Design of agricultural ontology based on levy flight distributed optimization and Naive Bayes classifier

Ontology is the fundamental representation used to link the various concepts of each field. It is widely utilized in the agricultural sector for the classification and expansion of new data and understanding. Besides this, the agriculture ontology enables optimization of feature weights to exact text from the source. However, the conventional methods exploit the vector space model for the indication of context. Nevertheless, it is not an exact solution since it has some shortcomings like undefined dimensionality and insufficiency of semantic data. Hence we propose a novel Naive Bayes Classifier using Levy Flight distributed optimization algorithm-based agricultural ontology to optimize the text mining. The dimensionality of the text tagging can be reduced with the aid of Principal Component Analysis and the dimensionally reduced data can be mapped by a feature mapping method. Further, the Naive Bayes classifier is exploited to estimate the weight of the text features. Furthermore, the Levy Flight distributed optimization algorithm is exploited to optimize the feature weight. Hence our proposed method provides better-optimized feature weight to perform the agricultural ontology. Besides, our proposed method is compared with existing GA-DAM, RENT, GRO, and CCE methods and concluded that our work has a better performance than all other works.