IJCST Logo










 

International Journal of Computer Science and Technology
IJCST 9.4 (Oct-Dec 2018)

S.No. Research Topic Paper ID Download
01

A Novel Image Fusion Method Which Combines Wiener Filtering, Pulsed Chain Neural Networks and Discrete Wavelet Transforms for Medical Imaging Applications

Fayadh Alenezi, Ezzatollah Salari

Abstract

Medical image fusion is an important tool for correct decision making in clinical diagnoses. It aims to increase image information content and provide complementary context for anatomical and physiological information by creating a single composite image from two or more source images. Even so, existing image fusion techniques do not completely resolve the inherent limitations of fused images, particularly when considering lost textural information and visual detail. This paper proposes a new technique that uses a space-variant Wiener filter followed by the enhancement of the filtered images with lateral inhibition and excitation in a feature-linking Pulse Coupled Neural Network (PCNN) under maximized normalization. The enhanced images are then fused using a Shift-Invariant Discrete Wavelet Transform (SIDWT). The resulting fused images are then evaluated with standard objective criteria and compared with results from previously existing image fusion methods, showing significant improvements in the main performance indicators.
Full Paper

IJCST/94/1/A-0919
02

To Accomplish OSN Spam Users by Systematically Integrating Features: ProGuard

Pavada Gowthami, B.J.M.Ravi Kumar

Abstract

OSN with 20 million introduces multi day third party applications are a noteworthy explanation behind the ubiquity and addictiveness of Facebook. Shockingly, programmers have understood the capability of utilizing applications for spreading malware and spam. The issue is as of now critical, as we find that something like 13% of applications in our data set are malicious. Up until now, the exploration network has concentrated on distinguishing malicious posts and crusades. Paper, we make the inquiry: given a Facebook application, would we be able to decide whether it is malicious? enter commitment is in creating Proguard Facebook’s Rigorous Application Evaluator apparently the primary device concentrated on identifying malicious applications on Facebook. To create ProGuard, we utilize data assembled by watching the posting conduct of 111K Facebook applications seen crosswise over 2.2 million clients on Facebook. We investigate the genetic system of malicious Facebook applications and recognize instruments that these applications use to proliferate. Curiously, we locate that numerous applications plot and bolster one another; in our dataset, we find 1,584 applications empowering the viral spread of 3,723 different applications through their posts. Long haul, we consider ProGuard to be a stage towards making a free guard dog for application appraisal and positioning, to caution Facebook clients before introducing applications And enhanced online framework and by expanding Internet network with the goal that we can maintain a strategic distance from extortion and conning.
Full Paper

IJCST/94/1/A-0920
03

Internet: A Boon or a Curse for Children

Jayveen Patel, Aadi Surana, Ekangi Patel

Abstract

As the access to the internet reaches mass market in all countries; this research paper identifies and evaluates the emerging research agenda, focusing particularly on children and young people and examines the effects of internet on children. The major purpose behind this was bloom in internet since the last decade. Whole world is now dependent on the internet. Every work is done on the internet and life has become easy. Basically, it was found that children are spending more and more time on the internet, which had some advantages and disadvantages. According, to a research use of home internet by children improved their scores in the standardized reading tests. Many children included in the study were African and American aged around 13 yrs. And were underperforming in school ages too. But another survey showed that children who spent most of the time on the internet were less physically active and fit as compared to other children. So, we can’t predict by these surveys only that internet can be a curse or boon. We have to see all the aspects of the internet and how it will affect the children’s life. Whatever the conclusion may be but internet is the need of hour. Anybody cannot imagine life without the internet. The results show that use of internet for younger children i.e., less than 15 years can be harmful, and their mind can wander from studies. The use should be limited down to enhance physical, mental and social health of a child.
Full Paper

