Prakash V S
@kristujayanti.edu.in
Assistant Professor II
Kristu Jayanti College
RESEARCH INTERESTS
Data Mining , Network Security , Network,
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Ms.Sangeetha J*, , Dr.Sinthu Janita, and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
In recent years, the online shopping and the online advertisement businesses is growing in a vast way. The reason behind this growth is, the peoples are not having sufficient time for go for a shop. Without seeing the quality of the product directly, the people are ready to buy the product by seeing the other user recommendation of the particular product. This leads an interest / the need to develop the researcher an innovative recommendation framework. Based on the opinion prediction rule, the huge size of words and the phrases which are presented in the unstructured data is modified as a numerical values. The sale of the particular product in an online shopping is depends on its description of the quality, the review of the customer. Based on the positive and negative polarity, an Inclusive Similarity-based Clustering (ISC) is proposed to cluster the extracted related keywords from the user reviews. To evaluate the strength, weakness of the product, estimate the respective features, as well as the opinions, the Improved Feature Specific Collaborative Filtering (IFSCF) model for the feature with aspect opinion is proposed. Finally the complete feedback of the product is estimated by propose the Novel Product Feature-based Opinion Score Estimation process. The main challenge in this recommendation system is the fault information estimation of the reviews and the unrelated recommendations of the bestselling or the better quality product. To neglect these issues, an Enhanced Feature Specific Collaborative Filtering Model based on temporal (EFCFM) is proposed in the recommendation system. Hence the developed EFCFM method is investigated by comparing along with the existing methods in terms of subsequent parameters, precision, recall, f-measure, MAE and the RMSE. The outcome shows that the developed EFCFM approach predicts the best product and produce the accurate recommendation to the customers.
S. Latha*, , Dr. V. Sinthu Janita Prakash, and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Computer Networks are prone to be attacked by a number of network attacks. To protect an individual system or the entire network from the malicious behaviour, a high level security system is needed. Intrusion detection system (IDS) is a system which give such protection to the network from the intrusions like misuse, unauthorised access etc. Even though many forms of new attacks come into practice, providing the security for the system from the known attack is also a challenging task. The solution is a Signature based IDS which is a potential tool to identify the known attack, sending alert and protect the networks. So a novel signature based IDS(ORG-IDS) with four phases such as Feature Selection, Classification, Optimized Rule generation and Pattern matching is proposed. For any efficient signature based IDS, it should have the signature rules in less number but it should be effective in identifying attacks with good time and memory complexity. In this paper, a new algorithm is proposed for Rule generation phase of proposed IDS to configure the rules by implementing Ant Colony Optimization Technique with Association Rule Mining . The parameters like number of rules, running time and memory utilization are measured and proved that this proposed algorithm outperforms the other existing algorithms.
Mrs. A. Sahaya Jenitha, , Dr. Janita pursed, and
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
This paper proposed on Poisson process-based algorithm is to carry out content-level deduplication for the streaming data. Since Poisson processes are meant to do the counting of different events happening over a period of time and space, it becomes appropriate to use it for identifying duplications of data as it gets streamed based on time and space, which can allow the deduplication process to be carried out in tandem. Some of the research on deduplication has been focusing on File-level and Block-level deduplication while the focus can be brought to content-level, as data get streamed lively and becomes more dynamic. With this approach, the content-level deduplication will allow the data to be scanned intelligently and at the same time, it can save the deduplication operation time. Also, streaming data has its randomness which is innately there and by having Poisson process based deduplication it will address the random behaviour of the data transfer and can work efficiently in the dynamically connected environment. The proposed method identifies the unique data to store in the Database. Based on the experimental result, the Poisson Process-based algorithm produce 0.912 Area Under Curve (AUC) accuracy on real-world streaming data, which means that if AUC is greater than 0.8 then the performance of algorithm is pretty good. So, the machine intelligence-based deduplication model produced reliable and robust deduplication on streaming data compared to existing approaches.
R. Merlin Packiam and V. Sinthu Janita Prakash
Springer Singapore
H. Krishnaveni and V. Sinthu Janita Prakash
Springer Singapore
A. Sahaya Jenitha and V. Sinthu Janita Prakash
Springer Singapore
J. Sangeetha and V. Sinthu Janita Prakash
Springer Singapore
R. Brendha and V. Sinthu Janita Prakash
Springer Singapore
J. Sangeetha and V. Sinthu Janita Prakash
IEEE
In recent days, the big data opinion mining is considered as an important research area because, the size of the user reviews increase in petabytes. Opinion classification is the process of identifying the positive, negative and neutral opinion from the text. The traditional data mining software tools find it difficult to manage the reviews because of its size. Hence, to provide an efficient big data opinion mining, an efficient Inclusive Similarity based Clustering (ISC) algorithm is proposed. During pre-processing the data is cleaned with Parts of Speech (PoS) tagger and sliced by using the proposed Threshold based Data Partitioning (TDP) algorithm. After pre-processing the dataset, the inclusive similarity algorithm computes the similarity between the consecutive reviews and constructs an adjacency matrix. The proposed ISC algorithm exploits the adjacency matrix and merges the clusters into a single cluster. The performance of the proposed method is validated with the existing algorithms for the metrics such as time consumption, memory utility and accuracy. The comparison results prove that the proposed ISC algorithm provides optimal results for all the metrics than the existing algorithms.
