协同过滤算法在高校图书馆个性化推荐中的应用研究

Application and research of collaborative filtering algorithm in personalized recommendation of college libraries

由于目前高校图书馆的个性化推荐效果不佳,研究协同过滤算法在高校图书馆个性化推荐中的应用。通过用户信息数据获取与显示,补全修正不完整数据。基于协同过滤处理数据,将隐式反馈转化为对图书资源类别的显式评分。用户与图书相似度定义及描述,利用不同关键词信息构成向量空间,计算相似度。建立高校图书馆个性化推荐模型,整合各模块功能实现个性化推荐。在实验论证分析中,文章方法在纵向推荐和横线推荐中均具有较高的查准率,满足高校图书馆对个性化推荐的应用要求,文章算法性能与传统算法性能对比,传统算法的平均绝对误差均高于文章算法,说明文章算法具有有效性。

The personalized recommendation effect of university library is not good, this paper studies the application of collaborative filtering algorithm in personalized recommendation of university library. By obtaining user information data, complete and correct incomplete data. Based on collaborative filtering algorithm to process data, translate implicit feedback into explicit ratings of book resource categories. Definition and description of similarity between users and books, use different keyword information to construct vector space, computing similarity. Establish personalized recommendation model of university library, integrate the functions of each module to achieve personalized recommendation. In the analysis of the experimental demonstration. The method presented in this paper has high accuracy in both vertical and horizontal recommendation, can meet the application requirements of personalized recommendation in university library, comparing the performance of the proposed algorithm with that of traditional algorithms, the average absolute error of the traditional algorithm is higher than that of the proposed algorithm, it proves that the proposed algorithm is effective.