IJCTA-Volume 8 Issue 4 / July-August 2017
Title/Author Name
Page No
BIG DATA Applications:A Technical Review
-Ankit Sarawagi,Rajeev Pandey,Raju Barskar
Big data are data sets which are beyond the power of storage and processing capacity. The huge quantity of information is generated from applications like social networks, Internet, sensors, Bio-informatics, Weather forecasting, etc. Processing of this much huge dataset using Traditional database systems is impractical. To process these data sets, two processing techniques Batch processing technique and non-batch processing technique are used. In Batch processing, huge data are gathered over a time-frame, processed and then produces output. Non-batch process involves continuous input of data and processed within a very little time-frame. With the presence of various documents, literatures and papers in the field of big data processing techniques, it was found difficult for the new scholars analyze the big data generated by various application areas with various processing techniques. In this paper, the suggestions of the most suitable and efficient processing techniques are given to process many types of datasets generated by various application areas, along with the challenges in processing and storing, and advantages of analyzing it. Finally, comparison table is made to suggest the efficient technique. Suggestions of more efficient techniques will help new scholars to opt better technique to handle big data generated by various applications and to give optimum results.
Keywords-Big Data, MapReduce, Apache Hadoop, Apache SPARK, Storm, S4.
A Survey on Sequence Alignment Based on Hadoop/MapReduce Big Data
-Prabhat Gupta,Rajeev Pandey,Anjna Deen
In Bioinformatics multiple sequence alignment (MSA) could be an essential step in a few bioinformatics examinations, and conjointly a NP-hard downside. To support the speed, precision and take into account the need of extensive scale arrangements arrangement, a wide Variety of MSA systems and programs are later created. We additionally consider future advancement of MSA techniques with connection to applying of all the more entirely unexpected advances and consequently the possibility of parallelization of MSA. All of those algorithms have demonstrated their potential to unravel several alignment issues. This paper provides AN in-depth survey of well-known sequence alignment algorithms and Hadoop map-reduce approach which is used to solve big data efficiently.
Keywords— Sequence Alignment, Multiple sequence Alignment, Big Data, Hadoop, Map-Reduce
IOT Enabled Immediate Response System for People in case of Road Accidents
-Vibhu Kapoor,Kopal Tripathi.Dr.Pradeep Kumar Singh
Every day, a large number of people die from accident injuries all over the globe. An effective approach to control the post accidents impact which may help in saving the life of the people by sending alert notifications to hospitals, police station and their relatives. Alert notification willhelp in getting the quick response in case of accident occurnce. Few systems are already in place as built in vehicle accident detection systems. However, these systems are very costly and lack in terms of availability. Whereas, to detect traffic accidents using smartphones is easily possible now because of the advances in connectivity and sensors deployed on the smartphones. The aim of this paper is to reduce the response time using Internet of Things(IOT) . The proposed system includes two phases; the detection phase and the notification phase. The detection phase detects car accident in all speeds. The notification phase is used to send detailed information such as the accident location to the hospital, police station and relatives of the person who met with an accident immediately after an accident is detected, for fast recovery. This paper includes two contributions; of using smartphone-based accident detection systems. First, it provides solutions to critical issues such as fake/false alerts by deploying onboard sensors todetect large accelerations. Second, it presents the system architecture of the prototype smartphonebased accident detection system.
Keywords-Internet of Things, Reponse Time, Sensors
A Survey on Differente Multiple Sequence Alignment and S.I Algorithm
-Renuka Devi,Aditya B,Abhinav Rajput
Multiple sequence alignment (MSA) could be a basic step in several bioinformatics analyses, and conjointly a NP-hard drawback. So as to boost the speed, accuracy and cater to the necessity of largescale sequences alignment, a wide variety of MSA strategies and software are afterwards developed. We also contemplate future development of MSA strategies with relation to applying of more totally different technologies and therefore the prospect of parallelization of MSA. Many swarm optimization algorithms are introduced since the first 60’s, biological process Programming to the foremost recent, gray Wolf optimization. All of those algorithms have demonstrated their potential to unravel several optimization issues. This paper provides AN in-depth survey of well-known optimization algorithms. Elect algorithms square measure shortly explained and compared with one another comprehensively through experiments conducted using thirty well-known benchmark functions. Their blessings and drawbacks also are discussed. Variety of applied mathematics tests square measure then dole out to work out the many performances. Keywords-Big Data, MapReduce, Apache Hadoop,Apache SPARK, Storm, S4.
