基于python和协同过滤推荐算法的电影评分预测系统源码.zip

import pandas as pd import numpy as np DATA_PATH = "ratings.csv" def load_data(data_path): ''' 加载数据 return:用户-物品评分矩阵 ''' # 设置要加载的数据字段类型 dtype = {"userId": np.int32, "movieId": np.int32, "rating": np.float32} # 加载数据（前三列），用户Id，电影Id，以及用户对电影的对应评分 ratings = pd.read_csv(data_path, dtype=dtype, usecols=range(3)) # 透视表（User-Movie的评分矩阵） ratings_matrix = ratings.pivot_table(index=["userId"], columns=["movieId"], values="rating") print("数据集加载完毕") return ratings_matrix if __name__ == '__main__': ratings_matrix = load_data(DATA_PATH) print("ratings_matrix:\n", ratings_matrix) def compute_pearson_similarity(ratings_matrix, based="user"): ''' 计算皮尔逊相关系数 ratings_matrix:用户-物品评分矩阵 based:user、item return:相似度矩阵 ''' # 基于皮尔逊相关系数计算相似度 # 用户相似度 if based == "user": print("开始计算用户相似度矩阵") similarity = ratings_matrix.T.corr() else: print("开始计算物品相似度矩阵") similarity = ratings_matrix.corr() print("相似度矩阵计算/加载完毕") return similarity if __name__ == '__main__': ratings_matrix = load_data(DATA_PATH) user_similar = compute_pearson_similarity(ratings_matrix, based="user") print("user_similar:\n", user_similar) item_similar = compute_pearson_similarity(ratings_matrix, based="item") print("item_similar:\n", item_similar) # 封装函数：User-Based预测给定用户对给定物品的评分值 # 方法说明：利用原始评分矩阵以及物品间两两相似度，预测指定用户对指定物品的评分。 # 如果无法预测，则抛出异常 def predict(uid, iid, ratings_matrix, user_similar): print("开始预测用户<%d>对物品<%d>的评分..." % (uid, iid)) # 1.找出uid用户的相似用户 similar_users = user_similar[uid].drop([uid]).dropna() # 筛选正相关用户 similar_users = similar_users.where(similar_users > 0).dropna() if similar_users.empty is True: raise Exception("用户<%d>没有相似用户" % uid) # 2.从uid用户的近邻相似用户中筛选出对iid物品有评分记录的近邻用户 ids = set(ratings_matrix[iid].dropna().index) & set(similar_users.index) finally_similar_users = similar_users.loc[list(ids)] # 3.结合uid用户与其近邻用户的相似度预测uid用户对iid物品的评价 sum_up = 0 # 评分预测公式分子值 sum_down = 0 # 评分预测公式分母值 for sim_uid, similarity in finally_similar_users.iteritems(): # 近邻用户的评分数据 sim_user_rated_movies = ratings_matrix.loc[sim_uid].dropna() # 近邻用户对iid物品的评分 sim_user_rating_for_item = sim_user_rated_movies[iid] # 计算分子的值 sum_up += similarity * sim_user_rating_for_item # 计算分母的值 sum_down += similarity # 计算预测的评分值并返回 predict_rating = sum_up / sum_down print("预测用户<%d>对电影<%d>的评分为：%0.2f" % (uid, iid, predict_rating)) return round(predict_rating, 2) # Item-Based 预测给定用户对给定物品的评分值 def predict(uid, iid, ratings_matrix, user_similar): print("开始预测用户<%d>对物品<%d>的评分..." % (uid, iid)) # 1.找出id物品的相似物品 similar_items = item_similar[iid].drop([iid]).dropna() # 筛选正相关物品 similar_items = similar_items.where(similar_items > 0).dropna() if similar_items.empty is True: raise Exception("物品<%d>没有相似物品" % iid) # 2.从iid物品的近邻相似物品中筛选uid用户评分过的物品 ids = set(ratings_matrix[uid].dropna().index) & set(similar_items.index) finally_similar_items = similar_items.loc[list(ids)] # 3.结合iid物品与其相似物品的相似度和uid用户对其相似物品的评分，预测uid对iid的评分 sum_up = 0 # 评分预测公式分子值 sum_down = 0 # 评分预测公式分母值 for sim_iid, similarity in finally_similar_items.iteritems(): # 近邻物品的评分数据 sim_item_rated_movies = ratings_matrix.loc[sim_iid].dropna() # uid用户对相似物品的评分 sim_item_rating_for_user = sim_item_rated_movies[uid] # 计算分子的值 sum_up += similarity * sim_item_rating_for_user # 计算分母的值 sum_down += similarity # 计算预测的评分值并返回 predict_rating = sum_up / sum_down print("预测用户<%d>对电影<%d>的评分为：%0.2f" % (uid, iid, predict_rating)) return round(predict_rating, 2) if __name__ == '__main__': # ratings_matrix = load_data(DATA_PATH) user_similar = compute_pearson_similarity(ratings_matrix, based="user") # 预测用户1对物品1的评分 predict(1, 1, ratings_matrix, item_similar) # 预测用户1对物品2的评分 predict(1, 2, ratings_matrix, item_similar) def predict_all(uid, ratings_matrix, user_similar): ''' 预测全部评分 uid：用户id ratings_matrix:用户-物品打分矩阵 user_similar:用户两两间相似度 return:逐个返回预测评分 ''' # 准备要预测的物品的id列表 item_ids = ratings_matrix.columns # 逐个预测 for iid in item_ids: try: rating = predict(uid, iid, ratings_matrix, user_similar) except Exception as e: print(e) else: yield uid, iid, rating if __name__ == '__main__': ratings_matrix = load_data(DATA_PATH) user_similar = compute_pearson_similarity(ratings_matrix, based="user") for i in predict_all(1, ratings_matrix, user_similar): pass ''' 添加过滤规则 ''' def _predict_all(uid, item_ids, ratings_matrix, user_similar): ''' 预测全部评分 :param uid: 用户id :param item_ids: 要预测的物品id列表 :param ratings_matrix: 用户-物品评分矩阵 :param user_similar: 用户两两间相似度 :return: 逐个返回预测评分 ''' # 逐个预测 for iid in item_ids: try: rating = predict(uid, iid, ratings_matrix, user_similar) except Exception as e: print(e) else: yield uid, iid, rating def predict_all(uid, ratings_matrix, user_similar, filter_rule=None): ''' 预测全部评分，根据条件进行前置过滤 uid:用户id ratings_matrix:用户-物品评分矩阵 user_similar:用户间两两相似度 filter_rule:过滤规则（只能四选一，否则抛出异常:“unhot”，“rated",[”unhot“，”rated“]，None return：逐个返回预测评分 ''' if not filter_rule: item_ids = ratings_matrix.columns elif isinstance(filter_rule, str) and filter_rule == "unhot": # 过滤非热门电影 # 统计每部电影的评分数 count = ratings_matrix.count() # 过滤出评分数高于10的电影，作为热门电影 item_ids = count.where(count > 10).dropna().index elif isinstance(filter_rule, str) and filter_rule == "rated": # 过滤用户评分过的电影 # 获取用户对所有电影的评分记录 user_ratings = ratings_matrix.loc[uid] # 评分范围是1-5，小于6都是评分过的，除此以外是没评分过的 _ = user_ratings < 6 item_ids = _.where(_ == False).dropna().index elif isinstance(filter_rule, list) and filter_rule == ["unhot", "rated"]: # 过滤非热门和用户已经评分过的电影 count = ratings_matrix.count() ids1 = count.where(count > 10).dropna().index user_ratings = ratings_matrix.loc[uid]