S. Latha and Sinthu Janita Prakash
IEEE
Network Security is playing a vital role in all types of networks. Nowadays the network is implemented in all places like offices, schools, banks etc. and almost all the individuals are taking part in social network media. Even though many types of network security systems are in use, the vulnerable activities are taking place now and then. This paper presents a survey about various types of network attacks mainly web attacks, and different Intrusion Detection Systems(IDS) which are in use. This may pave a path to design a new type of IDS which may protect the network system from various types of network attacks.
R. Brendha and V. Sinthu Janita Prakash
IEEE
Vehicular Ad Hoc Networks (VANETs) is an essential and emerging area of research in the field of Ad Hoc Networks. The main objective of deploying VANET is to improve the road safety and reduce the number of accidents. In VANET, routing is a difficult task because of the high mobility of nodes, which causes rapid changes of topology and to deliver a packet within a minimum period of time. Existing routing protocols are not sufficient to meet all the issues in routing. To provide best routing protocol, it is necessary to make an analysis of routing protocols in VANET. This paper starts with the basic architecture of VANET and provides a detailed description of various existing routing techniques with its advantages and disadvantages. Finally, this paper discusses an overview of the existing routing protocols for VANET.
R. Merlin Packiam and V. Sinthu Janita Prakash
IEEE
Today's world is flooded with unstructured information. Big data is not just a description of raw volume but it has to real issue of usability. The major part of information retrieval is giant experience in big data. The real challenge is identifying or developing most cost effective and reliable methods for extracting value from all the terabytes and petabytes of data now available. That's where big data analytics become necessary. Conventional analytics focused on structured data but these methods are not appropriate for large volume of unstructured data in order to extract knowledge. Text analytics is the way to extract significance from the unstructured text to find out patterns and transformations. The importance of text analytics is increased more in social media and business intelligence. This study reveals that big data text analytics can breed new insight to the world of text information and discusses various researches carried out in text analytics.
Pranav Koundinya, Sandhya Theril, Tao Feng, Varun Prakash, Jiming Bao, and Weidong Shi
IEEE Conference Publications
R. Malathi, M. Karthik kumar, S. Ramkumar, and V. Prakash
Maxwell Scientific Publication Corp.
The aim of this study is to weigh the pros and cons of SaaS ERP. The hype cycle surrounding Software as a Service (SaaS) has been escalating over the past several years, building to a crescendo entering 2012. Enterprise Resource Planning (ERP) holds a special place in the grand scheme of SaaS. While companies seemed willing enough to let the applications that surround and extend ERP reside in a SaaS environment, they were less willing to place their systems of record in a cloud they did not specifically own or control. The objective of this study is to prove that today the trend is changing and the advantages of ERP as SaaS appears to be winning.
S. Vanthana and V. Sinthu Janita Prakash
IEEE
A Mobile Ad-Hoc Network (MANET) is a collection of wireless mobile nodes forming a temporary network without using any centralized access point or administration. MANET protocols have to face high challenges due to dynamically changing topologies, low transmission power and asymmetric links of network. An attempt has been made to compare the performance of two On-demand reactive routing protocols namely AODV and DSR which works on gateway discovery algorithms and a proactive routing protocol namely DSDV which works on an algorithm to constantly update network topology information available to all nodes for MANETs on different scenarios. In this paper comparison is made on the basis of performance metrics such as throughput, packet loss and end-to-end delay, and the simulator used is NS-2 in Ubuntu operating system (Linux). The simulations are carried out by varying the packet size, number of connecting nodes at a time and pause time and the results are analyzed.
Varun S. Prakash, Yuanfeng Wen, and Weidong Shi
IEEE
Magnetic tapes have been a primary medium of backup storage for a long time in many organizations. In this paper, the possibility of establishing an inter-network accessible, centralized, tape based data backup facility is evaluated. Our motive is to develop a cloud storage service that organizations can use for long term storage of big data which is typically Write-Once-Read-Many. This Infrastructure-as-a-Service (IaaS) cloud can provide the much needed cost effectiveness in storing huge amounts of data exempting client organizations from high infrastructure investments. We make an attempt to understand some of the limitations induced by the usage of tapes by studying the latency of tape libraries in scenarios most likely faced in the backing up process in comparison to its hard disk counterpart. The result of this study is an outline of methods to overcome these limitations by adopting novel tape storage architectures, filesystem, schedulers to manage data transaction requests from various clients and develop faster ways to retrieve requested data to extend the applications beyond backup. We use commercially available tapes and a tape library to perform latency tests and understand the basic operations of tape. With the optimistic backing of statistics that suggests the extensive usage of tapes to this day and in future, we propose an architecture to provide data backup to a large and diverse client base.
