List of Articles Cold Start.

• Open Access Article
  • Abstract Page
  • Full-Text
  
  1 - A Trust-based Recommender System Using an Improved Particle Swarm Optimization Algorithm
  Sajad Ahmadian Mohammad Hossein Olyaee
  Introduction: Recommender systems are intelligent tools to help users find their desired information among a large number of choices based on their previous preferences in a way faster than search engines. One of the main challenges in recommender systems is the sparsit More

  Introduction: Recommender systems are intelligent tools to help users find their desired information among a large number of choices based on their previous preferences in a way faster than search engines. One of the main challenges in recommender systems is the sparsity of the user-item rating matrix. This means that users mainly tend to express their opinions about a few items, leading to a large portion of the user-item rating matrix being empty. Trust-based recommender systems aim to alleviate the sparsity problem using trust relationships between users. Trust relationships can be used to calculate similarity values between users and determine the nearest neighbors set for the target user. However, the efficiency of trust-based recommender systems depends on the correct selection of neighboring users for the target user based on the similarity values between users. Method: In this paper, a novel trust-based recommender system is proposed based on an improved particle swarm optimization algorithm. To this end, first, the similarity values between users are calculated based on the user-item rating matrix and trust relationships. Then, the improved particle swarm optimization algorithm is used to optimally weight the neighboring users of the target user. The main purpose of this algorithm is to assign an optimal weight to each user in the nearest neighbor set of the target user to predict the unknown items accurately. After the optimal weighting of neighboring users, unknown ratings are predicted for the target user. Results: The proposed method is evaluated on a standard dataset in terms of mean absolute error, root mean square error, and rate coverage metrics. Experimental results demonstrate the high efficiency of the proposed method compared to other methods. Discussion: We use the genetic algorithms operators and chaos-based asexual reproduction optimization algorithm to improve the original version of the particle swarm optimization algorithm. The genetic algorithms operators increase the exploration mechanism of the particle swarm optimization algorithm, leading to a decline in the probability of tapping into local optima. Moreover, the chaos-based asexual reproduction optimization algorithm is applied to the best solution to further search the area around the best solution. Manuscript profile