Application of 5D world map system to large news-article database for realizing context-diversity-responsive semantic associative search

This paper presents an application of 5D World Map System with Context-Diversity-Responsive Semantic Associative Search for a large database of environmental news articles. By applying 5D World Map System to data describing social and natural phenomena, a new news-article subscription environment is realized by the dynamic combination of various semantic, temporal and spatial 'context'. The new subscription environment enables to create, accumulate and visualize a series of analyzed results of semantic relations among phenomena as an interpreted social story. The system realizes the extraction of correlative and causal relations between phenomena which are potentially included in news articles. A semantic associative search method is applied to this system for realizing the concept that 'semantics' of words, documents, and events/phenomena vary according to the 'context'. The main feature of this system is to create various context-dependent patterns of social stories according to user's viewpoints and the diversity of context in phenomena dynamically. This system also provides an environment for analyzing the time-series change and spatial expansion of social and natural phenomena on a time-series multi-geographical space. In this paper, we show a prototype system applied for vast ten years of news-articles and several experiments about 'global warming' to clarify the feasibility of the system.