Knowledge Needs and Information Extraction: Towards an Artificial Consciousness

Chapter 15

Fields of Interest for Text Mining

In view of the definition of text mining which is heavily influenced by the opinion of users (be they experts in a domain or simply general public users), we have represented the practice of text mining in accordance with the technical uses and their context. We can distinguish two main axes for text mining. One, which is to be found fairly widely in the academic world but which, bit by bit, is seeping into the domains of research and development in large companies, is the discovery axis. This covers text-mining practices that are aimed not at accelerating access to information but rather to gathering information in order to uncover hitherto-unknown relations. The other axis is more mechanistic, as it aims to take advantage of the computational capacity of machines to deal with a vast quantity of data, so as to skim through the information quickly and summarize it. This axis is that of organization. We shall here enumerate sub-categories of the usage of these two axes, which benefit from the availability of large amounts of data, which may or may not arouse the interest of a large number of users.

15.1. The avenues in text mining

15.1.1. Organization

15.1.1.1. Decision support

15.1.1.1.1. Watchfulness

When a decision-maker wishes to take a decision, he must do so in full knowledge of the facts, and therefore must be well informed. The application of watchfulness or vigilance (also referred to as competitive intelligence) is intended ...