Search

Music recommender systems is an important field of research because of easy availability and use of online music. The most existing models only focus on explicit data like ratings and other user-item dimensions. A challenging problem in music recommendation is to model a variety of contextual information, such as feedback, time and location. In this article, we proposed a competent hybrid music recommender system (HMRS), which works on context and collaborative approaches. The timestamp is extracted from users listening log to construct a decision context behavior that extracted various temporal features like a week, sessions(as morning, evening or night). We used depth-first-search (DFS) algorithm which traverses the whole graph through the paths in different contexts. Bellman-Ford algorithm provides ranked list of recommended items with multi-layer context graph. We enhanced the process using particle swarm optimization (PSO) which produced highly optimized results. The dataset is used from Last.fm which contains 19,150,868 music listening logs of 992 users (till May, 4th 2009). We extract the properties of music from user's listening history and evaluate the efficient system to recommend music based on user's contextual preferences. Our system noticeably delivers the best recommendations regarding recall results when compared to existing methods. [ABSTRACT FROM AUTHOR]

Recommender systems play a significant role in e-commerce applications. The primary motive of a recommender system is to recommend some items or products to the users based on their previous ratings of other products in the online environment. In this article, we presented a hybrid collaborative filtering based recommender system that improved the accuracy of the recommendations. In our work, we adopted fuzzy c-mean (FCM) and a recent bio-inspired approach, which is artificial algae algorithm (AAA). We have used advanced multilevel Pearson correlation coefficient (PCC) to find the similarity between two users. Moreover, we discovered the rating which the user will most likely give to the movies which he has not given any ratings yet. By applying above-mentioned procedures, the quality of the recommendations is improved significantly. The proposed system succeeded to provide recommendations of better quality and accuracy when compared to other alternatives. We have experimented and evaluated our proposed recommender system on four real data sets: Movielens 100,000, Movielens 1 million, Jester and Epinion. We concluded that our proposed recommender system delivered better recommendations for all four datasets. The efficiency of the system was estimated by evaluation metrics such as mean absolute error (MAE), precision and recall and showed impressive results. This proposed system delivered best results as compared to our previous work (Katarya and Verma, 2016) [1]. [ABSTRACT FROM AUTHOR]

Recommender systems are information filtering tools that aspire to predict the rating for users and items, predominantly from big data to recommend their likes. Movie recommendation systems provide a mechanism to assist users in classifying users with similar interests. This makes recommender systems essentially a central part of websites and e-commerce applications. This article focuses on the movie recommendation systems whose primary objective is to suggest a recommender system through data clustering and computational intelligence. In this research article, a novel recommender system has been discussed which makes use of k-means clustering by adopting cuckoo search optimization algorithm applied on the Movielens dataset. Our approach has been explained systematically, and the subsequent results have been discussed. It is also compared with existing approaches, and the results have been analyzed and interpreted. Evaluation metrics such as mean absolute error (MAE), standard deviation (SD), root mean square error (RMSE) and t-value for the movie recommender system delivers better results as our approach offers lesser value of the mean absolute error, standard deviation, and root mean square error. The experiment results obtained on Movielens dataset stipulate that the proposed approach may provide high performance regarding reliability, efficiency and delivers accurate personalized movie recommendations when compared with existing methods. Our proposed system (K-mean Cuckoo) has 0.68 MAE, which is superior to existing work (0.78 MAE) [1] and also has improvement of our previous work (0.75 MAE) [2] . [ABSTRACT FROM AUTHOR]

In a web environment, one of the most evolving application is those with recommendation system (RS). It is a subset of information filtering systems wherein, information about certain products or services or a person are categorized and are recommended for the concerned individual. Most of the authors designed collaborative movie recommendation system by using K-NN and K-means but due to a huge increase in movies and users quantity, the neighbour selection is getting more problematic. We propose a hybrid model based on movie recommender system which utilizes type division method and classified the types of the movie according to users which results reduce computation complexity. K-Means provides initial parameters to particle swarm optimization (PSO) so as to improve its performance. PSO provides initial seed and optimizes fuzzy c-means (FCM), for soft clustering of data items (users), instead of strict clustering behaviour in K-Means. For proposed model, we first adopted type division method to reduce the dense multidimensional data space. We looked up for techniques, which could give better results than K-Means and found FCM as the solution. Genetic algorithm (GA) has the limitation of unguided mutation. Hence, we used PSO. In this article experiment performed on Movielens dataset illustrated that the proposed model may deliver high performance related to veracity, and deliver more predictable and personalized recommendations. When compared to already existing methods and having 0.78 mean absolute error (MAE), our result is 3.503 % better with 0.75 as the MAE, showed that our approach gives improved results. [ABSTRACT FROM AUTHOR]