Automatic Generation And Applications Of Ontology

Information is on the exponential increase with the rapid development of Internet; however, there are many kinds of difficulties that appear with the increase of information. Since the differences in thinking, understanding, knowledge, circumstance, and ways result in differences of understanding and cognition on the same things, which brings on a series of concepts, structures and methods that are of differences, repeated definitions and disagreements; moreover, which leads to many harmful results. The use of ontology-based techniques provides a considerablly successful solution to make agreements and commonness of understanding and cognition. This thesis studies technology of automatic generation of ontology and its applications and makes achievements as follows: 1. Make a systematic and profound study of the basic theories of ontology including its definition, primitive of modeling, and classification, which is the emphases of study in this thesis, and bases of technology of automatic generation of ontology. 2. Make much improvement in the key techniques of automatic generation of ontology including terminology extracting, concept learning, instance learning and relation learning. 3. Implement the system of automatic generation of ontology, OntoAGS, and make comparison with Text-To-Onto that is a famous system in the world. The experiment indicates OntoAGS is highly competitive with Text-To-Onto. 4. Make a study of applications of ontology to two domains such as text classification and information retrieval. The results of experiments indicate the improvement is significant by applying ontology to text classification. Ontology-based information retrieval is an important approach to overcome many disadvantages...