IJCST/94/1/A-0921
04

Detection of Abnormal Driving Behaviors Patterns with Advanced Notifications

Jailingu Gowri, Dr. G.Prasanna Lakshmi

Abstract

Road traffic injuries are the main source of death among individuals matured somewhere in the range of 15 and 29 years. Without supported activity, road traffic crashes are anticipated to end up the seventh driving reason for death by 2030. The greater part of the drivers are pompous and ignorant of their rash driving habits. Also, some strange driving practices are unapparent and are barely noticeable by drivers. To destroy the unsafe driving habits and to guarantee safe driving, the driver’s conduct ought to be observed. In this paper, we will investigate some examination works, which screen the drivers’ conduct. Driving conduct investigation is additionally a well-known course of Smartphone-based vehicular applications. Be that as it may, existing chips away at driving practices recognition utilizing dataset camera recordings can just give a coarse-grained result utilizing limits, i.e. recognizing anomalous driving practices from ordinary ones. Since limits might be influenced via auto compose and sensors’ affectability, they can’t precisely recognize the distinctions in different driving personal conduct standards.
Full Paper

IJCST/94/1/A-0922
05

An Efficient Secured Storage using Randomized Key Protocol in Cloud Computing

Malla Bhagyam, Praveena P

Abstract

Cloud computing opens new era in IT as it can give different flexible and adaptable IT services in a compensation as-you-go mold, where its clients can diminish the colossal capital interests in their very own IT foundation. In this theory, clients of cloud storage services never again physically keep up direct command over their data, which makes data security one of the real worries of utilizing cloud. Existing exploration work as of now enables data honesty to be checked without ownership of the genuine data record. Albeit a portion of the ongoing work based on BLS mark would already be able to help completely powerful data refreshes over settled size data squares, they just help refreshes with settled measured squares as essential unit, which we call coarse-grained refreshes. Accordingly, every little refresh will cause re-calculation and refreshing of the authenticator for a whole document square, which thusly causes higher storage and correspondence overheads. In this paper, we give a formal investigation to conceivable sorts of fine-grained data refreshes and propose a plan that can completely bolster approved inspecting and fine-grained refresh demands. Based on our plan, we likewise propose an upgrade that can drastically decrease correspondence overheads for checking little updates. Hypothetical examination and exploratory outcomes show that our plan can offer upgraded security and adaptability, as well as altogether bring down overhead for enormous data applications with countless little updates, for example, applications in online life and business exchanges.
Full Paper

IJCST/94/1/A-0923
06

Design Model for Information Classification in Big Data Environment

Prachi Paliwal, Gourav Sitlani

Abstract

Deep learning is a standard machine learning approach which has accomplished a hug of advancement with completely traditional machine learning in the domain of areas. The current issue centers around how to extract and classify the greatest, huge scale real dataset. Consequently, the task manages proposing a framework that can efficiently contribute a to a extremely proficient and precise capable prediction model. In the human health services domain: A point to point analysis of the patient records and patient live condition. Change and improvement of the existing frameworks by appending or replacing manual work with the utilization of huge data and information based artificial intelligence forms. Applying deep learning to these domains has been one among the prevalent points of research. The model given by us has the capability of maintaining high a long term accuracy. In case of an NDLCM network, a stacked NDLCM layer makes possible learning a high level temporal feature without any need of fine tuning and preprocessing that may otherwise be important in case of another technique. In this proposed paper, we construct a deep learning computational model which uses the normal back propagation neural network training set that helps to build a precise prediction model.
Full Paper

IJCST/94/1/A-0924
07

Data Security using Secured Key Protocol in Cloud Computing

Maddi Vedh, Praveena P

Abstract

In Cloud computing innovation there isapractice of critical strategy issues, which incorporate issues of protection, security, namelessness, broadcast communications limit, government reconnaissance, unwavering quality, and obligation, among others. Be that as it may, the most imperative between them is security and how cloud supplier guarantees it. Encryption is an outstanding innovation for securing touchy data. This paper exhibits an outline of security issues and furthermore breaks down the plausibility of applying encryption calculation for data security and protection in cloud Storage. It likewise attempted to cover the different algorithms utilized by specialists to take care of the open security issues. In this paper we have talked about cloud computing security issues, component, challenges that cloud service supplier look amid cloud building and introduced the figurative investigation of different security algorithms.
Full Paper

IJCST/94/1/A-0925
08

User Psychological Stress Discovery and Analysis based on Social Network stream Usage

