An Efficient DS-LSTMM Ontology for Paddy - Agricultural Data Processing

Mining of unstructured and textual data is challenging task due to the complexity of data. Even unstructured data fails to provide clear information about data required. In the field of agriculture, data availability is vast in range which increases the complexity. To overcome those challenges, ontology concept is derived for data mining to process data. This paper focused on gathering relevant information about Paddy data for ease of processing. To extract data this paper developed a Domain-Specific Long-Short Term Memory Model (DS-LSTMM) is proposed for agricultural data processing. The proposed DS-LSTMM uses deep learning technique for data extraction. Initially, the proposed DS-LSTMM uses the construction of domain-specific data in hierarchical order. Based on the domain-specific order, agricultural data are categorized. Finally, using deep learning technique paddy data is extracted from constructed hierarchical structure. The performance of proposed DS-LSTMM is comparatively examined with search using keyword-based and Ontology-based search. The results imply that propose DS-LSTMM exhibits improved performance.