Implementación de un sistema de detección de señales débiles de futuro mediante técnicas de minería de textos

Nowadays, one of the biggest threats for companies is not being able to cope with the constant changes occurring in the market by not predicting them well in advance. For this reason, the development of new processes that facilitate the detection of future phenomena and significant changes is a key component for correct decision making that can mark a correct course in the company. A business intelligence based architecture system is proposed to allow discrete changes or weak signals detection in the present that are indicative of more significant phenomena and transcendental changes in the future. In contrast with current available works, which are focused on structured information sources or, at most, with only a single type of data source, in this paper the detection of these signals is done quantitatively from various kinds of heterogeneous and unstructured documents (scientific articles, journalistic articles and social networks) on which text mining techniques are applied. The system has been tested in the study of the future of solar panels sector, obtaining promising results that can help business experts in the recognition of new driving factors of their markets and the development of new opportunities.