scispace - formally typeset
Search or ask a question

Showing papers in "International Journal of Computer Applications in 2020"


Journal ArticleDOI
TL;DR: This paper has performed handwritten digit recognition with the help of MNIST datasets using Support Vector Machines (SVM), Multi-Layer Perceptron (MLP) and Convolution Neural Network (CNN) models.
Abstract: The reliance of humans over machines has never been so high such that from object classification in photographs to adding sound to silent movies everything can be performed with the help of deep learning and machine learning algorithms. Likewise, Handwritten text recognition is one of the significant areas of research and development with a streaming number of possibilities that could be attained. Handwriting recognition (HWR), also known as Handwritten Text Recognition (HTR), is the ability of a computer to receive and interpret intelligible handwritten input from sources such as paper documents, photographs, touch-screens and other devices [1]. Apparently, in this paper, we have performed handwritten digit recognition with the help of MNIST datasets using Support Vector Machines (SVM), Multi-Layer Perceptron (MLP) and Convolution Neural Network (CNN) models. Our main objective is to compare the accuracy of the models stated above along with their execution time to get the best possible model for digit recognition.

26 citations


Journal ArticleDOI
TL;DR: This study will use data exploratory and mining techniques to extract hidden patterns using python to seek better performance in predicting heart diseases to reduce the number of tests require for the diagnosis of heart diseases.
Abstract: Heart disease, an example of cardiovascular diseases is the number one notable reason for the death of many people in the world. Of recent, studies have concentrated on using alternative efficient techniques such as data mining and machine learning in the diagnosis of diseases based on certain features of an individual. This study will use data exploratory and mining techniques to extract hidden patterns using python. By this, machine learning algorithms (logistic linear regression, decision tree classifier, Gaussian Naïve Bayes models) will be developed to predict the presence of heart diseases in patients. This will try to seek better performance in predicting heart diseases to reduce the number of tests require for the diagnosis of heart diseases. The k-fold cross validation approach will be used in assessing the resulting models for receiver operating characteristic (ROC) curves (sensitivity against specificity). The dataset was collected from UCI machine learning repository which contains information on patients with heart disease. The dataset has 14 attributes and measured on 303 individuals. General Terms Algorithms, pattern recognition, supervised learning, machine learning, heart disease.

20 citations



Journal ArticleDOI
TL;DR: A research work has been done on Bengali Sports news comments published in different newspapers to train a deep learning model that will be able to categorize a comment according to its sentiment.
Abstract: Identifying and categorizing opinions in a sentence is the most prominent branch of natural language processing. It deals with the text classification to determine the intention of the author of the text. The intention can be for the presentation of happiness, sadness, patriotism, disgust, advice, etc. Most of the research work on opinion or sentiment analysis is in the English language. Bengali corpus is increasing day by day. A large number of online News portals publish their articles in Bengali language and a few News portals have the comment section that allows expressing the opinion of people. Here a research work has been done on Bengali Sports news comments published in different newspapers to train a deep learning model that will be able to categorize a comment according to its sentiment. Comments are collected and separated based on immanent sentiment. The deep learning algorithms that have been used are Convolutional Neural Network (CNN), Multilayer Perceptron, Long Short-Term Memory (LSTM). General Terms Sentiment Analysis, Deep Learning, Emotion Classification

13 citations


Journal ArticleDOI
TL;DR: It is found that the k-nearest neighbor (kNN) machine learning algorithm exhibits excellent accuracy in detecting malware and also reviews different tools for ransomware detection, classification and analysis.
Abstract: Internet of Things (IoT) is being considered as the growth engine for industrial revolution 4.0. The combination of IoT, cloud computing and healthcare can contribute in ensuring well-being of people. One important challenge of IoT network is maintaining privacy and to overcome security threats. This paper provides a systematic review of the security aspects of IoT. Firstly, the application of IoT in industrial and medical service scenarios are described, and the security threats are discussed for the different layers of IoT healthcare architecture. Secondly, different types of existing malware including spyware, viruses, worms, keyloggers, and trojan horses are described in the context of IoT. Thirdly, some of the recent malware attacks such as Mirai, echobot and reaper are discussed. Next, a comparative discussion is presented on the effectiveness of different machine learning algorithms in mitigating the security threats. It is found that the k-nearest neighbor (kNN) machine learning algorithm exhibits excellent accuracy in detecting malware. This paper also reviews different tools for ransomware detection, classification and analysis. Finally, a discussion is presented on the existing security issues, open challenges and possible future scopes in ensuring IoT security.

