Research Cell: An International Journal of Engineering Sciences

Sanjay Bajpai ¹, Kanak Saxena ²

Affiliations
1 Department of Computer Applications, Lakshmi Narain College of Technology, Bhopal (MP), IN
2 Department of Computer Applications, Samrat Ashok Technological Institute, Vidisha (M.P.), IN

Abstract

Information hiding in images has given a new dimension to provide data security during transmission. In this paper, we present an algorithm by which embedding capacity of the image is extended to a large extent. Each character of the secret message is segmented into three parts in the forms of bits to make it more secure and each pixel of the colored image is compartmentalized into three components, that is, red, green and blue to enhance the embedding capacity. Various permutations and combinations are done on the segmented parts of the message and components of the pixel to make it more robust and efficient. Our algorithm imposes no constraints regarding the type of pixel chosen for embedding or the particular region of the image. This enhanced algorithm is supported by our experimental results that show that a secret message of sufficient long length can be embedded in the colored cover image causing least distortions.

Keywords

Segmentation, Embedding Capacity, Distortion, Robustness.

Full Text

   

J. S. Raikwal ¹, Kanak Saxena ²

Affiliations
1 Department of Information Technology, I.E.T, D.A.V.V, Indore, IN
2 Department of Computer Application, S.A.T.I., Vidisha, IN

Abstract

Web mining is the field having applications in variety of domains like e-commerce, e-learning, digital libraries etc. As the popularity of website related to medical and health car continues to grow, so web mining also derived its applications in medical and health care domain. This paper provides a survey of web mining applications in medical and health care domain and focuses on some research work in this domain.

Full Text

   

Umesh Banodha ¹, Kanak Saxena ¹

Affiliations
1 Samrat Ashok Technological Institute, Vidisha, IN

Abstract

The architecture style helps to determine the relationship of components and connectors that can be used with a set of constraints at any instance of architectural descriptions. With increase in size of software systems the software Architecture style presents a new set of design problems for systems to serve the needs of specific domains with integration of components. It has been recognized both by Academia and software industries as the most promising approach to tackle the problems in the era of Finance, Communications, Medical, Run-time applications etc. Different style has different system characteristics. In this paper we propose to model the architecture of medical process Re-engineering based on Pipe and Filter Architectural style. Medical process re-engineering model provides clinical staff, patients, and other individuals with knowledge and person - specific information, intelligently filtered and presented at appropriate times to enhance health and health care with the concept of using software re-engineering, which helped to improve understanding about the processes and led to the conclusions with deletion of subsequent improvements to those processes. We discuss the properties that are interesting to be analyzed with the process of design and reengineering.

Keywords

Architecture Style, MPR Model, Reengineering, Reusability, Software Architecture.

Full Text

   

Vijay Kumar Verma ¹, Kanak Saxena ²

Affiliations
1 Dept. of Computer Science & Engg., Lord Krishna College of Technology, Indore, IN
2 Department of Computer Applications, Samrat Ashok Technological Institute, Vidisha (M.P.), IN

Abstract

Fining frequent item set is a key issue in data mining; the Apriori algorithms use candidate itemsets to generate Frequent item set , but this approach is highly time-consuming because of self joining and prunining . To look for an algorithm that can avoid the generating of vast volume of candidate itemsets, DSIM (Data-Set Intersection Method) algorithm uses set intersection method to find the maximal frequent itemset.This process is performed by deleting items in infrequent 1-itemset and merging duplicate transaction repeatedly; the process is performed by generating intersections of transactions and deleting unneeded subsets recursively. This algorithm differs from all other methods which are used for discovering maximal frequent itemset.

Keywords

Data Mining, Maximum Frequent Itemsets, Candidate Itemsets, Intersection.

Full Text

   

Mohd. Shahnawaz ¹, Kanak Saxena ¹

Affiliations
1 Department of Computer Applications, Samrat Ashok Technological Institute, Vidisha, M.P, IN

Abstract

In this paper we compare various statistical method of estimation of missing data values. Missing data estimation is a part of data farming. Data Farming is a process to grow the data & provides a more comprehensive understanding of the possible outcomes, and offers the opportunity to discover outliers, surprises. Many times data mining task use existing data collected for various other purposes, such as daily transactional data, monitoring & control data. Sometimes, the data set might be missing some values, to estimate these missing values various statistical methods exist in the literature. In this paper a comparison among these methods is given with implementation & comparative results on the real life data set. This research work will be helpful to understand the effect of missing values on the mining process.

Keywords

Data Farming, Missing Data, Error Factor, Least Square Method.

Full Text