The Semantic Construction Of CADAL Chinese Literature Chronicle System

CADAL Chinese Literature Chronicle System aims to build a research system based on the rich historical and literary resources in CADAL. More than just digitizing these resources, we organize and systemize various types of digital resources in cross-media forms, and provide multi-dimensional digital visualization services on axis of characters, time, place, activity, etc., so that users can retrieve knowledge more precisely and completely. To maximize the value of the historical and literary resources, we work on the semantic construction of CADAL Chinese Literature Chronicle System. For the particularity of literary resources, we do text mining on literary works, so as to provide services on the dimension of semantic. We perform LDA topic modeling and text clustering on ancient literary works, automatically create tag cloud for each extracted topic, provide a method for generating stop-words for LDA. By optimizing the storage of geospatial data and appling ontology-based semantic web technology, we put explicit semantic to geographic entities, improving the knowledge organization on geographic dimension. We store maps data on PostGIS, use automated program to publish maps through GeoServer, and write SLD file for each map layer. We use ontology to model geographic data, and provide a fuzzy matching approach based on geospatial sampling points, to solve the problems caused by the frequent changes of place names and administrative divisions across dynasties.