Abstract:Abstract:A Recommender system (RS) is an information filtering software that helps users with a personalized manner to recommend online products to Users and give suggestions about the products that he or she might like. In e-commerce, collaborative Movie recommender system assist users to select their favorite movies based on their similar neighbor's movie ratings. However due to data sparsity and scalability problems, neighborhood selection is more challenging with the rapid increasing number of users and mo… Show more
“…It is used to estimate the average absolute deviation between the actual and the predicted rating values. A lower MAE provides good recommendation quality [35]. The formula for calculating MAE is defined in (8).…”
A recommendation system is a software used in the e-commerce field that provides recommendations for customers to choose the items they like. Several recommendation systems have been proposed; however, collaborative filtering is the most widely used approach. The main issue in collaborative filtering is how to implement a similarity algorithm that can improve performance in the recommendation system. Several similarity algorithms based on user rating value have been developed, and recently a similarity algorithm has been developed that combines the user rating value and the user behavior value. However, the existing research is still based only on a single user behavior value, which is the genre data. Therefore, we propose a new similarity algorithm that considers not only the genre data but also the user profile data (namely age, gender, occupation, and location). The new similarity we are proposing is called User Profile Correlation-based Similarity (UPCSim). The user profile correlation similarity was obtained by calculating the correlation coefficient between the user profile data and the user rating or behavior values. An experiment was done to compare the accuracy of the UPCSim algorithm with that of the previous algorithm. The experiment results show that the UPCSim algorithm can improve the recommendation performance MAE by 1.64% and RMSE by 1.4% compared to the previous algorithm.
“…It is used to estimate the average absolute deviation between the actual and the predicted rating values. A lower MAE provides good recommendation quality [35]. The formula for calculating MAE is defined in (8).…”
A recommendation system is a software used in the e-commerce field that provides recommendations for customers to choose the items they like. Several recommendation systems have been proposed; however, collaborative filtering is the most widely used approach. The main issue in collaborative filtering is how to implement a similarity algorithm that can improve performance in the recommendation system. Several similarity algorithms based on user rating value have been developed, and recently a similarity algorithm has been developed that combines the user rating value and the user behavior value. However, the existing research is still based only on a single user behavior value, which is the genre data. Therefore, we propose a new similarity algorithm that considers not only the genre data but also the user profile data (namely age, gender, occupation, and location). The new similarity we are proposing is called User Profile Correlation-based Similarity (UPCSim). The user profile correlation similarity was obtained by calculating the correlation coefficient between the user profile data and the user rating or behavior values. An experiment was done to compare the accuracy of the UPCSim algorithm with that of the previous algorithm. The experiment results show that the UPCSim algorithm can improve the recommendation performance MAE by 1.64% and RMSE by 1.4% compared to the previous algorithm.
“…Their model's result shows that their system was achieved a better performance compared to other existing methods. On the other hand, Vimala Vellacichamy and Vivekanandan Kalimuthu [20], have also proposed a collaborative recommender system to reduce a data Sparsity and Scalability issues. They apply FCM (Fuzzy C-Mean) technique to cluster users into different groups and Bat optimization to obtain the right number of clusters initially.…”
Section: Related Workmentioning
confidence: 99%
“…They apply FCM (Fuzzy C-Mean) technique to cluster users into different groups and Bat optimization to obtain the right number of clusters initially. According to [20], Fuzzy Bat Clustering method is performed in two steps. In the first step, FCM groups users into different groups based on their appreciation they have given for each item in the past.…”
Section: Related Workmentioning
confidence: 99%
“…Understanding the online user's needs and desires is viewed as a significant for the present customer situated electronic business showcase. So, to overcome the problems facing in traditional recommender systems, a lot of research papers [ 6,8,11,12,13,14,18,19,20], have been done in CF by combining traditional recommendation approaches with that of modern recommendation approaches such as semantic based approaches and cross domain based approaches. The rest of this paper work is arranged as pursues: Section 2 describes related work.…”
A recommender framework is a data refining engines that seeks to foresee the rating for customers and things from enormous information to suggest their preferences. Movie suggestion frameworks give a system to help customers in arranging customers with practically identical interests. This causes a recommender framework basically a focal piece of sites and internet business application. In this study, we have developed a collaborative movie recommender system using crow search and K-means algorithm. This article centers on the movie suggestion proposal frameworks whose essential goal is to recommend a recommender framework through information bunching and computational insight. We have used Elbow method and Silhouette score to select right k number of clusters and calculate errors in each cluster respectively. We have used evaluation metrics standard deviation, mean absolute error, and root mean absolute error to evaluate the performance of the proposed system. The experiment result shows 0.635 MAE and 0.758 RMSE which indicates that our framework accomplished better execution contrast with other existing approaches.
“…If there was not enough information to build a solid profile for a user, the recommendation could not be provided correctly. V. Vellaichamy, and V. Kalimuthu, [17] developed a hybrid collaborative movie recommender system, which was the combination of both Fuzzy C Means clustering (FCM) and bat optimization technique. This combined methodology improves the recommendation quality.…”
Web page recommendation systems are used to recommend the future web page views to World Wide Web (WWW) users. Existing personalized web page recommendation systems are still limited to several problems such as, cold start, sparse data structures (Sparsity), and no diversity in the set of recommended items (Content Overspecialization). In order to overcome these difficulties, we proposed a new personalized web recommendation system by considering the different contributions of the training samples (MSNBC dataset, and castings technology international dataset). At first, the web pages are converted into number of sequences. An effective clustering algorithm called as Possibilistic Fuzzy C-Means (PFCM) is employed to identify the neighbourhood of each user profile. Then the association rule mining technique, Rapid Frequent Pattern Growth (RFPG) algorithm is used for mining the frequent web pages. Finally, the experimental outcome shows that the proposed approach delivers the high priority of web pages and also recommends the related web pages. Finally, the experimental outcome shows that the proposed approach improved accuracy in web page recommendation up to 50-60% compared to the existing methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.