N.Ch.Sriman Narayana Iyengar

Privacy has in recent times become an astounding akin to an oxymoron. It can either be embellished or marred with technology; confiscating more consideration in many data mining applications. We are focusing on information safety measures in order to preserve the individual's privacy, so that no personal information can be gained by the hacker from the data. Under the modern

There is a need for a CAPTCHA as a result of the abuse of automated 'bots' [6]. The problem of spamming resulted as the bots intruded into the internet [3]. However, there were many solutions provided for stopping these bots from entering the internet. The best and most feasible solution proposed till now has been the CAPTCHA or the fuzzy

Distributed Denial of Service (DDoS) is a type of attack in the application layer initiated from the various hosts to a single web server. The aim of this attack is to consume all the resources of the targeted system by exploiting the vulnerability. We proposed a mathematical model called Recurrence Quantification Analysis (RQA) for detecting the DDoS attacks by computing entropy and determinism of selected packet attributes. To detect the anomalies and check the performance we considered the live traffic traces from the network and various RQA parameters like entropy, laminarity and determinism were used to determine the uncertainty or randomness in the dataset.

Nowadays, health care is one of the most important subjects in our life. Increasing level of patients demand across the world obliges health care services in a more flexible and uniform manner. So that patients can get useful health services in the home environment without going to the health center personally. The e-healthcare is new kinds of health care system

Medical diagnosis processes vary in the degree to which they attempt to deal with different complicating aspects of diagnosis such as relative importance of symptoms, varied symptom pattern and the relation between diseases themselves. Rough set approach has two major advantages over the other methods. First, it can handle different types of data such as categorical, numerical etc. Secondly, it does not make any assumption like probability distribution function in stochastic modeling or membership grade function in fuzzy set theory. It involves pattern recognition through logical computational rules rather than approximating them through smooth mathematical functional forms. In this paper we use rough set theory as a data mining tool to derive useful patterns and rules for kidney cancer faulty diagnosis. In particular, the historical data of twenty five research hospitals and medical college is used for validation and the results show the practical viability of the proposed approach.

Chronic Kidney Disease (CKD) is usually characterized by a gradual loss of the functioning which the kidney does over time due to various factors. Early prediction and treatment save the kidney and halts the progress of CKD. CKD disease is being viewed as global public health issue for the past decade. The greatest threat for this deadly disease is developing countries where getting therapy is very expensive. The importance of predicting individuals who are at risk of CKD as well as applying clustering techniques cannot be underestimated since these can modify the progression of the disease. Identifying the silent killer disease early offers best opportunities for implementing possible strategies for lessening the probability of kidney loss. Neuro-fuzzy algorithm is applied to determine the risk of CKD in patients. Predictions done using neuro-fuzzy gave an accuracy of 97 percent. Using selected features, prediction for CKD disease is done so as to identify the risk. The results of the prediction are clustered to identify the percentage of patients with a high risk of having kidney disease who have a higher probability of being diabetic. Using hierarchical clustering three clusters formed show that there is a strong relationship between chronic kidney and diabetes.

An intelligent predictive model using deep learning is proposed to predict the patient risk factor and severity of diabetics using conditional data set. The model involves deep learning in the form of a deep neural network which helps to apply predictive analytics on the diabetes data set to obtain optimal results. The existing predictive models is used to predict the severity and the risk factor of the diabetics based on the data which is processed. In our case Firstly, a feature selection algorithm is run for the selection process. Secondly, the deep learning model has a deep neural network which employs a Restricted Boltzmann Machine (RBM) as a basic unit to analyse the data by assigning weights to the each branch of the neural network. This deep neural network, coded on python, will help to obtain numeric results on the severity and the risk factor of the diabetics in the data set. At the end, a comparative study is done between the implementation of this model on type 1 diabetes mellitus, Pima Indians diabetes and the Rough set theory model. The results add value to additional reports because the number of studies done on diabetes using a deep learning model is few to none. This will help to predict diabetes with much more precision as shown by the results obtained.

Healthcare Organizations have seen an alarming rise in cyber-attacks in the recent years. One way a hacker could get control was by breaking into a medical network to gain access over the active medical devices that patients rely on for their survival. Our network model proposes a low-interaction and a medium-interaction honeypot based intrusion detection system using Dionaea and Kippo SSH to secure our internal network and study the activities of the intruders. We also look at a possible Metasploit attack and Brute force attack logged by Dionaea and Kippo SSH which prepares the Malware Analysis report of the suspicious file downloaded.

