Showing papers in "International Journal of Innovative Research in Computer and Communication Engineering in 2017"

A Survey on Machine Learning: Concept,Algorithms and Applications

Abstract: Over the past few decades, Machine Learning (ML) has evolved from the endeavour of few computer enthusiasts exploiting the possibility of computers learning to play games, and a part of Mathematics (Statistics) that seldom considered computational approaches, to an independent research discipline that has not only provided the necessary base for statistical-computational principles of learning procedures, but also has developedvarious algorithms that are regularly used for text interpretation, pattern recognition, and a many other commercial purposes and has led to a separate research interest in data mining to identify hidden regularities or irregularities in social data that growing by second. This paper focuses on explaining the concept and evolution of Machine Learning, some of the popular Machine Learning algorithms and try to compare three most popular algorithms based on some basic notions. Sentiment140 dataset was used and performance of each algorithm in terms of training time, prediction time and accuracy of prediction have been documented and compared.

Comparative Analysis of Clustering Techniques in Network Traffic Faults Classification

Abstract: Ubiquitous high-speed communication networks play a crucial role in the modern life, demanding the highest level of reliability and availability. Due to the rapid growth of computer networks in terms of size, complexity and heterogeneity, the probability of network faults increases. Manual network administration is hopelessly outdated; complex automated fault diagnosis and management are essential to ensure the provision and maintenance of high quality service in computer networks. Guaranteed Service with higher levels of reliability and availability for real-time applications can be achieved with a systematic approach for real-time classification of network faults, which helps in well-informed (often-automated) decision making. In this paper we discuss three different data mining algorithms as part of the proposed solution for network fault classification: K-Means, Fuzzy C Means, and Expectation Maximization. The proposed approach can help capture abnormal behavior in communication networks, thus paving the way for real-time fault classification and management. We used datasets obtained from a network with heavy and light traffic scenarios in the router and server and built a prototype to demonstrate the network traffic fault classification under given scenarios. Our empirical results reveal that the FCM is more accurate while causing computational overhead. The other two algorithms attain almost the same performance.

Career Counselling Using Data Mining

Abstract: Selecting an appropriate career is one of the most important decisions and with the increase in the number of career paths and opportunities, making this decision have become quite difficult for the students. According to the survey conducted by the Council of Scientific and Industrial Research’s (CSIR), about 40% of students are confused about their career options. This may lead to wrong career selection and then working in a field which was not meant for them, thus reducing the productivity of human resource. Therefore, it is quite important to take a right decision regarding the career at an appropriate age to prevent the consequences that results due to wrong career selection. This system is a web application that would help students studying in high schools to select a course for their career. The system would recommend the student, a career option based on their personality trait, interest and their capacity to take up the course

Comparative Analysis of Various Optimization Methodologies for WDM System using OptiSystem

Abstract: A wavelength division multiplexing (WDM) network are widely used in modern telecommunication infrastructure, which are expected to support a large variety of services with different requirements in terms of latency, bandwidth, reliability and many other features. The gain flatness of Erbium Doped Fiber Amplifier (EDFA) plays an important role in WDM network. The main drawback of EDFA is its unequal gain spectrum. The purpose of this project is to investigate different gain flatness techniques for EDFA in order to optimize the WDM performance, as well as to achieve a given bit error rate (BER), gain flatness and noise figure of EDFA through optimized fiber length and pump power. A commercial software, known as ‘Optisystem’ is used in this project for implementing the physical model of WDM system with different gain flatness techniques. At the end, the results obtained from the simulation of WDM systems implemented with different gain flatness techniques are compared with each other.

Node Transmission Power Value Optimization in MANET

Abstract: Mobile Ad-hoc Network (MANET) is normally recognized in zones someplace infrastructural services such as base station, routers etc. do not happen or have been damaged due to natural hardship. They have numerous pressures such as bandwidth, computational volume and battery power of each node as of their infrastructure-less nature. Power preservation is serious to appropriate actions of MANET. Countless scientists have been provided several mechanisms to diminish the power consumption variable transmission range of nodes are one such tool taken into interpretation. This paper studies the impact transmission range of routing protocols by designing a simulator in Qualnet. We note an obvious impact of variable transmission range on power consumption. All extra protocols are defined in terms of Packet Delivery Ratio (PDF), End-To-End Delay (ETED), average jitter rate, throughput etc.