Adigarla Damodara Gowrishankar, Praveena P

Abstract

The social networks are work to interchange images, recordings and reviews between the users. The users can associate from anyplace and whenever. The check in areas and time information are kept up under the Location Based Social Networks (LBSN). Area Based Social Networks (LBSNs) gives the office to get to users’ areas, profiles and online social associations. High versatile, high volume and high speed information things are handled utilizing huge information mining models. The social network information esteems are dissected to gauge the client behavior and stress levels. The connections of client communications and stress are broke down for the stress detection process. The social network information esteems are parsed and three kinds of characteristics are separated for the stress detection process. Stress-related literary, visual and social characteristics are extricated for the stress detection process. The hybrid model consolidates the Factor Graph and the Convolutional Neural Network (CNN) for stress detection. Tweet content and social connection are investigated in the hybrid model. The computerized stress revelation process is work with hybrid model, setting and association information. The area and time information are spoken to in the setting information. The stress detection process is enhanced with spatio worldly highlights. The hybrid model is improved to recognize staggered stress classifications. The inadequate client connation parameters are incorporated with the stress detection process. The hybrid model is improved to recommend treatment levels. Computerized stress discharge messages and recommendations are given by the framework.
Full Paper

IJCST/94/1/A-0926
09

Artificial Intelligence is Everywhere

Chandresh Phirke, Churchil Moondra, Gourav Sharma

Abstract

Artificial intelligence is the next stage of evolution. We are in an era in which we are surrounded by artificial intelligence knowingly or unknowingly. In this paper, we have discussed the overview of artificial intelligence starting from its history, goal, types, branches, applications to its misuse and threats. Artificial Intelligence has seen an immense progress in recent years. This progress is a result of various endeavours and diligent work contributed by many industrious scholars, mathematicians, analysts and various researchers. Artificial Intelligence (AI) technology can play an imperative role in economic development, resource conservation, and environmental protection by increasing efficiency. This paper gives us a better understanding of the field of Artificial intelligence and can be helpful in pursuing any branch of the same.
Full Paper

IJCST/94/1/A-0927
10

Implementation of Secured Mail Server for Group Communication

Arunadevi M

Abstract

Electronic communication is an emerging technique where we send information from sender to receiver in the form of E- Mail. To send and receive an Email, each user should have an ID. That ID must be locked with the unique password. The password is in the form of text. It may be alphabetical, numbers, alphanumerical and etc. Email servers provide the constraints to set the passwords, for the users. Even most of the servers secured, Black hat hackers hack the account and access the information. A graphical password is an authentication system that works by having the user select from images in a specific order, presented in a Graphical User Interface (GUI). The most common computer authentication method is to use alphanumerical usernames and passwords. This method has been shown to have significant drawbacks. For example, user tends to pick a passwords that can be easily guessed. On the other hand, if a password is hard to guess, then it is often hard to remember.
Full Paper

IJCST/94/1/A-0928
11

Fog Computing based Smart Home Controller

Dr. R. Hemalatha, P.Saranya

Abstract

The Internet of things (IoT) will be the Internet of future, as we have seen a large raise in smart home/city, smart connected vehicles, wearable technology, and smart grid. Cloud computing cannot solve all troubles due to its own disadvantages. Applications, for instance real, augmented reality, real time streaming and time gaming, are high in latency sensitive to install on cloud. The newest fashion of computing paradigm is to push elastic resources such as computation and storage to the border of networks, which inspires the promising computing paradigm of fog computing as a result of occurrence of all over the place connected smart devices relying on cloud services. Fog computing remains information and computation close to end users at the edge of network, and thus offers a new services and breed of applications and to end users with low latency, high bandwidth, and location-awareness, what’s more, hence gets the name as fog is comparably a cloud near the ground. Home automation systems are group of smart devices that permit different functions within a house or building, such as light and heating, energy monitoring, air conditioning, plug control and temperature metering, etc. Typically, these devices are smart sensors that are applied with low power communication protocol like ZigBee only can be controlled from an Internet gateway. These days, there are number of home automation products on the market for customers. Client can utilize application on smart phone to control the items they purchased. The control command can go through a cloud-based server and be directed to the corresponding gateway, and internally reach to the sensor devices, which is referred to as “Cloud-based mode” system.
Full Paper

