No abstract

In this paper a classification for natural images is proposed using hybrid features. The objective of this paper is to develop an image content based classifier, which can perform identity check of a natural image. Here we have extracted wavelet and color features from a captured natural image to classify out of three groups. The developed technique is able to classify translation and rotation invariant matching among natural images using feed forward back propagation neural network. The database contains several hundreds of natural images of three groups namely coast, Forest, Mountain for classification and found good classification rate.

No abstract

In privacy preserving data mining, utility mining plays an important role. In privacy preserving utility mining, some sensitive itemsets are hidden from the database according to certain privacy policies. Hiding sensitive itemsets from the adversaries is becoming an important issue nowadays. The existing paper utilized two algorithms; such as HHUIF and MSICF are conceal the sensitive itemsets, so that the adversaries cannot mine them from the modified database. But, the performance of this method lacks if the utility value of the items are same. To solve this problem, in this paper a Modified MSICF algorithm (MMSICF) is proposed. The proposed MMSICF algorithm is a modified version of existing MSICF algorithm. The MMSICF algorithm computes the sensitive itemsets by utilizing the user defined utility threshold value. The threshold value selection plays a major role in this paper and it is determined by the hybridization of Artificial Bee Colony (ABC) and Genetic Algorithm (GA). In order to hide the sensitive itemsets, the frequency value of the items is changed. The proposed MMSICF reduces the computation complexity as well as improves the hiding performance of the itemsets. The algorithm is implemented and the resultant itemsets are compared against the itemsets that are obtained from the conventional privacy preserving utility mining algorithms.

No abstract