نوع مقاله : مقاله علمی
نویسندگان
1 گروه مدیریت فناوری اطلاعات، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی، تهران، ایران،
2 گروه مدیریت فناوری اطلاعات، واحد علوم و تحقیقات، دانشگاه آزاد اسلامی، تهران، ایران
چکیده
کلیدواژه‌ها
عنوان مقاله [English]
نویسندگان [English]
To provide location-based services in conventional social networks, the gap between these networks and the physical world must be bridged by adding the location dimension. One of the solutions developed to add the location dimension to the social network structure is location-based social networks. Today, the use of recommender systems has enabled the discovery and search for places of interest to individuals in the shortest possible time. The fundamental principle of location-based recommender systems is determining the degree of similarity between users. In this research, two criteria were presented for determining the spatial similarity of users, considering the security aspects of social data. The first criterion, by focusing on common locations and visit history, divides shared visits into two groups, simultaneous and non-simultaneous, and calculates their combined average. The second proposed criterion, by integrating three key measures—custom similarity (focusing on spatial-temporal commonalities), cosine similarity (analyzing movement direction), and the Pearson correlation coefficient (measuring linear behavioral relationships)—enables a multi-dimensional evaluation of user behavior. The proposed criteria were compared with common similarity measurement methods, including the Pearson correlation coefficient, cosine similarity, and the Jaccard coefficient. The comparison results show that both proposed criteria perform better than the Jaccard coefficient, but they demonstrate weaker performance compared to cosine similarity and the Pearson coefficient. However, to assess the accuracy of the proposed criteria using the mean absolute error and root mean square error metrics, the results indicate a significant superiority of the proposed methods.
کلیدواژه‌ها [English]
)