Ontology Matching and Extraction
Loading...
Date
item.page.authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Civilization started with agriculture, and it remains vital even today. There is a vast amount of data related to agriculture sector is available online in the form of documents, tables, and spreadsheets. In order to use this data for the benefit of farmers it is necessary to convert data into structured knowledge. Ontology can be use to represent knowledge by defining relationships between terms and rules. They are effective in tasks like information retrieval, extraction, and question answering. In this thesis, we propose an ontology-based question-answering system for the agriculture domain. We address the challenges of developing an ontology for a vernacular language and develop a system for agriculture domain. Gujarat is a leading producer of groundnut crops. However, the region lacks a precise knowledge representation or ontology for groundnut farming. To address this, we compared existing groundnut ontologies to evaluate their suitability for Gujarat farmers. This thesis includes detailed research and analysis of existing groundnut ontologies, focusing on overlapping concepts, differences in structure or concepts, unique concepts, and missing concepts. Then we discuss the critical thinking of existing groundnut ontologies to get some answers and manually checked how Gujarati (regional language) concepts fit into them. After analyzing the compatibility of these ontologies with Gujarati agricultural data, we found that they could be adapted with modifications. Using this analysis, we manually constructed a groundnut ontology in the Protégé tool with the necessary updates. This thesis also introduces an ontology-based interactive interface designed to provide farmers in Gujarat with information related to groundnut crops. The proposed QA interface converts natural language question into SPARQL Query and provides answer using the backbone ontology. The overall performance of the system is at par with the existing semantic QA system. We consider gujarati as a language and groundnut as a crop to perform our exp