海量文本信息的价值评估模型及算法*--以新浪微博为例

摘要

[目的/意义]互联网平台每天都在产生海量的文本信息，唯有快速评估这些信息的价值，才能及时发现高价值信息并避免信息迷失。[方法/过程]以新浪微博为例，探讨了海量文本信息的价值评估模型及算法。首先，利用社会网络分析法筛选并构建了信息价值评估指标体系；其次，对各级指标进行了可通过计算机快速执行的量化表达；最后，运用构建的信息价值评估模型，对取自新浪微博的文本信息样本进行了价值测算。[结果/结论]对所构建的模型及算法，可以快速计量海量信息的价值，并迅速完成对海量信息的价值分类排序，有助于及时发现和有效利用高价值信息。%Purpose/Significance] The only way to discover the high value information and avoid the information loss of the vast amount of text message produced in the internet platforms every day is to evaluate the value of these information quickly. [ Method/Process] An evaluation algorithm model for the mass information was developed based on Sina microblog. Firstly, an evaluation index system for the mass information was established based on Social Network Analysis. Secondly, the indicators in the model were quantified for fast calcula-tion by computers. Finally, the model was verified with a sample taken from Sina microblog. [ Result/Conclusion] The algorithm model facilitates the fast measurement of the information value, quick sorting of the mass information according to their relative value, and helps to identify and utilize the high value information timely.