The overabundance of information on the internet and ecommerce has resulted to the development of recommender system to discover interesting items or contents that are recommendable to the user. The recommended items might be of no interest or unwanted to the users and can make users to lose interest in the recommendations. In this work, a Collaborative Filtering (CF) based method which exploits the initial top-N recommendation lists of an item-based CF algorithm based on unwanted recommendations penalisation is presented.  The method utilises a relevance feedback mechanism to solicit for users preferences on the recommendations while popularise similarity function minimises the chances of recommending unwanted items. The work explains the proposed algorithm in detail and demonstrates the improvements required on existing CF to provide some adjustments required to improve subsequent recommendations to users. 

The usefulness of Collaborative filtering recommender system is affected by its ability to capture users' preference changes on the recommended items during recommendation process. This makes it easy for the system to satisfy users' interest over time providing good and quality recommendations. The Existing system studied fails to solicit for user inputs on the recommended items and it is also unable to incorporate users' preference changes with time which lead to poor quality recommendations. In this work, an Enhanced Movie Recommender system that recommends movies to users is presented to improve the quality of recommendations. The system solicits for users' inputs to create a user profiles. It then incorporates a set of new features (such as age and genre) to be able to predict user's preference changes with time. This enabled it to recommend movies to the users based on users new preferences. The experimental study conducted on Netflix and Movielens datasets demonstrated that, compared to the existing work, the proposed work improved the recommendation results to the users based on the values of Precision and RMSE obtained in this study which in turn returns good recommendations to the users.

Recommender Systems are systems that aid users in finding relevant items, products, or services, usually in an online setting. Collaborative Filtering is the most popular approach for building recommender system due to its superior performance. There are several collaborative filtering methods developed, however, all of them have an inherent problem of data sparsity. Covering Reduction Collaborative Filtering (CRCF) is a new collaborative filtering method developed to solve the problem. CRCF has a key feature called popular items extraction algorithm which produces a list of items with the most ratings, however, the algorithm fails in a denser dataset because it allows any item to be in the list. Likewise, the algorithm does not consider the rating values of items while considering the popular items. These make it to produce less accurate recommendation. This research extends CRCF by developing a new popular item extraction algorithm that removes items with low modal ratings and similarly utilizes the rating values in considering the popular items. This newly developed method is incorporated in CRCF and the new method is called Improved Popular Items Extraction for Covering Reduction Collaborative Filtering (ICRCF). Experiment was conducted on Movielens-1M and Movielens-10M datasets using precision, recall and f1-score as performance metrics. The result of the experiment shows that the new method, ICRCF provides a better recommendation than the base method CRCF in all the performance metrics. Furthermore, the new method is able to perform well both at higher and lower levels of sparsity.

Collaborative filtering recommender system suffers from data sparsity problem due to its reliance on numerical ratings to provide recommendations to users. This problem makes it difficult for the system to compute accurate similar neighbours for the items and provide good quality recommendations. Existing methods fail to pre-process the missing ratings of the new items and to predict cold items to the active users which lead to poor quality recommendations. In this work, a sparsity reduction method is presented to improve the quality of recommendations. The method utilises Bi-Separated clustering algorithm to cluster the ratings matrix simultaneously into users and items bi-clusters based on ratings classification. It also employs Bi-Mean Imputation algorithm to fill the missing ratings in the bi-clusters using the estimated means. The method then performs the traditional collaborative filtering process on the new rating matrix for cold items prediction. The experimental results demonstrated that compared to the existing method, the proposed BiSCBiMI improves density of the rating matrix by 5.75%, 10.73% and 7.35% as well as Mean Absolute Error (MAE)  of the new items prediction for all of the considered datasets.  The results indicated that, the proposed approaches are effective in reducing the data sparsity problem as well as items prediction, which in turn returns good quality recommendations.