Design a collaborative filtering-based recommendation system for an e-commerce platform specializing in [PRODUCT_CATEGORY]. The system should handle [NUMBER_OF_USERS] users and [NUMBER_OF_ITEMS] items, with a focus on addressing the [SPECIFIC_CHALLENGE] (e.g., cold-start problem, scalability, or data sparsity). Provide a step-by-step implementation plan including: 1. Data preprocessing steps for [TYPE_OF_DATA] (e.g., implicit/explicit feedback) 2. Choice of similarity metric (e.g., cosine, Pearson) and justification 3. Model evaluation using [METRIC] (e.g., RMSE, precision@k) 4. Scalability optimization for [PLATFORM] (e.g., AWS, on-premise) Include code snippets in [LANGUAGE] for key components and explain how to handle [EDGE_CASE].