IJCST/94/1/A-0929
12

Sentiment Analysis on Myocardial Infarction Using Tweets Data

Dr. R. Hemalatha, M.B.Monicka

Abstract

In 2016, the survey reports that 1.7 Million people die of Myocardial Infarction (MI), due to less medication facilities, less prevention care and treatment planning is top most analysis of effective disease risk assessment, through this we have take prevention using sentiment analysis of recent advancements, the text analytics have opened up new potential of using the rich information of tweet analysis, to identify the relevant risk factors in MI. To tackle the MI risk factors tweet analysis gives more remedy and care factors by users, also this leads to decrease of MI in India. Our system plays a machine learning approach using sentiment analysis using tweet dataset. Nowadays people suffering from MI such as cardiac arrest, high blood pressure, congestive heart failure etc. Twitter is an excellent resource for the MI Patients since they connect people who have with similar conditions and experiences. It provides the knowledge sharing about MI, plays a vital role through Opinion Mining system.
Full Paper

IJCST/94/1/A-0930
13

Review Paper on Low Power Design for Power and area Efficient Exploitation of Carry Select Adder

Piyush Choudhary, Heena Gadwanshi

Abstract

An adder is a combinational digital circuit which accomplishes accumulation of number. In numerouscomputers and additional processors adders are used in arithmetic logic units and additional parts of processors where they are used to compute addresses, increment and decrement operators, In circumstance where two’s complement or one’s complement is presence used to signify negative number. Adders are the essential units in numerous electronic circuits particularly in circuits used for performancearithmetic operations. An optimized is required to avoid whichever degradation at the circuit level in the output voltage, less delay in seriouspath. In this paper, we study and analysis the number of research paper and determine the performances circuit have been compared to earlier designed XNOR gate delay, power indulgence and Power Delay Product (PDP) minor Power-Delay product, and area saving due to lower transistor counts and different structures.
Full Paper

IJCST/94/1/A-0931
14

Golden Batch Identification Using Statistical Tools as a Part of Asset Performance Management

Modhuli D. Goswami

Abstract

This paper discusses use of data analysis using statistical methods for prediction of the Golden Batch in a sample industrial dataset (Wine manufacturing). The Golden Batch identification is extremely important as it allows us to reach, a plant process operating curve, which delivers the most optimal product mix,along with enhanced quality of final product. PCA(Principle Component Analysis) and PLS (Partial Least Square Regression) analysis are the tools used for analyzing the data set.These Data sets are the telemetered and archived process/plant parameters from a large number of such product runs. Principal Components Analysis (PCA) is used for dimension reduction and Partial Least Square Regression (PLS) is used for predictive analysis, RMSE is used to get the error between actual and predicted dataset. Business Managers, Process and control engineers can use this method to detect batch to batch error and recipe of the Golden batch which when implemented in other batches, results in optimal plant and product output, thus increasing efficiency of the manufacturing and increasing profits.
Full Paper

IJCST/94/1/A-0932
15

An Efficient Methodology Designed for Designing Travel Package

T. Paramesh, Dr. Molli Srinivasa

Abstract

Tourism is one of the major enlightenments behind the improvement and support of any areas. Its dependability helps in the development of financial estimation of any nation. Furthermore, in these ongoing years, recommendation system is ending up so mainstream among online applications, for example, internet business, etc. This paper attempts to diagram how recommendation system is valuable to tourism. A huge travel and vacationer dataset which incorporates travel packages, will be misused and in the wake of examining it, records will be isolated into gatherings according to the traveler, territory, season and in particular in the premise of appraisals. The customized rundown of vacationer package recommendation will be created and suggested based on community oriented evaluating. Moreover, the idle relationship among the whole vacationer will be investigated with the goal that package can be prescribed for comparative visitors in a gathering. This methodology can be utilized both by the travel offices and the travel bunches at low upkeep and cost. Thus, the cocktail approach can well speak to the substance of the service and the issues went ahead substation. In light of this cocktail approach is created for customized service recommendation by thinking of some as extra factors including the issues summon by ABT system, the services included and the initiate issue of new services.
Full Paper

IJCST/94/1/A-0933