| Resumen |
In order to compute intelligent answers to complex questions, using the vast amounts of information existing in the Web, computers have (1) to translate such knowledge, typically from text documents, into a data structure suitable for automatic exploitation; (2) to accumulate enough knowledge about a certain topic or area by integrating or fusing these data structures, taking into account new information, additional details, better precision, synonyms, homonyms, redundancies, apparent contradictions and inconsistencies found in the incoming data structures to be added; and (3) to perform deductions from that amassed body of knowledge, most likely through a general query processor.
This article seeks to solve point (2) by using a method (OM, Ontology Merging), with its algorithm and implementation, to fuse two ontologies (coming from Web documents) without human intervention, producing a third ontology, taking into account the inconsistencies, contradictions and redundancies between them, thus delivering an answer close to reality. Results of OM working on ontologies extracted from Web documents are shown. |
