Since past few years there is tremendous advancement in electronic commerce technology, and the use of credit cards has dramatically increased. As credit card becomes the most popular mode of payment for both online as well as regular purchase, cases of fraud associated with it are also rising. In this paper we present the necessary theory to detect fraud in credit card transaction processing using a Hidden Markov Model (HMM). An HMM is initially trained with the normal behavior of a cardholder. If an incoming credit card transaction is not accepted by the trained HMM with sufficiently high probability, it is considered to be fraudulent. At the same time, we try to ensure that genuine transactions are not rejected by using an enhancement to it(Hybrid model).In further sections we compare different methods for fraud detection and prove that why HMM is more preferred method than other methods.Keywords-Credit card , fraud , Hidden Markov Model ,Hybrid model 978-1-4673-0126-8/11/$26.00 c 2011 IEEE

With the ever-increasing usage of internet, the availability of digital data is in tremendous demand. In this context, it is essential to protect the ownership of the data and to be able to find the guilty user. In this paper, a fingerprinting scheme is proposed to provide protection for Numeric Relational Database (RDB), which focuses on challenges like: 1. Minimum distortion in Numeric database, 2. Usability preservation, 3. Nonviolation of the requirement of blind decoding. When the digital data in concern is numeric in nature the usability of data needs to be keenly preserved, this is made possible by achieving minimum distortion.

Deduplication is the process of determining all categories of information within a data set that signify the same real life / world entity. The data gathered from various resources may have data high quality issues in it. The concept to identify duplicates by using windowing and blocking strategy. The objective is to achieve better precision, good efficiency and also to reduce the false positive rate all are in accordance with the estimated similarities of records. Various Similarity metrics are commonly used to recognize the similar field entries. So the main focus of this paper is to applying appropriate similarity measure on appropriate data to properly identifying the duplicates.

No abstract

Information age demands omnipresence of data. Large data sets are created, maintained and outsourced to the third party experts for data mining. Knowledge and patterns are extracted by using advanced data mining algorithms that assist the decision makers to ensure quick, correct and effective decisions to be made in this world of competition. The outsourcing of these large data sets faces the problem of theft and loss of ownership. The problem of data theft can be handled by fingerprinting i.e. embedding buyer specific marks along with ownership identification marks which further leads to the challenge of knowledge preservation. Thus, a technique which performs fingerprinting with knowledge preservation on numeric relational data to be outsourced is proposed here. It is ensured that the usability constraints are not violated. Knowledge preservation is achieved by optimizing the error to be inserted using Particle Swarm Optimization (PSO), a natureinspired optimization algorithm. Collusion attack is very well-known in the context of fingerprinting techniques. Here, the proposed system provides a mechanism for avoiding collusion. The proposed system is independent of the primary key.