A data-stream-based abnormal data mining in web texts environment

摘要

The stability and self-adaption for combination texts must be processed in Web Texts Environment. Therefore a language and technology method for self-adapting environment of web texts is needed. To do this, we have built an adaptive data-stream method in which the abnormal data mining process is started. The resource consumption of abnormal data in a text includes the resource consumption of error text and the total resource consumptions of relating with the previously executed texts which are dependent on the error text. In this paper an adaptive data-stream method is applied to implement the Abnormal Data Mining in Web Texts Environment. Proved by simulation verification, we proposed this adaptive data-stream method is efficient for solving the problem of abnormal data mining in web texts environment.