Safeguard of security: firewalls

Abstract: This article is based on network security and security issues Day to day hackers and intruders are attacking on packets data with occurring disturbing traffic flood in source to destination way Many techniques and types are helping us to secure our data from attackers IP address, port number using in network security firewall for passing information on original server to clients This paper will introduce various methods of network security using firewalls The security that firewalls supply is only as superior as the strategy they are configured to realize

Challenges in Using Mobile Technology for Data Collection in Research Settings

Abstract: Data collection is one of the important components of public health systems and research settings The application of mobile technology to perform various tasks has increased in the recent past This increase in usefulness has come at the expense of the usability of these devices in some contexts The rapidly growing use of mobile technologies has also created a demand for mobile-based data collection solutions to bridge the information gaps in the health research in the developing countries Studies conducted on mobile usability models found that usability is usually measured in terms of three attributes; effectiveness, efficiency and satisfaction The three attributes are very important in ensuring accurate and timely data collection for public health systems The rapid growing use of mobile technologies has also led to a demand for mobile-based as a mode of data collection This study reviews the challenges that hinder the effective use of mobile technology such as internet network, infrastructure, installation and operation costs, and capacity of personnel and uptake of this new technology This review will show the factors influencing the implementation of mobile technologies, particularly open source applications in research data collection and reporting systems for the developing world It is important to look at the feasibility of mobile technologies, particularly open source technologies in improving the research data collection and reporting systems for the developing world This review paper defines the challenges of deploying mobile technology in a remote environment and then discusses how to addresses these challenges to provide an adaptable systems

A Robust Deep Learning Approachto Detect Nuclei in Histopathological Images

Abstract: Automated cell detection in histopathology images is challenging due to large variations in size, density, and batch variations. Nuclei detection provides useful information for evaluating cancer progression and prognosis. The performance of most classical nuclei detection methods relies on appropriate data selection. These methods require experts to create useful features. On the other hand, deep learning can extract feature sets from the data automatically, not requiring the design of feature extractors by experts. In this work, a new enhancement for deep learning approach is proposed to learn a continuous mapping from H&E image patches centered around nucleus centroids to nuclear distance maps. Our approach formulates the problem as a continuous regression problem and builds a fully convolutional regression network. In this method, it handles partial detections and irregular-shaped, neighboring nuclei, and different nuclei sizes and color. We train the network with the colorectal dataset which is publicly available dataset. The work is evaluated with the human bone marrow dataset without re-training and superior results are achieved.

A Proposal for Factors Influencing Biometrics Technology Security

Abstract: Biometric technology is being considered as a convenient and secure method of identification which eliminates the need to remember complex password, nor smart cards, keys and the like. This report is designed to help exploring main influencing factors and attitudes concerning this technology by informal interviews and web based survey specially in health care. This report will help in identifying recent security trends in potential organizations/institutions but also forecast its impact on future security concerns. After that the analysis will further help recommending a guide to boom its presence as emerging technology.

Network Intrusion Prevention System Using Machine Learning Techniques

Abstract: Secured data communication over networks is always under threat of intrusions and misuses. A Network Intrusion Prevention and Detection System (IPDS) is a valuable tool for the defense-in-depth of computer networks. Network IPDS look for known or potential malicious activities in network traffic and raise an alarm whenever a suspicious activity is detected. The Intrusion Detection Systems most commonly used in enterprise networks are signature-based, because they can efficiently detect known attacks while generating a relatively low number of false positives. Anomaly-based detection systems usually produce a relatively higher number of false positives, compared to the misuse-based or signature-based detection systems because only a fraction of the anomalous traffic is derived from intrusion attempts. As a matter of fact, it has been shown that the false positive rate is the true limiting factor for the performance of IDS, and that in order to substantially increase the Bayesian detection rate, P (Intrusion |Alarm), the IDS must have a very low false positive rate. One-class classification algorithms pursue concept learning in absence of counter examples, and have been shown to be promising for network anomaly detection. This project aims to use one-class classifier that is One-Class Support Vector Machines to detect network attacks that bear form of port-scan attacks for very low false positive rates.

