<![CDATA[The rapid growth of digital platforms has led to a massive increase in user-item interaction data, most of which is recorded as implicitfeedback such as purchase history and reading duration. Unlike explicit ratings, implicit feedback presents fundamental challengesincluding extreme sparsity and ambiguity, as non-interaction does not necessarily indicate disinterest. This study develops a Top-Nbook recommendation system by implementing Item-Based Collaborative Filtering (IBCF) as a baseline and Alternating LeastSquares (ALS) as the primary model, addressing the challenge of a highly sparse dataset (13,876 users × 123,069 items). Themethodology involves transforming raw interaction data into Compressed Sparse Row (CSR) format, performing systematichyperparameter tuning on the ALS model, and implementing rigorous data preprocessing techniques to optimize performance.The experimental results demonstrate a progressive improvement trajectory: the baseline IBCF model achieved a Mean AveragePrecision at 10 (MAP@10) score of 0.0186, the initial ALS model improved to 0.0243, hyperparameter tuning further increasedperformance to 0.0257, while the integration of systematic data preprocessing (K-Core filtering) with advanced hyperparametertuning ultimately yielded a score of 0.0622 representing a 234% improvement over the baseline. Qualitative analysis reveals thatwhile IBCF produces monotonous and homogeneous recommendations limited to similar publishers and series, ALSprovides substantially more diverse and exploratory results by capturing latent behavioral patterns across multiple genres andthematic categories. This study concludes that the optimized ALS model, combining matrix factorization with rigorous dataengineering, is significantly more effective than in handling sparse implicit feedback, delivering superior rankingaccuracy, enhanced recommendation diversity, and improved user discovery experience.]]>