Lower back pain can occur due to various reasons involving any body part such as the interconnected network of spinal cord, nerves, bones, discs or tendons in the lumbar spine. Understanding and knowing about the origins of this disorder and getting a diagnostic treatment that determines the underlying reason, is the primary step in achieving an effective and efficient cure. Despite of heavy spending on resources such as time and money, devoted to lower back pain research methodologies, fruitful management remains a significant goal and lower back pain continues to be a cause of considerable concern on the primary care setting. One of the reason for this, could be the degrees of importance for researching into specific domain like these, are regularly developed by researchers and funding bodies, with less consideration of the needs of primary care practitioners. This study aims to determine the research priorities of primary care practitioners who manage low back pain on a day-today basis and to identify whether a person is abnormal or normal using collected physical spine data of 381 patients with 12 parameters. This study also identifies the degree of importance of each parameter used in the classification and ranks those parameters accordingly.

As part of our research work Predictive analytics, we are interested to perform experiments on the areas, Supply Chain Risk Management, Credit Scoring and Bankruptcy Prediction. When comparing to previous studies on this topic, our research is novel in the following areas. All the experiments carried out in this paper have used three different application specific data repositories that are described in detail in Design and implementation section. Focused on making use of traditional predictive techniques Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA) and compared their performance with respect to Accuracy, Misclassification, Precision, Recall, prevalence and F-Score. we compared the performance of classification algorithms like: Naive Bayes, K Nearest Neighbor with respect to Error in Classification. Analyzed the performance of Model Averaging generation algorithm with respect to average of Markov Blanket size, Neighbourhood size and Branching factor. The main finding from our research is LDA is very good choice when modeling Supply Chain Risk Management, Credit Scoring and Bankruptcy Prediction.

Variations in the cost for the same diagnosis among different hospital providers is a great concern to the public at large. With huge amounts of data being availed every second, utilising the data for the benefit of the society is commendable. In this research a neuro-fuzzy approach is proposed for Medicare payments data. Machine learning clustering algorithms on neuro-fuzzy results are compared to understand the variations in price for same treatment and diagnosis among different healthcare providers. Cluster analysis has been applied in various domains to help reveal hidden structures. Cluster analysis has not been well exploited in healthcare claims datasets, the reason being that healthcare expenditure data is highly skewed which make analysis complicated. The Inpatient charges is a large dataset that has 163065 and 12 attributes describing amounts paid by Centers for Medicare and Medicaid Services (CMS) to different healthcare providers using different Diagnostic Related Group (DRGs).

In this paper, a mathematical model with flexible negotiation strategies for agent based negotiations is developed which can be applied suitably in bilateral/multilateral multi-issue negotiation environments. Unlike the existing approaches for offer value computation for the negotiation issues, this model considers not only the reservation values but also the offer values proposed in the preceding negotiation round. This approach for offer value computation enables the traders to reach consensus much quicker than the existing approaches. This model considers the compelling urge of the trader in buying/selling a product based on which the reservation values are adjusted automatically at the end of the negotiation process in order to reach consensus in a deal which is otherwise not possible. The formula devised in this model to determine the concession speed of each negotiation issue handles the dynamicity of the negotiation environment and reflects the importance of each negotiation issue from the traders' perspective. The effectiveness of the proposed strategies is evaluated using various hypothetical cases representing the real-world negotiation scenarios in an e-commerce environment. The test results show that the proposed negotiation strategies are able to optimize the utility process and also improve the rate of reaching consensus in the negotiation process .

Recent studies have shown that ''hacktivists'' can mount serious attacks on automobiles. The automotive On-Board Diagnostic (OBD) interface enables an efficient way to access information of the in-vehicle electronic system and leaves way for unauthorized access by an intruder. We discover that remote exploitation is feasible via a broad range of attack points such as mechanic tools, CD players, Bluetooth and Tire Pressure Monitoring System. Wireless communications channels allow long distance vehicle control. Finally, the paper discusses the drawbacks of Seed-Key Mechanism to authenticate and provide an extra layer of authentication to help build a safer automotive ecosystem.

To create a system that provides a comparison of multiple algorithms that may be tested in the Prisoner's Dilemma decision problem using two subjects in a dual agent environment. As an addition to understanding the effects of various algorithms and logic that helps influence a single agent's decision, our system aims at analysing the performance of the same algorithms in iterative and multi agent systems. The results are obtained by using concepts of Swarm Intelligence, Multiple Agent Systems and Super Agents within the testing system. The results of the research are to expose the advantages and disadvantages of each schema to help plan investments, predict outcomes and for real world application of the Prisoner's Dilemma in fields of Environmental Sciences, Psychology, Economics and many more such fields.