Usability and Accessibility Evaluation of Windows 10

Abstract: The purpose of our study is to evaluate, assess and appraise the usability of Windows 10 operating system in light of its usability by users of variable diversity in origin, genre and intellectual standards. Obviously, every version of Windows operating system has different features which can be either easy or sometimes difficult for the user to understand and perceive. The difficulty is often triggered because of Human-Computer Interaction mistakes, brought about not only by the design but often due to the introduction of a new mental model. Hidden functions and modified features of updated versions contribute significantly to the predicaments of the users in performing basic tasks. The main emphasis of our study is to collect maximum specifics in detail, highlighting the influence of the transition from Windows 8 to Windows 10, on the users. Consequently, this data can then be utilized by the developers to evaluate the extent of the problems and later analysis in resolving them. Our study can play a major role in ensuring prompt satisfaction of the users, achieving good user experience and opinion. [1,2]

Emotional Recognition Using Facial Expression by Emoji in Real Time

Abstract: Present research focuses on the role of emoji’s in facilitating emotional recognition with the help of pictorial depictions of facial expressions. Today is the era of fast and dynamic internet and communication technologies. Hence, the communication is convenient as compared to the past. Use of communications through different channels, such as mobile phones and computers, are very common in today’s era. E-mails, text messaging, blog entries, and comments are some of the forms of communication which are very common today. To enhance the experience of communication, emojis were developed by the Japanese mobile companies such as Vodafone. Emoji’s are the pictorial depiction of the facial expression of human beings. They are very helpful in the facilitation of human emotional experiences. This research thesis investigates emotional recognition using facial expression by emoji in real time. Moreover, it also develops the parameters of measuring the facial expression and understanding the facial emotion recognition in real time. The application developed includes six human expressions, which include neutral, fear, anger, happy, sad, and surprise emotions. These expressions are the actual expressions which are being conveyed in human beings. The investigations of such expression are important because of their ability to better express human emotions and the way they facilitate communications among the people. Recommendations on further research will be provided for researchers.

Semantic-Based Video Retrieval Survey

Abstract: There is a tremendous growth of digital data due to the stunning progress of digital devices which facilitates capturing them. Digital data include image, text, and video. Video represents a rich source of information. So, there is an urgent need to retrieve, organize, and automate videos. Video retrieval is a vital process in multimedia applications such as video search engines, digital museums, video-on-demand broadcasting. In this paper, the different approaches of video retrieval are clearly and briefly categorized. Moreover, the different methods which try to bridge the semantic gap in video retrieval are discussed in more details.

Multi-hop Mobile Wireless Mesh Network Testbed Development and Measurements

Abstract: Mesh networks are envisioned to play a major role in the future of Smart-Cities and Internet of Things technologies, and can be a great ally in executing mission-centric tasks like search and rescue, and assist first responders. In this paper, we demonstrate the feasibility of development of a mobile mesh network testbed with the help of commercial-off-the-shelf (COTS) components and perform measurement tests in an indoor environment. We used open-source software to enable the testbed, such as Linux as the operating system, and Optimized Link State Routing (OLSR) protocol to enable the wireless mesh network. The physical components include open-source hardware like Raspberry Pi computers, and mobility is enabled through the use of a mobile robot Pioneer 3. We evaluate the network in various scenarios, first establishing benchmark measurement in a stationary setup, and then comparing with a mobile variation.

Alleviate HOL Blocking and DoS Attack in IMS Based NGN

Abstract: IMS based NGN uses TCP or UDP as transport layer protocol to carry multimedia data across different networks and even to carry signal among core components such as Call/Session Control Functions (CSCF). Due to using of TCP or UDP, IMS based NGN is facing TCP’s SYN Denial of Service (DoS) attack, Head of Line (HOL) blocking crisis. To overcome these troubles the SCTP with its features can be used in IMS based NGN.

