A Study on the method of quantifying the diffusion situation of the word-of-mouth communication

Abstract

With the popularization of social networks,word-of-mouth communication on the internet is greatly impacting companies and their products . Companies are required to quickly and appropriately respond by analyzing the cause of the expansion of the reviews. However, constantly listening to the content of the review is a correspondingly laborious task.Here we describe a method to automatically detect the situation where the word of mouth spreads, based on the time series change of the review data quantity of a specific keyword.
In this method, not only detection of abnormal values ??but also quantification of diffusion magnitude and diffusion period of reviews can be made possible. In this paper, we explain the features of the developed method.