Abstract− The optimized fertilizer usage for better yield of rice cultivation is influenced by key factors like soil fertility, crop variety, duration, season, nutrient content of the fertilizer, time of application etc., It is observed that 60 percent of yield gap in tamilnadu is due to farmers lack of knowledge on key factors and informal sources of information by pesticide dealers. In this study the major contributing factors for fertilizer requirement and optimum crop yield were analyzed based on rough set theory. In data analytics perspective the nutrient plan is sort of multiple attribute decision-making processes. To reduce the complexity of decision making, key factors that are indiscernible to conclusion are eliminated. Our rough set based approach improved the quality of agricultural data through removal of missing and redundant attributes. After pretreatment the data formed as target information, then attribute reduction algorithm was used to derive rules. The generated rules were used to structure the nutrition management decision-making. The precision was above 88% and experiments proved the feasibility of the developed decision support system for nutrient management.

Cloud computing focus on utility of virtualised network with
infrastructure and application services through data centres. Currently,
maintenance of patient’s medical records and reports are through centralised
system without sharing facility with other health centres. There is no freedom
to the patients to obtain their medical information like treatment reports,
discharge summary, etc., via mobile devices. A model proposed in our work
where the medical information and reports will be stored in an ecosystem
with synchronisation, so that they can be shared through different healthcare
clouds and can be accessible to the patients’ mobile devices. Randomised
alphanumeric cipher (RAC) algorithm used to enhance security for keeping
medical reports of patients in cloud with availability and interoperability.

Abstract: Locating relevant Information in Peer-to-Peer (P2P) system is a challenging problem. Conventional approaches
use flooding to locate the content. It is no longer applicable due to massive information available upfront in the P2P systems.
Sometime, it may not be even possible to return small percent of relevant content for a search if it is an unpopular content. In
this paper, we present Adaptive Semantic P2P Content Indexed System. Content Indices are generated using topical semantics
of documents derived using WordNet Ontology. Similarities between document hierarchies are computed using information
theoretic approach. It enables locating and retrieval of contents with minimum document movement, search space and nodes
to be searched. Results illustrate that our work can achieve results better than Content Addressable Network (CAN) semantic
P2P Information Retrieval system. Contrary to CAN semantic P2P IR system, we have used content aware and node aware
bootstrapping instead of random bootstrapping of search process.

Cloud defines a new age of computing solution that
provides services to customers with its unique features of agility
and multi-tenancy. As the critical resources are hosted at cloud
provider’s end, security is a big challenge in cloud computing. If the
cloud environment is compromised and attackers get the access of
core data centers, the availability of the critical resources becomes a
big concern for the service consumers. Denial of Service and
Distributed Denial of Service kind of attacks are launched towards
cloud environment to make the resources unavailable for legitimate
users. In this paper we propose a fuzzy logic based defense
mechanism that can be set with predefined rules by which it can
detect the malicious packets and takes proper counter measures to
mitigate the DDoS attack. Also a detailed study of different kind of
DDoS attack and existing defense strategies has been carried out.
Keywords: DoS, DDoS, fuzzy logic, anomaly

Security is the governing dynamics of all walks of life. Here we propose a secured medical diagnosis system. Certain specific rules are specified implicitly by the designer of the expert system and then symptoms for the diseases are obtained from the users and by using the pre defined confidence and support values we extract a threshold value which is used to conclude on a particular disease and the stage using Rule Mining. “THINK” CAPTCHA mechanism is used to distinguish between the human and the robots thereby eliminating the robots and preventing them from creating fake accounts and spam’s. A novel image encryption mechanism is designed using genetic algorithm to encrypt the medical images thereby storing and sending the image data in a secured manner.

Abstract This paper proposes the utilization of mobile agents to increase the performance in maintaining the consistency of web caches. Most popular browsers like Internet Explorer and Netscape are providing options to cache the frequently used web pages in the client sites itself in order to provide faster access to the clients and to reduce the network overhead due to the reduction in number of messages transferred. In this environment, the difficulty is in maintaining the web caches in a consistent state by updating the web pages that are ...

Mobile agent systems provide a great flexibility and customizability to distributed applications like ebusiness and information retrieval in the current scenario. Security is a crucial concern for such systems, especially when they are to be used to deal with money transaction. Mobile agents moving around the network are not safe as the remote hosts that accommodate the agents can initiate

Performance and scalability of the Web server is significantly affected by its architecture and software contention problem. This paper proposes agent-based architecture to manage user sessions of an e-business system. Each thread of control from the connection pool of the Web server is used to create an agent and is used to manage the sessions of multiple clients. It improves the scalability and performance, through the efficient usage of Web server software resources like connection threads. The proposed model is implemented ...

Abstract: E-Business must be highly secured and scalable to provide efficient services to millions of clients on the web. This paper proposes a new approach based on shared objects to improve security and mobile agents to improve scalability. The e-business uses shared objects and mobile agents to update the clients automatically with new information. The agent that resides in the database server is informed about the new information by triggering a function. Then the agent updates the shared object which is accessed by ...