CBIR Systems: Results of Feature Extraction with Color Feature Comparison with Standard Database

Abstract: Image retrieval is a distinguished field in digital image processing. Images can be extracted from a big collection of images on the basis of text, color and structure. In systems, using features which users are connected to like color get many similar and relevant images. In a typical Content based image retrieval (CBIR) system, the visual content of the images in the database are extracted and displayed by multi-dimensional feature vectors. The feature vector of the images in the database is a feature database. Most sought out systems represent images with color feature most related to user and shape feature to relate image to exact object. In this paper we present introduction to CBIR systems, research work on Color scheme with HSV color space making use of color histogram. A comparison with Corel dataset and evaluation with parameters precision, recall and f-measure are computed.

Question Analysis and Classification for Question Answering System

Abstract: Question answering(Q/A) are part of information retrieval (IR) research in which users instead of providing a few keywords to retrieve a set of documents, actually provide complete questions and expect from the QA system to get the most relevant answers(s). This way an efficient Q/A system will be more helpful in providing an accurate answer to a question instead of a query based information retrieval system. No one could have predicted that question answering system would become an indispensable technology for Information Retrieval that would enable the creation of new technologies for information retrieval. The Q/A systems have now been recognized as one of the challenging problem in IR. The task of question classification is one of the most important steps for the efficiency and relevancy of any QA system design. After going through the available literature on the QA systems, we found that there could be different classification of questions. This paper first discusses the general architecture of a QA system and then proposes a more elaborative classification of questions along with some major approaches used for classification in order to design a suitable question answering (QA) system.

ACO Method Base Data Flow Quality Testing Approach for Component Based Software Development

Abstract: In light of programming designing segments (CBSE) it has been centred on the advance related to arranging and sending programming segments outlines gathered from programming parts. Quality affirmation (QA) for CBSE is another theme in the product advancement Exploration Segment. In this report, an enhanced stream of information a model based QA is introduced for CBSE utilizing Subterranean insect settlement advancement calculation (ACO) to streamline the given code for programmed era and prioritization of the ideal way stream Control Chart (CFG) Basic leadership Process, the consequence of an enhanced test stage for the QA demonstrate with less multifaceted nature. So the approach proposed on the COA premise is additionally utilized for create test information to meet the produced way gatherings. This archive is a further proposed approach connected a program module. The outcomes demonstrate that a superior test is performed by applying a diagram in view of the ACO-based parts proposed programming. The proposed approach gives finish programming scope not as much as repetition.

Cognitive Radio Spectrum Sensing andSecondary User Allocation

Abstract: The increased demand of services and applications on the wireless communication networks in the future requires wider ranges of spectrum. This enforces the designers to develop new techniques and solutions to use the scarce spectrum efficiently. One of the proposed solutions is moving to new higher frequencies, but it is observed that the licensed spectrum is not completely used where some parts are unused and others are not utilized. Cognitive radio is an intelligent technique used in wireless networks to take advantage of unused or underutilized spectrum. This may be performed by allocating secondary users in the spectrum holes in order to enhance the efficiency of radio resource allocations. In this paper; spectrum sensing methods and secondary user allocation are performed to utilize the spectrum holes. The simulation Model is carried out using Matlab simulation program in order to study the performance of the proposed cognitive radio system.