Varun S. Prakash, Xi Zhao, Yuanfeng Wen, and Weidong Shi
Springer Berlin Heidelberg
Tao Feng, Varun Prakash, and Weidong Shi
IEEE
In this paper, to enhance identity acquisition procedures in smartphones and make the process transparent to the user, a novel User Identity Sensing approach leveraging the unified fingerprint enabled touch panel that combines multiple capacitive TFT based fingerprint sensors directly with the touch screen panel of the smartphone is proposed. The solution passively fulfills mobile users identity management during natural user-device touch interactions and requires neither password nor extra actions from the user, which makes it highly user friendly. To demonstrate the feasibility of such a unified, user identity sensing based design, an investigation of the hardware metrics is performed. Simulation experiments are conducted to evaluate the system with touch data collected from 25 smartphone users. The resulting observations and simulation results provide guidance for an efficient design of the hardware and criteria that need to be satisfied for the development of an operational prototype.
Xi Wang, Xi Zhao, Varun Prakash, Zhimin Gao, Tao Feng, Omprakash Gnawali, and Weidong Shi
IEEE
Detecting Person-Of-Interest (POI), e.g., fugitives, criminals and terrorists in public spaces is a critical requirement of many law enforcers and police officers. In realty, most law enforcement personnel cannot effectively differentiate POIs from millions of faces and thus demand a portable assistant to recognize faces, in order to take the golden opportunity taking the POIs into immediate custody. Unfortunately, current face recognition systems are stationary and limited to a small scale of POI datasets. In this paper, we investigate a wearable computerized-eyewear based face recognition system. This system is a portable device which can accompany a police officer during patrolling or other tasks. The eyewear is connected to a cloud based face recognition system via wireless networks. Facial images captured by the mounted camera are sent to the cloud for identity retrieval. When the system finds a POI, it would alert officers via overlaying a virtual identity tag on the real POI's face on the transparent screen of the eyewear. We provide approaches to greatly minimize recognition time, including leveraging the large storage and high computational capacities provided by the cloud. The cloud enables nationwide POI database and supports parallel computing for face recognition.
Weidong Shi, Xi Wang, Xi Zhao, Omprakash Gnawali, Katherine Loveland, and Varun Prakash
IEEE
Due to hypersensitivity to sound, patients with autism spectrum disorders (ASD) can feel frustrated and even profoundly fearful when talking with multiple speakers. This exacerbates their impairments in social interaction and communication. We propose a fully interactive system that allows ASD patient to focus on a single auditory stream (a person's voice) according to their preference during conversations. The system has the capacity to filter out other speakers' voices based on distinguishing their locations. The experimental results have demonstrated our prototyping system works reliably in regular conversations.
Weidong Shi, Xi Wang, Xi Zhao, Varun Prakash, and Omprakash Gnawali
IEEE
Prosopagnosia (PA), or the inability to recognize faces, is widespread around the world, with no effective cure for the disease. We propose a wearable system to improve prosopagnosic's daily life. The system utilizes a real-time face recognition application that runs on a smartphone, and a portable eyewear that displays descriptive information when a person appears in range of camera mounted on the eyewear. The system helps these patients identify faces, reminds them to make social reactions and ultimately prevents them from developing social anxiety disorders, characterized by fear and avoidance of social situations that may cause embarrassment. Lastly, we implement the system in a real life demonstration to test its performance.
N.Venkata Subramanian, V. Prakash, and K.S. Ramanujam
Maxwell Scientific Publication Corp.
The promise of the cloud is appealing: reduced costs, greater agility, flexibility, scalability and potentially greater security. At the same time, IT organizations recognize that the cloud introduces a number of issues related to security, data integrity, compliance, service level agreements and data architecture that must be addressed. Therefore, the adoption of cloud services is being tempered by a significant level of uncertainty. Numerous surveys indicate that the top concerns for moving to the cloud are security, performance and availability. In other words, enterprises are looking for assurances that they are not adding risk to the business by leveraging the cloud. For many, moving to the cloud is still a leap of faith. Different cloud deployment models-public, private, or hybrid have different security vulnerabilities and risks. Generally, risk increases from greater degrees of multitenancy among increasingly unknown participants. The objective of this article is to insist the fact that cloud security begins with and adds to, well-defined enterprise security; it also introduces a new cloud security model called Cloud Security NXT.