Bidisha Roy

OBESTDD: Ontology Based Expert System for Thyroid Disease Diagnosis

This study presents a new approach in developing an ontology based expert system to diagnose thyroid diseases. Ontology represents explicit specification of knowledge in a specific domain of interest in the form of concepts and relations among them. Its merit is that the represented knowledge is both computer and human readable. The adoption of ontology in healthcare has facilitated domain experts and non-experts to perform knowledge representation tasks with great ease. This paper presents an ontology describing the domain of thyroid diseases and its symptoms. The thyroid gland is one of the most important organs in the body since it is responsible for controlling metabolism. Thyroid disease can be difficult to diagnose because symptoms are easily confused with other conditions. This study proposes at diagnosing thyroid diseases with an ontology based expert system called as OBESTDD (Ontology Based Expert System for Thyroid Disease Diagnosis). It uses ontology to model the contents of domain knowledge and rules to infer the related diagnosis.

Secure data transmission using video

In todays world transmitting data in a safe and secure fashion is difficult, especially when highly sensitive data is involved. The data should be ideally robust to security threats. The paper aims at proposing a methodology which employs dual level of security using cryptography and steganography to hide the secret text data using video as a cover. A blend of existing and novel techniques is used for hiding the data. Cryptography aims at concealing the secret data and steganography hides the existence of the data. Since a video is addressed as a collection of frames, a frame selection logic is incorporated which inserts data into the frames in a random fashion so that the data cannot be retrieved easily by the attacker. Considering the goal of crypto and steganography the analysis of the methodology can be done on the basis of visual perceptibility, error ratios and histogram comparison of video frames before and after hiding data. Use of two techniques with distinct goals to hide the data increases the robustness as well as simple frame selection logic prevents the attacker to obtain the data easily, but at the same time simplifies the senders job as a rule index of frames is maintained.

Suggested techniques for clustering and mining of data streams

The buzz word in research is Big Data. Big Data gets characterized by 5 Vs: Volume, Velocity, Variety, Veracity and Value of data. Volume in order of penta bytes, velocity which refers to click stream data in various domains, variety comprising of heterogeneous data, veracity indicating the cleanliness of data and value emphasizing on the return on investment for companies who invest in Big Data technologies. This Big Data is better modeled not as persistent tables but in the form of transient data streams which need different clustering and mining techniques to be effectively processed and managed. In this paper some suggestions on online learning through clustering and mining of stream data are presented.

Enhanced data processing using positive negative association mining on AJAX data

Knowledge discovery is the process of analyzing data from different perspectives and summarizing it into useful information. Association rule mining is a data mining process used widely in traditional databases to find the positive association rules. Association rules are created by analyzing data for frequent patterns and by using the criteria support and confidence to identify the most important relationships. However, there are some other challenging rule mining topics like negative association rule mining. In this research, a rule mining approach has been proposed that provides efficient and secure solution using positive and negative association rule computation on Asynchronous JavaScript and XML (AJAX) data. By using AJAX, we get the search result in the form of semantic data. Whenever data search from database is intended, the next possible word of the search will be made available.