Life Cycle of Source Program - Compiler Design

Abstract: This Research paper gives brief information on how the source program gets evaluated and from which sections source code has to pass and parse in order to generate target code or predicted output In addition to that, this paper also explains the concept of Pre-processors, Translators, Linkers and Loaders and procedure to generate target code This paper concentrates on Concept of Compiler and Phases of Compiler

Intelligent Detection for Cyber Phishing Attacks using Fuzzy rule-Based Systems

Abstract: Cyber phishing attacks are increasing rapidly, causing the world economy monetary losses. Although various phishing detections have been proposed to prevent phishing, there is still a lack of accuracy such as false positives and false negatives causing inadequacy in online transactions. This study constructs a fuzzy rule model utilizing combined features based on a fuzzy inference system to tackle the foreseen inaccuracy in online transactions. The importance of the intelligent detection of cyber phishing is to discriminate emerging phishing websites with a higher accuracy. The experimental results achieved an excellent accuracy compared to the reported results in the field, which demonstrates the effectiveness of the fuzzy rule model and the feature-set. The findings indicate that the new approach can be used to discriminate between phishing and legitimate websites. This paper contributes by constructing a fuzzy rule model using a combined effective feature-set that has shown an excellent performance. Phishing deceptions evolve rapidly and should therefore be updated regularly to keep ahead with the changes.

Protected Data Reserve over Cloud

Abstract: Cloud-based outsourced storage relieves the client's burden for storage management and maintenance by providing a comparably low-cost, scalable, location-independent platform. However, the fact that clients no longer have physical possession of data indicates that they are facing a potentially formidable risk for missing or corrupted data. To avoid the security risks, audit services are critical to ensure the integrity and availability of outsourced data and to achieve digital forensics and credibility on cloud computing. Provable data possession (PDP), which is a cryptographic technique for verifying the integrity of data without retrieving it at an un-trusted server, can be used to realize audit services. Cloud computing has been emerged solution to the rising storage costs of IT industry With the high costs of data storage devices as well as the rapid rate at which the data begins generated it proves costly for enterprises or individual users to frequently update their hardware Apart from reduction in storage cost the user’s data to large data centres, which are remotely located, on which user does not have any control.

A Survey on Dynamic and Public Auditing with Fair Arbitration for Cloud Data

Abstract: Cloud clients no more extended physically have their information, so how to guarantee the trustworthiness of their outsourced information turns into a testing assignment. As of late proposed plans, for example, "provable information ownership" and "verifications of retrievability" are intended to address this issue, however they are intended to review static document information and thusly absence of information elements bolster. Besides, danger models in these plans more often than not accept a fair information proprietor and concentrate on recognizing an untrustworthy cloud specialist organization in spite of the way that customers may likewise make trouble. This paper proposes an open evaluating plan with information progression support and decency mediation of potential debate. Specifically, we outline a list switcher to kill the constraint of list utilization in label calculation in current plots and accomplish proficient treatment of information progression. To address the decency issue so that no gathering can get out of hand without being distinguished, we additionally amplify existing danger models and receive signature trade thought to configuration reasonable mediation conventions, so that any conceivable question can be genuinely settled. The security examination demonstrates our plan is provably secure, and the execution assessment exhibits the overhead of information elements and question discretion are sensible.

Performance Analysis of Machine Learning Techniques to Predict Mental Health Disorders in Children

Abstract: Mental disorders are quite common in children. The commonly found childhood mental disorders are anxiety disorders; depression and attention deficit disorder. Diagnosis of these problems at early stage helps the professionals in treating it at beginning stage and to improve the patient’s health. Therefore the need to treat common mental health disorders that are found in children which lead to complicated problems, if ignored at early stage. Machine learning Techniques can be applied for analyzing patient’s history to diagnose the problem. In this research three machine learning techniques have been identified and compared based on their performances on several scales of accuracy on selected attributes to diagnose five basic mental health disorders. The basic aim is to find the technique which is most accurate. The data set is containing sixty attributes for analyzing and measuring the performance of techniques. Ignoring the irrelevant attributes that do not have much effect, twenty-five attributes were found as important to diagnose the disorder. Applying Feature Selection algorithms on the attribute set, thirteen attributes were found. Accuracy of the selected attribute set on three machine learning techniques were compared viz., Multilayer Perception, LAD Tree and Multiclass Classifier. It is clear by the results that the Multiclass classifier produces much accurate results on set of selected attributes.