Analysis of Scientific Paper Recommendation System Using Article Side Information
-Rachana R.Jadhav,Prof.N.R.Wankhade
Retrieving scientific article recommendation system plays important role for research in the field of academia. Most of the existing systems have designed integrated techniques for all target researchers therefore some of the algorithm generates recommendations for the researchers. In this paper we proposed different method to generate recommendation which is based on side information available in the papers such as citations and contents. The contents will be the abstracts of that paper. We use various similarity measurements like Euclidean Distance, cosine distance, Jaccard Coefficient, dice coefficient and SMTP for text processing distance measures calculating similarity between contents in the document. As a result we found that SMTP works better than other four similarity measures. We proposed a content based recommendation along with collaborative filtering which considers contents or unique words from the paper to recommendation purpose where top k recommendations were more closely similar to the target researcher’s document. We perform an experiment on real world dataset and compared results demonstrate that proposed system gives more accurate results than baseline system.
Keywords— Content based recommendation, SMTP, Unique words, text processing, Collaborative Filtering, Article Recommendation.
Millennium of Virtual Reality in e-Commerce: Proposed Virtual Environment and prototype
-Sakshi Taaresh Khanna
The research paper is aimed at examining the online shopping market from various angles as well as analysing and evaluating the effect of virtual reality on it. The paper begins with introducing various fundamentals of Virtual Reality to familiarize the reader with basic concepts and gradually explains the working of VR Systems. The paper also aims to focus on major developers in the VR industry and further to the myriad of applications that VR can have in various domains. The paper mainly aims at the application of VR in online shopping, its concepts and the current researches. This paper proposed the features of Virtual environment for virtual stores and prototype of a VR based online shopping system.
Keywords: Virtual Reality, Immersive experience, E-commerce
The Indian IT Act-2000 in context to addressing the needs of the Cyber Crimes in India
-Prashant Vats,Prof. (Dr.) Sanjay Singh
Rapid advancements in Information Technology sector have revolutionized work and personal lives of people globally. Technology has entered every sphere of life like banks, work place, social networking, stock markets, shopping etc. resulting in sharing of one’s personal information with every bit of machine one comes across. With the availability of personal information on a single click, the data is vulnerable to cyber-crime. In mid-90s liberalization of Indian economy resulted in manifold increase in e-transactions. Therefore, the need to bring technology under legislation was felt. With this objective in view, Parliament of India, passed the Information Technology Act in 2000. This paper describes IT Act 2000, and discusses important dimensions of amendments in 2008.Further, various notifications issued till date have also been discussed. An attempt has also been made to touch areas, left unaddressed in the foregoing legislations.
Keywords: IT Act 2000, Cyber Crime, IT Act Amendment 2008.
Cubic & Quartic Equation based Secure Authentication among Internet of Things with AES Encryption
-Dharminder Singh,Ambica Verma
The two major security paradigms for IOTs are authentication & encryption mechanisms, which have many variants. In this paper, the work has been carried over the enhancement of the proposed security model by designing the authentication model with set of algebraic equations. The multi-column based complex key generation is designed around the algebraic equations, specifically cubic and quartic equations. The authentication keys and data encryption is another security paradigm of the proposed model, which utilizes the advanced encryption standard (AES), which has been used for the implementation of the high security protocols. The performance of the proposed model has been analyzed under the different scenarios with variable number of nodes (50, 100 and 150) with decreasing transmission range of 75, 50 and 25 respectively. The proposed model has been recorded with minimum projected resource readings at 1.09, 5.49 and 10.96 percent in the scenarios with 50, 100 and 150 nodes respectively, whereas the maximum readings are 1.95, 9.81 and 19.63 percent for similar scenarios. The maximum value of entropy has been recorded at 2.98, 3.39 and 3.29 for proposed model against the 2.26, 2.53 and 2.39 existing model, which shows the robustness of the proposed security model for IOTs. The minimum values of entropy are recorded between 1.99 and 2.16 for proposed model against the existing range of 1.61 and 1.77 for the entropy, which again shows the similar trend to the latter analysis. Hence, it shows the clear improvement of proposed model against the existing model in all experiments.
