High-utility itemset mining is one of the highly researched area. Many research enthusiasts have discovered various techniques and algorithms to mine high-utility itemsets from transaction databases. One of the limitations of the existing high-utility itemset mining techniques is that there is no any generalized framework for applying the custom combinations of input parameters and any other constraints for mining high utility itemsets. This paper proposes a novel customizable framework to discover customized high utility itemsets (C-HUI). Users can customize the constraints and/or input parameters as per their requirements. A novel C-HUIM algorithm is used to discover customized high utility itemsets (C-HUI) from real-life datasets. The experimental results of the proposed framework and C-HUIM algorithm highlight the effectiveness of the approach.

A steganography is an art of hiding confidential data into digital media such as image, audio, video etc. Texture synthesis uses the concept of the patch which represents an image block of source texture where its size is user specified. A texture synthesis process resamples a smaller texture image and provides a new image with arbitrary size and shape. Instead of using an existing cover image to hide messages, the algorithm conceals the source texture image and embeds secret messages using the process of texture synthesis. This allows extracting the hidden messages and source texture from a stego synthetic texture. This offers the advantages like, First, it provides the embedding capacity that is proportional to the size of the stego texture image. Second, the reversible capability inherited from this includes functionality, which allows recovery of the source texture. And third, there will be no image distortion since the size of the new texture image is user specified.

Privacy is the key factor to handle personal and sensitive data, which in large chunks, is stored by database management systems (DBMS). It provides tools and mechanisms to access and analyze data within it. Privacy preservation converts original data into some unknown form, thus protecting personal and sensitive information. Different access control mechanisms such as discretionary access control, mandatory access control is used in DBMS. However, they hardly consider purpose and role-based access control in DBMS, which incorporates policy specification and enforcement. The role based access control (RBAC) regulates the access to resources based on the roles of individual users. Purpose based access control (PuBAC) regulates the access to resources based on purpose for which data can be accessed. It regulates execution of queries based on purpose. The PuRBAC system uses the policies of both, i.e. PuBAC and RBAC, to enforce within RDBMS.

In the current scenario of the business world, the importance of data analytics is quite large. It certainly benefits the businesses in the decision-making process. Sequential rule mining can be widely utilized to extract important data having variety of applications like e-commerce, stock market analysis, etc. Predictive data analytics using the sequential rule mining consists of analyzing input sequences and finding sequential rules that can help businesses in decision making. This article presents an approach called M_TRuleGrowth that generates partially-ordered sequential rules efficiently. The authors conducted an experimental evaluation on real world dataset that provides strong evidence that M_TRuleGrowth performs better in terms of execution time.