Data Mining and Knowledge Engineering

D. N. Vasundhara ¹, M. Seetha ²

Affiliations
1 Jawaharlal Nehru Technological University, Hyderabad, Telangana, IN
2 Department of Computer Science Engineering, G Narayanamma Institute of Technology and Science, Hyderabad, Telangana, IN

Abstract

Image classification is one of classical problem for many aspects of remote sensing in which to extract some of the important spatially variable parameters, global change studies and environmental applications. In the literature, various classification methods have been developed for classification of images such as Artificial Neural Networks (ANN), Support Vector Machines (SVM), Fuzzy Support Vector Machines (FSVM), Genetic algorithms (GA) and Decision Trees (DT). In this paper, we propose a combined scheme for spatial image classification, which is composed of FSVM and ANN techniques. The proposed classification scheme consists of two main steps: Firstly, we separate the each image into class by an ANN classifier based on the features of images and in the second step, the FSVM classifier has been applied on the output of ANN.  This can be denoted as ANN_FSVM classifier. A comparison of techniques for spatial data has been given in this paper.

Keywords

Artificial Neural Network, Feature Extraction, Fuzzy Support Vector Machine, Image Classification.

Full Text

     

K. Prasanna ¹, M. Seetha ², M. Sankara Prasanna Kumar ³

Affiliations
1 Department of Computer Science and Engineering, Annamacharya Institute of Technology & Sciences, Rajampet, Andhra Pradesh, IN
2 Department of Computer Science and Engineering, G. Narayanamma Institute of Technology & Sciences for Woman, Hyderabad, Andhra Pradesh, IN
3 Department of Computer Science and Engineering, Annamacharya Institute of Technology & Sciences, Rajampet, IN

Abstract

Association rule mining aims at generating association rules between sets of items in a database. Now a day, due to huge accumulation in the database technology and incredible growth in high dimensional dataset, conventional data base methods are inadequate in extracting useful information. Such large high dimensional data gives rise to a number of new computational challenges not only the increased in number of data objects but also in the increased in number of features/attributes. However, it is becoming very tedious to generate association rules from high dimensional data, because it contains different dimensions or attributes in the large data bases. To improve the high dimensional data mining task, it must be preprocessed efficiently and accurately. In this paper, an Apriori based method for generating association rules from large high dimensional data is proposed. It constitutes 1) Preprocessing and generalizing the data base dimensions; 2) generating high dimensional strong association rules using support and confidence. It can be seen from experiments that the mining algorithm is elegant and efficient, which can obtain more rapid computing speed and sententious rules at the same time It was ascertained that the proposed method is proved to be better in support of generating association rules.

Keywords

Association Analysis, Apriori Algorithm, Pre Processing, High Dimensional Data, Support, Confidence, Data Mining.

Full Text