Development of Intelligent Prediction System using Data Mining Techniques
Keywords:
Introduction, Data Mining Tools and Techniques, Experimental Dataset Description, Feature Selection, Issues Regarding Feature Selection, Concern Regarding Cluster Analysis, Clustering Algorithm, Classifier Validation Methods, Statistical Analysis of Classifier, Summary, Feature Selection in Static Environment, Single Feature Subset Selection, Generation of Reduct Constructing Directed Minimal Spanning Tree using Rough Set Theory (GRG), Multiple Feature Subset Selection, Feature Selection in Dynamic Environment, Incremental Feature Subset Selection, Dynamic Reduct Generation using Rough Set Theory (DRED ), Comparative Analysis of DRED and IFS Method, Classification Analysis, Application of Data Mining Techniques, Designing of a Predictive Model in the Field of Agriculture, Rice Diseases, Conclusions and Future Research, BibliographySynopsis
Now a day’s, everything is being done through electronic media which generates huge amount of data in every moment. Most of the time, data are not static rather they are dynamic and transactional in nature. Retrieval of some interesting information from generated data is a very challenging task. Generally, each data set contains a large number of instances with huge number of features. Therefore, relevant feature selection and classification is one of the main objectives of data mining technique for knowledge discovery both in static and dynamic environment. Though many research works have been conducted for the data analysis of static and incremental data, still it is an ongoing research to handle with newly generated high dimensional data sets to obtain meaningful interpretations. The concerned issues are major requirements and challenges have been addressed in the book by developing optimal feature selection and classification algorithms using the concept of Rough Set Theory, Graph Theory, Genetic Algorithm, Particle Swarm Optimization, and so on. Developed algorithms have been applied in various benchmark datasets and in the real world agricultural field to classify and predict efficiently the unknown objects as a task for data analysis. All the methods are very useful with respect to Big Data analytics as it provides the optimal solutions.
This book is primarily intended to serve as a reference book for graduate and master degree students of computer science domain and researchers of any domain of various Colleges and Universities. We hope this book will provide the necessary guidance to the students to work on this data analysis domain. The book is actually an outgrowth of our research experience for the last several years.
