Hitesh Gupta

Information Hiding using Least Significant Bit Steganography and Blowfish Algorithm

Security is the most challenging aspects in the World Wide Web. In present time information sharing and transfer has increased exponentially. So to find out best solution this is providing necessary protection of our data against malicious attacks from intruders. Cryptography and Steganography are the two major techniques for secret communication. Cryptography converts information from its original form (plaintext) into unreadable form (cipher text); where as in steganography the secret message is hidden into the cover medium. There are many different techniques are available for cryptography and stagnography. In this paper two techniques BLOWFISH algorithm for cryptography and LSB approach for stagnography are used. First encryption of data is done by using BLOWFISH algorithm which is one of the most powerful techniques and then hide encrypted message using LSB approach. Our proposed model gives two layers of security for secret data.

SMDS: Secure Model for Cloud Data Storage

Cloud computing is a computing technique, where a large group of systems are connected to private or public networks, where data owner can store his data on remote systems and frees himself from storage burden and uses the data ondemand, anytime, everywhere. As, a Cloud data user does not possess direct control of his data, security is one of the few challenging issues which needs to be addressed. Security in Cloud computing can be addressed in many directions viz. authentication, integrity, confidentiality and many more. Data integrity or correctness is an issue where there may be some unauthorized alteration in the data without consent of the data owner. In this paper we address the issue of storage correctness in Cloud computing and propose operational algorithms which may be used to build a complete solution.

Global K-Means (GKM) Clustering Algorithm: A Survey

means clustering is a popular clustering algorithm but is having some problems as initial conditions and it will fuse in local minima. A method was proposed to overcome this problem known as Global K-Means clustering algorithm (GKM). This algorithm has excellent skill to reduce the computational load without significantly affecting the solution quality. We studied GKM and its variants and presents a survey with critical analysis. We also proposed a new concept of Faster Global K-means algorithms for Streamed Data sets (FGKM-SD). FGKM-SD improves the efficiency of clustering and will take low time & storage space. the randomly chosen sets. Every run should be initialized using the K final centroid locations from one of the run of 10 subsets. The K center location that we get from this run will be used to initialize the K-means algorithm for the complete data set. The Global K-Means algorithm (The GKM algorithm) (2) is the incremental approach of clustering. We can dynamically add one cluster center at a time using deterministic global search procedure from suitable initial positions. It is consists of N (Where N is the size of the dataset) executions of K- means algorithm. Experimental results of the algorithm show that GKM algorithm considerably out performs the K-means algorithms.

Mining Strong Valid Association Rule from Frequent Pattern and Infrequent Pattern Based on Min-Max Sinc Constraints

Rule mining is very efficient technique for find relation of correlated data. The correlation of data gives meaning full extraction process. For the mining of rule mining a variety of algorithm are used such as Apriori algorithm and tree based algorithm. Some algorithm is wonder performance but generate negative association rule and also suffered from multi-scan problem. In this paper we proposed IMLMS-PANR-GA association rule mining based on min-max algorithm and MLMS formula. In this method we used a multi-level multiple support of data table as 0 and 1. The divided process reduces the scanning time of database. The proposed algorithm is a combination of MLMS and min-max algorithm. Support length key is a vector value given by the transaction data set. The process of rule optimization we used min-max algorithm and for evaluate algorithm conducted the real world dataset such as heart disease data and some standard data used from UCI machine learning repository.

A Review: Data mining over Multi-Relations

In this paper, Multi-relational data mining enables pattern mining from multiple tables. Multi-relational data mining algorithms can be used as practical proposal to overcome the deficiency of conventional algorithms. Multi-relational data mining algorithms directly extract frequent patterns from different registers in efficient manner without need of transfer the data in a single table will, on the other hand, used the available memory space is not enough to ensure the production of large amounts of data. For this reason, and the use of space, algorithms are an integral care for the prospection of large repositories. The paper provides the overview of multi relation data mining techniques and classification algorithms. It also defines the frequent pattern mining. The presented paper discussed the various architecture and issues related to multi table data mining. A lot of literature has been proposed in this area. Some of them has discussed in this paper.

Classification of end users over distributed network by using fuzzy temporal correlation of HTTP and TCP log

Digital communication is a back bone of the communication over the world. This is openly available for all users. This is a major cause of attack in the network. Now how it is possible to catch the attacker. So here digital investigation takes place. This paper proposed an approach which is used the two basic concepts one is temporal data mining and another is fuzzy association rules. Using log files it is possible to classify the attacker from the normal user. This paper uses the time based investigation which gives the efficient result to detect the end user if it is attacker.

Digital Image Watermarking in Robust Feature Region Set

image watermarking is widely used for copyright protection of digital information. The effectiveness of a digital watermarking technique is indicated by the robustness of embedded watermarks against various attacks. A new method for watermarking is suggested is feature based watermarking. For getting robust watermark, the watermark should be embedded in silent part of the data and for these significant features of data is used. In this paper few methods of feature extraction as Harris Laplacian, Laplacian-of-Gaussian, Susan, Gilles are applied for feature extraction. Robust Non overlapping regions against different attacks are selected for watermarking. Comparison for robust feature selection is done against different feature extraction methods. In next stage those regions are pruned to get minimal primary feature region set using pruning algorithm and watermark is embedded in selected regions and then again results of extracted watermark is compared against different feature selection methods for robustness.