Information overload is a major problem for many internet users which occurs due to overwhelming amounts of data made available to a user. In order to deal with this problem filtering tool, like Recommender System is required for providing relevant information for the users which personalizes the search according to user preferences. The Collaborative Filtering Recommender System finds the nearest neighbour set of active user by using similarity measures on the rating matrix. This paper proposes different variations of K-nearest neighbors (KNN) algorithm with different similarity measures namely cosine, msd, pearson and pearson baseline for Movie Recommender System. These different variations of KNN algorithms have been implemented for real data from MovieLens dataset and compared on accuracy metrics like fraction of concordant Pairs, mean absolute error, mean squared error, root mean squared error, precision@k and recall@k for Movie Recommender System. For real life application, Movie Recommender System filtering tool may be used as plugin by customizing the web browser.

信息过载是许多互联网用户面临的主要问题，这是由于可供用户使用的大量数据所致。为了处理此问题，需要使用Recommender System这样的筛选工具，以便为用户提供相关信息，从而根据用户的喜好对搜索进行个性化设置。协作过滤推荐系统通过使用评分矩阵上的相似性度量来找到活动用户的最近邻居集。本文针对电影推荐系统，提出了不同的K近邻算法（KNN）算法，它们具有不同的相似性度量，即余弦，msd，pearson和pearson基线。已针对MovieLens数据集的真实数据实现了KNN算法的这些不同变体，并在诸如一致性对分数的分数等准确性指标上进行了比较，电影推荐系统的平均绝对误差，均方误差，均方根误差，precision @ k和callback @ k。对于现实生活中的应用，可以通过自定义Web浏览器将Movie Recommender System筛选工具用作插件。