Abstract
Recommendation Systems (RS) are essential for personalized item suggestions, but traditional approaches face limitations in accuracy, scalability, efficiency, and cold-start problems. This paper introduces the HRS-IU-DL model, a hybrid RS that leverages advanced techniques to enhance accuracy and relevance. The model integrates user-based and item-based Collaborative Filtering (CF) to analyze user-item interactions, Neural Collaborative Filtering (NCF) to capture non-linear interactions, and Recurrent Neural Networks (RNN) to identify sequential patterns in user behavior. Additionally, Content-Based Filtering (CBF) using Term Frequency-Inverse Document Frequency (TF-IDF) is employed to analyze item attributes. By combining CF, NCF, RNN, and CBF, the HRS-IU-DL model addresses challenges such as data sparsity, the cold-start problem, and the need for personalized recommendations. The proposed model utilizes N-Sample techniques to recommend the top 10 similar items based on user-specified genre, employing methods like Cosine Similarity, Singular Value Decomposition (SVD), and TF-IDF. Evaluation of the HRS-IU-DL model is conducted on the publicly available Movielens 100k dataset using train and test sets. Performance metrics, including Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), Precision, and Recall, are employed to assess the model's performance. The results demonstrate that the HRS-IU-DL model outperforms state-of-the-art methods and achieves significant improvements across these metrics.