A. Meenakshi ¹, R. Aghila ², P. Suganthi ², S. Kavya ²

Affiliations
1 Department of Information Technology, K.L.N. College of Information Technology, Sivagangai, Madurai – 630612, Tamil Nadu, IN
2 Department of Computer Science and Engineering, K.L.N. College of Information Technology, Sivagangai, Madurai – 630612, Tamil Nadu

Abstract

Background: Knowledge Engineering is an engineering discipline that involves integrating knowledge into computer systems in order to solve complex problems normally requiring a high level of human expertise. Knowledge Engineering is the technique applied by knowledge engineers to build intelligent systems: Expert Systems, Knowledge Based Systems, Knowledge based Decision Support Systems, Expert Database Systems etc. Methods: This work builds a Knowledge Base using Knowledge Base Markup Language (KBML) which is derived from XML architecture. All the Meta information is stored in a KBML file whereas the actual data may be available in any data source. Findings: This system also provides facilities to search/add the contents to and from the Knowledge Base dynamically. The experimental results show that the system provides a high precision, recall and f-measure values which proves the high relevance of the retrieved values. Applications: Builds an intelligent system for Edaphology domain which concerns with plants and its related soil features. It provides the edaphologists and agriculturists by retrieving relevant and useful information about the plants. This is of huge importance as plant growth and yield are directly dependent on the soil features.

Keywords

Data Source, Edaphology, Knowledge Base Markup Language, Meta Information

Full Text

   

A. Meenakshi ¹, P. Suganthi ², R. Aghila ², S. Nirmala ²

Affiliations
1 Department of Information Technology, K.L.N. College of Information Technology, Madurai – 630612, Tamil Nadu, IN
2 Department of Computer Science and Engineering, K.L.N. College of Information Technology, Madurai – 630612, Tamil Nadu, IN

Abstract

Background: Data volume with respect to modern systems has been growing outwardly at a rapid rate. The systems also face a tough task of frequent re-scanning of the large datasets stored because of the updation process. Frequent re-scanning, updation and large corpus data leads to having larger retrieval time and decreased efficiency. Hence, the data retrieved by automated systems may not be efficient and accurate. Knowledge Based System using Dynamic decision Quad tree is employed in this paper. The Quad trees designed in this chapter helps in retrieving knowledge about the suitable plants for the given kind of soil. Methods: The Quad trees designed in this paper helps in retrieving knowledge about the suitable plants for the given kind of soil. The Dynamic decision Quad tree is built using the knowledge and information given by Edaphologists and domain experts. The system is made of two modules, namely Dynamic Quad Tree construction module and information retrieval module. In the first module, the obtained data is transformed to a dynamic Quad tree based knowledge data structure. In the information retrieval module, knowledge retrieval is carried out with the help of constructed knowledge base (XML). Findings: Efficient techniques have been developed and tailored for solving complex soil datasets using data mining. The proposed scheme exploits clustering and dynamic decision tree based system for better storage in edaphology compared to existing systems. Dynamic decision quad trees are tailored to make best retrieval in soil databases. Applications: Assist Edaphologists and agricultural experts in obtaining the right crops/plants for the given soil characteristics.

Keywords

Data mining, Edaphology, Information Retrieval, Knowledge Extraction

Full Text