摘要
图书馆知识过载现象愈发严重,无法满足用户知识结构多元化需求,为此提出了改进支持向量机的图书馆书籍自动推荐方法。该方法从用户借阅历史、登录信息等获取书籍信息,并利用书籍信息和用户对书籍评分建立样本集,以该样本集作为基础建立用户兴趣模型,利用该模型获取书籍特征词库和书籍索引表,利用中文分词工具包提取书籍特征词并建立书籍特征索引库,计算书籍特征权值后,将书籍特征权值作为支持向量机的输入,利用自适应粒子群算法改进支持向量机参数后,经过训练支持向量机书籍分类推荐模型,输出最终书籍推荐列表,实现图书馆书籍自动推荐。实验结果表明:该方法可有效依据用户兴趣为用户推荐书籍,且书籍评分均高于9.80,所推荐的书籍被用户采纳率较高。
The phenomenon of library knowledge overload is becoming more and more serious,which can not meet the diversified needs of users'knowledge structure.Therefore,this paper puts forward a library book automatic recommendation method based on improved support vector machine.This method obtains the book information from the user's borrowing history and login information,uses the book information and the user's score on the book to establish a sample set,takes the sample set as the basis to establish the user interest model,uses the model to obtain the book feature thesaurus and Book index ta-ble,uses the Chinese word segmentation toolkit to extract the book feature words and establish the book feature index data-base,After calculating the book feature weight,the book feature weight is used as the input of support vector machine.After using the adaptive particle swarm optimization algorithm to improve the parameters of support vector machine,after training the support vector machine book classification recommendation model,the final book recommendation list is output to realize the automatic recommendation of library books.Experimental results show that this method can effectively recommend books for users based on user interests,and the book scores are all higher than 9.80,and the recommended books are highly adopted by users.
作者
贺海侠
HE Haixia(BeijingNormalUniversity At Zhuhai,Zhuhai,Guangzhou 519080,Chian)
出处
《自动化与仪器仪表》
2022年第1期144-147,152,共5页
Automation & Instrumentation
基金
2018年广东图书馆学会基金项目:项目编号:GDTK1828。
关键词
支持向量机
图书馆
书籍自动推荐
书籍特征权值
推荐列表
support vector machine
library
Automatic book recommendation
Book feature weight
Recommended list