Enhanced Text Recognition and Extraction Approach from Image Email
-Mallikka Rajalingam
Text extraction and text line detection is the foundation of document image analysis. Since many years, a large number of text detection methods have been proposed, where these methods depend on convinced assumptions of documents with various font style, font size, distorted text, uneven lighting, complex background and low resolution. In this paper, reveals k-nearest neighbor rule as a generic text-line detection and text extraction approach that can be applied on a complex mail document images. The performance evaluation of transition map generation and it compares with other two models is presented in this paper. Experimental analysis shows that image based text Optical Character Recognition (OCR) method is to extract the text from the colour image and detection of advertised mails is very efficient than that of the other existing methods.
Keywords- Text extraction, Text-line detection, Optical Character Recognition, KNN rule, and image mail document.
Face Sketch Recognition by Enhancing Evolutionary-Based Model
-Ali Salem Ali Bin Sama
Face sketch recognition models play a vital role for investigators in solving criminals. The task of these developed models is to perform matching between a drawn sketch image, and photo images at the database of law enforcement agencies. A forensic artist based on the description of an eyewitness draws the sketch image. However, one of the main challenges in problem of face sketch recognition is the variation between the drawn sketches and the original face photo. This work introduces an Evolutionary-Based Model (EBM) to tackle the problem of face sketch recognition. In particular, EBS is employed to perform two different operations, which are parameters tuning during the training phase, and localization of sketch facial components (e.g. face, eyes, nose, and mouth) at sketch recognition phase. To evaluate the effectiveness of the proposed model, a number of face sketch dataset images are used including CUHK, AR, and FERET. Experimental results show that the proposed method provides better identification performance compared to existing methods.
Keywords: Face Sketch Recognition, Evolutionary Algorithms, CUHK face sketch dataset, FERET sketches, AR sketches and HOG features.
Enhanced Feature Extraction Technique for Iris Template Generation
-Sarika B.Solanke,Ratnadeep R.Deshmukh
In the process of iris recognition localization and segmentation are the primary preprocessing step that locate the iris and segments it from the remaining part of the image. The normalization and feature extraction are very important and crucial stages which prepare the iris image to extract the required unique features and create the templates that can be compared to find the uniqueness among the two irises. Normalization allows transformation of iris region into a fixed size dimensions that can be compared. In short normalization produces constant dimension size images of every localized input iris image which are ready to further processing and comparison of such images is possible due to their nature of constant dimension size. In general the normalization process prepares the segmented image for feature extraction process. This paper discusses the enhanced normalization process based on Daugman’s rubber sheet model and feature extraction is based cumulative sum based change analysis. Iris features are extracted and the iris template is generated by horizontally and vertically grouping the iris texture features as iris codes.
Keywords : Iris Recognition ,Black hole search mehod, Camus & Wilde’s Method, Daugman’s Integro Differential operator, Rubber sheet model, Cumulative Sum Based Change Analysis
Melanoma Skin Cancer Analysis Using Thresholding Method
-Dr.A.Mercy Rani,S.Maheshwari
Image processing is a dynamic research area in which medical image processing is a highly challenging field. Medical imaging techniques are used to capture the interior portions of the human body for medical diagnosis. Skin cancer is a most dangerous skin diseases and it is identified as an uncontrolled growth of abnormal skin cells. It is caused by ultraviolet radiation from sunshine. Melanoma is the most hazardous type of skin cancer. In this proposed method, Melanoma skin cancer is diagnosed using image processing technique. The different types of digital lesion images have been analyzed based on image acquisition, pre-processing, and image segmentation technique. The image segmentation technique is applied to extract the affected portion of the skin input image. After the skin lesion region is extracted the ABCD (Asymmetry, Border, Color and Diameter) features and TDS (Total Dermatoscope Value) are calculated. The experimental results are evaluated and the Melanoma skin cancer is diagnosed based on the TDS values of the image.
