基于文献内容的图书推荐机制研究

Research on Book Recommendation Mechanism Based on Literature Content

每年发布的各种"好书榜单"已经大大超过了普通读者的阅读量,即图书在专家筛选后仍然存在信息过载的问题,需要进一步结合专家意见、群体智慧和个性化需求,构建更准确的基于文献内容的阅读推荐机制。文章引入了专家书目维、协同过滤维和情感分析维,通过"引入权重值—加权求和—计算比值"的算法步骤,分别得出专家推荐指数、协同推荐指数和情感推荐指数。文章认为该图书推荐机制既考虑了社会主流价值观的要求,也匹配了读者的个人阅读口味,能够把真正适合的好书推荐给读者。

The Books on the lists of the year has largely exceeded the reading quantity of ordinary readers,which means after the selections of experts the books still reflect the problem of information overloading,so we need to find a book recommendation mechanism based on literature content which combines with expert opinion, collective wisdom and personal needs. This article introduced the dimensions of expert recommendation, collaborative filtering and emotional analysis,and by calculation of weight values, weighted sum and ratio, got the three indexes of experts ＇recommendation, collaborative recommendation and emotion recommendation. This article holds that the book recommendation mechanism not only takes into account the requirements of the mainstream values of society,but also matches the reader＇ s personal reading taste. So it can recommend the best books to the readers.