基于读者阅读偏好的数字资源智能聚类系统设计研究  

Design of digital resource intelligent clustering system based on readers′reading preference

作　　者：王玉兰 WANG Yulan(Xi’an Shiyou University,Xi’an 710065,China)

机构地区：[1]西安石油大学,陕西西安710065

出　　处：《现代电子技术》2021年第8期154-158,共5页Modern Electronics Technique

基　　金：陕西省自然科学基金资助项目(2018JQ6093)。

摘　　要：传统数字资源智能聚类系统设置的权值不具有独立性,导致系统只能处理体量较小的数字资源且不具备可扩展性,因此基于读者阅读偏好,设计全新的数字资源智能聚类系统。硬件方面,在微控制器外围重新设计晶振电路。在软件方面,根据读者阅读偏好相似性度量结果,设计数字资源预处理模块,通过设置独立性更好的TF⁃IDF权值,实现系统对数字资源的智能聚类。实验测试表明,与传统设计下的两组系统相比,此次设计的系统扩展性更好,面对大、中、小体量的数字资源,都有极佳的聚类效果。As the weight value set by the traditional intelligent clustering system of digital resources is not independent,which results in the system processing smaller digital resources only and has no scalability,a new digital resource intelligent clustering system is designed based on the readers′reading preference.In the aspect of hardware,the crystal oscillator circuit is redesigned outside the microcontroller.In the aspect of software,the digital resources preprocessing module is designed according to the measurement results of similarity of readers′reading preferences,and the intelligent clustering of digital resources is realized by setting TF⁃IDF weights with better independence.The experimental test results show that in comparison with the two groups of systems under the traditional design,the designed system has better scalability,and has excellent clustering effect in the face of digital resources with large,medium and small volume.

关 键 词：数字资源 智能聚类 读者偏好 晶振电路 权值设置 系统设计 

分 类 号：TN919-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]