Index Terms— Melanoma Skin Cancer, Image Acquisition, Image Pre-Processing, Image Segmentation, ABCD Rule, Melanoma Skin Cancer Detection.
Enhanced the Reliability and Security of Communication in Vehicular Adhoc Network Using COM-AODV Model
-Nidhi Saxena,Dr.Shilpa Sharma
The advancement of communication technology drives the concept of intelligent transporting systems. The intelligent transportation systems need the reliable and secured communication. The reliability and security of communication depends on the performance of vehicular ADHOC network. For communication, no any standard protocol is available. Some authors are used the extended from of AODV routing protocol, for the reliability and security of communication design COM-AODV Model. In this paper proposed the COM-AODV Model. The COM-AODV Model is basically work is filter. The filtration process depends on the cooperative mechanism of MAC layer and network layer. The processing of COM-AODV Model depends on the cross-platform protocol. The proposed COM-AODV Model simulated in MATLAB software and used sumo traffic for the simulation of proposed model. The proposed model compare with some other protocol of VANET network. The experimental results show that better improvements instead of previous algorithm of reliable communication.
Keywords: - VANET, MAC, Network layer, COM-AODV Model, RSU, Security.
Image Segementation Techniques to Detect Nail Abnormalities
Medical Image Processing plays a vital role in diagnosing various diseases like skin cancer, brain tumor, breast cancer and diseases related to lungs and heart. Human nail acts as a window of the human body for diagnosing the diseases. The segmentation techniques to extract the infected nail regions and their shape attributes are calculated and analyzed. Initially the nail image is segmented using Watershed, Thresholding and K-means segmentation Techniques. Then, the Shape features of the segmented nail region are extracted which can be further used in the diagnosis of nail diseases. The results of the three segmentation techniques are compared and analyzed based on these features.
Keywords: Medical Image Processing, Finger Nail, Nail Disease, Watershed, Thresholding and K-means segmentation
A Noval Image processing Technique To Extract Facial Expressions from Mouth Regions
Emotion recognition is the process of identifying a human emotion, most typically from facial expressions. Different types of facial expressions are Joy, Sadness, Fear, Disgust, Surprise, and Anger. In this paper, an image processing technique to recognize various facial expressions from mouth regions is proposed. The mouth regions are initially located by means of Viola-Jones algorithm and cropped. Then Region Based Segmentation is applied to segment the mouth region. Morphological area filling and boundary extraction methods are applied to extract the boundary of the mouth region. Since morphological operations are used the shape and size features are retained. Then the area of the mouth region is calculated from the number of white pixels extracted and the range of values for each emotion is identified. The proposed technique is executed on various emotional images (natural, joy, angry, surprise) of two different persons. The results are analyzed and their performances are evaluated.
Keywords: Emotion recognition, Viola-Jones, Region Based Segmentation, Morphological area Extraction, Feature Extraction.
Different Aspect of Community Detection Methods in Social Network
-Pranavati Jadhav,Dr.Vijaya Babu
Social Network defined as a set of human, companies, programs, computers, information and knowledge processing entities. They are grouped together by set of relationships for information flow in social network. Social network analysis leads to method to analyse patterns of social relationship between social entities. Community Detection integrated with social network analysis as branch of mathematical sociology. The paper exposes different, multi-views of social network analysis in terms of members, organization, social websites, and social web services etc. The paper reviews various community detection methods for views of users. Main goal of the paper is to impart basic knowledge for researchers who are working on community detection in Social Networks
Analysis of Extended Performance for clustering of Satellite Images Using Bigdata Platform Spark
Due to the recent emergence Clustering techniques have been widely adopted in many real world data analysis applications, such as customer behavior analysis, targeted marketing, digital forensics, etc. As the satellite imagery is getting generated at a higher rate than the previous decades, it becomes essential to have better solutions in terms of accuracy as well as performance. In this paper, we are proposing the solution over big data which performs the clustering of images using different methods viz. Scalable K-means++, Bisecting Kmeans and Gaussian Mixture. Since the number of clusters is not known in advance in any of the methods, we also propose a better approach of validating the number of clusters using Simple Silhouette Index algorithm and thus to provide the better clustering possible.