A Reengineering Process of Online Registration Services

Abstract: In the past few years, in almost every semester, the University of Baghdad (UOB) has accepted thousands of students. All these years the registration procedures have been carried out through computerized systems, in a way students waste time through shifting the schedule book to decide which courses to register. Almost all students want to be registered virtually at the same time, which creates chaos. In order to overcome such matters, extra staff members are assigned to help in reducing the workload and time during the registration period. This solution still causes other problems such as higher costs for the additional expenses, wages, and time consumption for both students and employees. The project attempts to analyze, evaluate and reengineer the current e-Registration system at the University of Baghdad. This effort aims to identify weaknesses and difficulties in order to suggest and recommend solutions to improve the current process and strategically add services to the university’s competitive advantage. In this study efforts were made to apply the business process reengineering practice (BPR) to this system. This process is used to automate many of the manual registration tasks such as admissions, early registrations, addition of chair process and the payment system to resolve some major problems that were faced by students during the registration process.

Color Images Encryption using Cipher System with Different Types of Random Number Generator

Abstract: Image encryption is a wide science in a nowadays and is used in research of information security, and a lot of image encryption algorithms have been introduced to protect the personal images from unauthorized access. The proposal algorithm generated random password seed which is used as a key to three types of Pseudo Random Number Generator; Linear Feed Back Shift Register, Non-Linear Feed Back Shift Register and BLUM BLUM SHUB. The algorithms are applied to the colour images with large random keys. The experimental results of comparisons between these algorithms, shows that BBS have better performance than LFBSR and NLFBSR.

An Affordable Hand-glove for the Blind using Ultrasonic Sensors

Abstract: This paper presents the implementation of an assistive glove for the blind. The basis of this glove is a technology that helps the visually impaired in detecting any obstacles that may appear in their path within 100 cm in any direction of the glove. When the user of this glove encounters an object within 100 cm, the glove alerts the user with a loud beeping sound and heavy vibrations. One of the key USPs of this glove is its extremely low manufacturing cost as against the technologies used in other gloves which cost almost ten times of what the glove presented in this paper does. The technology used in this glove involves heavy communications between the Arduino UNO and Ultrasonic Sensors like the HC-SR04. This glove can easily be marketed as a social impact project considering its ease of use and a very competitive price.

Personality Based E-Recruitment System

Abstract: Nodes in Mobile Ad Hoc Networks (MANETs) are limited battery powered. That’s why energy efficient routing has become an important optimization criterion in MANETs. The conventional routing protocols do not consider energy of the nodes while selecting routes which leads to early exhaustion of nodes and partitioning of the network. This paper attempts to provide an energy aware routing algorithm. The proposed algorithm finds the transmission energy between the nodes relative to the distance and the performance of the algorithm is analyzed between two metrics Total Transmission energy of a route and Maximum Number of Hops. The proposed algorithm shows efficient energy utilization and increased network lifetime with total transmission energy metric.

Profiling REST Services with Advent of Aspect Oriented Modelling

Abstract: REST, a communication protocol is used for communicating via APIs. It is widely used today for intercommunication among the services and the users. However, it does have the constraint of constructing servers to be stateless. Stateless servers are restricted from storing any information about the clients. Hence profiling such resources becomes a serious issue, as session is not retained. AOP enables implementing cross cutting concerns to be applied to applications without disturbing their business logic. This paper aims to overcome this challenge of resources profiling using REST and AOP concepts. It presents a plug and play approach for profiling REST resources using Aspect Oriented Programming.

A Probabilistic Black hole & gray hole attacks Detection Scheme towards Efficient Trust Establishment in Delay-tolerant Networks-Review

Abstract: Malicious and selfish nature displays a serious threat against routing in Delay/Disruption Tolerant Networks (DTNs). Due to the unique characteristics of network, designing a misbehavior detection scheme in DTN is regarded as a great research. In this paper, we propose iTrust, misbehavior detection scheme that uses a probability, for secure DTN routing towards efficient trust establishment. The basic idea of iTrust is introducing a regularly available Trusted Authority (TA) to judge the node’s behavior based on the routing evidences collected and probabilistically checking.