当前位置: 首页 > 文章 > Obsolescence determination of network information based on “double-proportion” method
数据与情报科学学报(英文)
2013,6
(4)
Position: Home > Articles > Obsolescence determination of network information based on “double-proportion” method
Journal of Data and Information Science
2013,6
(4)
Obsolescence determination of network information based on “double-proportion” method
作 者:
Juncheng Wang;Kai Yang;Xiang Li;Zhu Mengxia
关键词:
dpm;obsolescence;news;comments;webpages;determinatio
摘 要:
Purpose: It is necessary to determine whether the value of network information exists or not from users' perspective. This study is to develop a new approach, called “double-proportion” method (DPM), to determine the point of information obsolescence. Design/methodology/approach: Firstly, DPM is produced and introduced with statistics in great detail. Secondly, two data sets from Qiangguo Forum and NetEase News are used to validate DPM. Finally, the distributions of lifetime of the two sets are compared and reasons for the differences are revealed. Findings: In Qiangguo Forum , posts with no comments within 17 h can be identified as obsolescence, which indicates that approximately 83% of posts information has died within 24 h, 92% died within 36 h and almost all died within 54 h; While in NetEase News , news without comments within 28 h can be identified as obsolescence. Research limitations: This study adopted comments as the only behavior indicator, for further study other indicators like page views, republished times need to be used, to exploit better solutions for information obsolescence. Practical implications: According to obsolescence determination, network managers can optimize webpages more rationally, delete aged information in time, preserve valuable information, help search engines ignore aged webpages, optimize reptiles' mechanisms, and help advertisers choose webpages rationally to advertise online. Originality/value: Firstly, DPM is designed from users' perspective to describe the lifetime of information; Secondly, DPM is easy to use. DPM merges and improves former methods by taking the long-tail theory and Pareto principle into consideration.