Keyword: Images, Distributed Processing, Scalable Kmeans++, K-means Clustering, Bigdata, Datamining, Security, Gaussian Mixture
Cooperative PSO-ACO Approach for Job Scheduling and Load Balancing in Public Cloud Network
-Indra Nath Sahu,Dr.Jitendra Sheetalani
Resource allocation and management of job is major issue in public cloud networks. For the allocation of resource and management of job used scheduling techniques. The conventional and dynamic job scheduling techniques increases the time span and failure of jobs are occurred. The failure of job and resource raised the problem of load balancing. For the management of load balancing used various dynamic methods based on probability theory and heuristic function. In this paper proposed the hybrid methods of load balancing for public cloud networks. The hybrid methods are combination of two well know heuristic function particle swarm optimization and ant colony optimization. The particle swarm optimization maps the job according to their resource and the ant colony optimization works as scheduler of jobs. The proposed algorithm simulated in cloudSim simulator and measure two parameter one is data processing time and other is response time.
Keywords: - Cloud Computing, Load Balancing, ACO, PSO, CloudSim.
Pattern Extraction and Analysis of Health Care Data Using Rule Based Classifier and Neural Network Model
-Rahul Deo Sah,Dr.Jitendra Sheetalani
The mining of health care data is important aspect for the prediction and estimation of critical disease of previous record. For the process of health care data mining various tools and technology are used. In concern of technology data mining algorithm are used. Data mining offers various algorithms for the purpose of mining. The bucket of data mining technique is association rule mining, clustering, classification and regression. The association rule mining technique is very important phase of pattern analysis. The association rule mining estimate the correlation coefficient for the relative attribute for the mining process. The clustering and classification technique play major rule in health care database. The process of clustering defines the process of similar pattern on the basis of iteration. Instead of data the classification technique guided grouping technique on the basis of certain guidance. In this paper also discuss the process of pattern analysis of health care data using different classification based on rule mining technique.
Keywords: - Classifier, Rule Based Classifier, KNN, SOM, Neural Network.
Analysis of Query Optimization in DBMS and Proposed Optimization Techniques
-Dr.Amit Mishra,Daniel Tsado,Usama M Gana,Elijah Joseph,Abdulganiyu Abdulrahman
Query optimization is a function of many relational database management systems in which multiple query plans for satisfying a query are examined and a good query plan is identified. A broad work in query optimization has been on since the early days of ‘70s. The development of Object-Oriented database systems started in the mid days of 80’s, this resulted in order to measure up to the requirements of applications beyond the data processing applications which are served by relational database systems. It has been observed that it is difficult to capture the extensiveness and complexity of this large body of a work. Therefore, we wish to focus on techniques of analysing query optimization in relational as well as Object-Oriented database Management Systems and we wish to present an analysed view of this field based on existing works. The query execution engine implements a set of physical operators. An operator takes as input one or more data streams and produces an output data stream. Examples of physical operators are (external) sort, sequential scan, index scan, nested loop join, and sort-merge join. We refer to such operators as physical operators since they are not necessarily tied one-to-one with relational operators. The simplest way to think of physical operators is as pieces of code that are used as building blocks to make possible the execution of SQL queries. The edges in an operator tree represent the data flow among the physical operators. We use the terms physical operator tree and execution plan (or, simply plan) interchangeably. The execution engine is responsible for the execution of the plan that results in generating answers to the query. Therefore, the capabilities of the query execution engine determine the structure of the operator trees that are feasible.
Keywords:SQL,QueryOptimization,OODBMS,Optimization techniques.
IJCTA © Copyrights 2015| All Rights Reserved.

This work is licensed under a Creative Commons Attribution 2.5 India License.