import numpy as np
from sklearn.metrics.pairwise import cosine_similarity

# User-Item rating matrix (rows=users, cols=items)
ratings = np.array([
    [5, 3, 0, 1, 4],
    [4, 0, 0, 1, 3],
    [0, 1, 2, 4, 0],
    [1, 0, 3, 5, 1],
    [0, 4, 0, 0, 5]
])

# Compute user similarity
user_sim = cosine_similarity(ratings)
print("User Similarity Matrix:")
print(np.round(user_sim, 2))

# Predict rating for user 0, item 2 (currently 0)
def predict_rating(user_id, item_id, ratings, similarity, k=3):
    # Find k most similar users who rated this item
    item_ratings = ratings[:, item_id]
    rated_mask = item_ratings > 0
    
    if not rated_mask.any():
        return ratings[user_id].mean()
    
    sim_scores = similarity[user_id]
    sim_scores[~rated_mask] = 0
    
    # Get top-k similar users
    top_k_idx = np.argsort(sim_scores)[-k:]
    top_k_sims = sim_scores[top_k_idx]
    top_k_ratings = item_ratings[top_k_idx]
    
    # Weighted average
    if top_k_sims.sum() == 0:
        return ratings[user_id].mean()
    
    return np.dot(top_k_sims, top_k_ratings) / top_k_sims.sum()

pred = predict_rating(0, 2, ratings, user_sim)
print(f"Predicted rating for user 0, item 2: {pred:.2f}")

import numpy as np
from sklearn.metrics import precision_score, recall_score

def precision_at_k(actual, predicted, k):
    """Calculate Precision@K"""
    predicted_k = predicted[:k]
    relevant = set(actual)
    hits = len(set(predicted_k) & relevant)
    return hits / k

def recall_at_k(actual, predicted, k):
    """Calculate Recall@K"""
    predicted_k = predicted[:k]
    relevant = set(actual)
    hits = len(set(predicted_k) & relevant)
    return hits / len(relevant) if len(relevant) > 0 else 0

def ndcg_at_k(actual, predicted, k):
    """Calculate NDCG@K"""
    dcg = sum(1 / np.log2(i + 2) for i, item in enumerate(predicted[:k]) if item in actual)
    ideal = sum(1 / np.log2(i + 2) for i in range(min(len(actual), k)))
    return dcg / ideal if ideal > 0 else 0

# Example
actual_relevant = [1, 2, 3, 4, 5]
predicted = [1, 6, 3, 7, 8, 2, 9]

print(f"Precision@3: {precision_at_k(actual_relevant, predicted, 3):.3f}")
print(f"Recall@5: {recall_at_k(actual_relevant, predicted, 5):.3f}")
print(f"NDCG@5: {ndcg_at_k(actual_relevant, predicted, 5):.3f}")