A theoretical model of meaning on textual big data

Abstract

In recent years, the importance of the “meaning” that users find through their experiences of using products and services has been pointed to in the product development and marketing fields. Also, the development of a methodology to quantitatively evaluate this meaning has been advocated. The background to this is the increasing sophistication of technologies to collect and analyze vast amounts of text data, including users generated data, through the development of Internet media such as Web2.0 and CGM. Attempts are also being made to quantitatively verify users' evaluations of certain products or services by analyzing “Big Data,” which is the large-scale collection of text data output by users. Here, there remains room to investigate how meaning can be defined, as meaning for the user, which is regarded as subjective, has generally been treated quantitatively by statistical analysis. Therefore, in this research we examine a theoretical model of meaning that can be quantitatively extracted by analyzing text. First, we examine the argument of previous research that defines the meaning as “translation in a different language.” Then, we argue that in terms of the translation, the generation of meaning can be understood in a symbols process and a signals process. And we attempt to construct a theoretical model of meaning that is quantitatively detectable from textual big data.