outperform collaborative filtering which is the competitor approach in the field of recommendation systems. This study explores how to increase the accuracy of the Random Walk approach in a recommendation system graph model by changing the rating mechanism. In our approach we replace the conventional rating systems with a new system which is capable of detecting negative opinions as well as positive ones and gathering both like/unlike opinions in the process of rating by users. After that we design a scoring algorithm, which can be used to rank products according to expected user preferences, in order to recommend top–rank items to potentially interested users. We tested our algorithm on a standard database, the MovieLens data set, which contains data collected from a popular recommender system on movies, and we compared our algorithm with other popular ranking techniques. The result shows that our approach provides better recommendations with less memory and time complexities.
کلیدواژه ها:
نویسندگان
Zainabolhoda Heshmati
Faculty of New Sciences and Technologies, University of Tehran
Yasamin Sadat
Faculty of New Sciences and Technologies, University of Tehran