12 citations


Journal ArticleDOI
TL;DR: Results of performance evaluation of the developed system shows that it has very low level of distortion as revealed by the Signal to Noise Ratio (SNR) and the compression ratio obtained is also equal to one (1), which shows that the cover audio file is identical to the resultant stego file.
Abstract: Hiding text in a digital audio file format has been a major challenge because of the Human Auditory System (HAS) and how the digital audio will be converted into analog form for text to be embedded into it. Several techniques that include but are not limited to Least Significant Bit, Echo Hiding, Phase Coding, Parity Coding and so on have been proposed in both research communities and the academia. Audio steganography hides data in selected audio files. Several audio steganography works exist, but their major limitations include their inability to embed information in multiple audio file formats, high distortion rate and low level of robustness of their resultant stego files. This research attempts to proffer solution to the obvious challenges of the previous works by developing an efficient and robust audio steganography system for the security of information whether in store or on transit across the Internet. Results of performance evaluation of the developed system shows that it has very low level of distortion as revealed by the Signal to Noise Ratio (SNR). The compression ratio obtained is also equal to one (1), which shows that the cover audio file is identical to the resultant stego file.

12 citations


Journal ArticleDOI
TL;DR: The researchers classify those incidents against the Cybersecurity vulnerabilities in Blockchain technology and explain the methods of risk measures according to the Information Security Risk Assessment (ISRA) Models.
Abstract: Blockchain technology has become a paradigm shift to digital transactions. It has brought massive potentials in many fields, such as financial services, energy, healthcare and Internet of Things. As often occurs with innovative technologies, it has suffered from several critical Cybersecurity threats and vulnerabilities. The complicated relation between Cybersecurity risk management and companies strategic and operational objectives which make identifying, analyzing, and controlling the relevant risk events as a major challenge. In this paper, the researchers classify those incidents against the Cybersecurity vulnerabilities in Blockchain technology and explain the methods of risk measures according to the Information Security Risk Assessment (ISRA) Models.

11 citations



Journal ArticleDOI
TL;DR: In this article, the authors use penetration testing to assess vulnerabilities and conduct attacks on Wireless Equivalent Privacy (WEP), Wi-Fi Protected Access (WPA) and 802.11i (Wi-Fi2) security protocols.
Abstract: The use of wireless network as a medium of communication has tremendously increased due to its flexibility, mobility and easy accessibility. Its usage is inevitable at hotels and restaurants, airports, organizations and currently predominant in homes. As large number of devices connect to wireless network, valuable and sensitive information are shared among users in the open air, attackers can easily sniff and capture data packets. This paper aims at using penetration testing to assess vulnerabilities and conduct attacks on Wireless Equivalent Privacy (WEP), Wi-Fi Protected Access (WPA) and 802.11i (WPA2) security protocols. The penetration testing was conducted using Kali Linux with its Aircrack-ng tools.

10 citations


Journal ArticleDOI
TL;DR: The results suggest that awareness of social engineering is a positive predictor of security-protective practices above and beyond the predictability power of possessing information security knowledge.
Abstract: Social engineering has become one of the biggest security threats facing organizations. Rather than relying upon information security technical-related shortcomings to break into computer networks, social engineers make use of employees’ individual and organizational traits to deceive them. In such a scenario, it is crucial for organizations to ensure that their employees not only possess sound knowledge about information security but also about the concept of social engineering and threats emerging from social engineering attacks. This study aims to test whether awareness of social engineering can predict and explain individuals’ securityprotective practices. We conducted a survey of 265 employees working in different organizations in Saudi Arabia. The results suggest that awareness of social engineering is a positive predictor of security-protective practices above and beyond the predictability power of possessing information security knowledge. Thus, to reduce the probability of potential consequences of social engineering attacks, our study suggests that organizations should not only strive to enhance employees’ security knowledge but should also invest in increasing employees’ awareness of social engineering.

9 citations


Journal ArticleDOI
TL;DR: This paper is proposing the system with the help of face recognition to develop the automatic door lock and unlock system that keeps the owner informed in the real time about the unknown person at the door of the home.
Abstract: In the today’s world the security of the home is the prime concern. The traditional methods of securing our home are very easy to break and lead to theft. To protect the home, we need to install the costly security system. To overcome this problem, we are presenting IoT based solution where we can setup a smart home security system. In this paper we are proposing the system with the help of face recognition to develop the automatic door lock and unlock system. It also gives us the facility to monitor our home remotely and take appropriate action if anything goes wrong. The Pi camera will be attached to the Raspberry pi accompanied with Passive Infrared and other sensors. Camera captures an image of the person in front of the door, then real-time face recognition is done using local binary pattern (LBP) [3]. If person’s image matches with one of the home members or identified person then the door will unlock otherwise won’t. Email containing image of the unknown person will be sent to the homeowner to his Gmail Id. The proposed system keeps the owner informed in the real time about the unknown person at the door of the home.This information will help the user to take necessary actions.

Journal ArticleDOI
TL;DR: Comparison of Windows, Unix, Linux, Mac, Android and iOS operating systems based on the OS features and their strengths and weaknesses to provide some guides to both end-users and developers guiding them in taking decisions about operating systems that are most suitable for them.
Abstract: Varieties of operating systems (OS) have emerged over the years having different features and functionalities. Understanding the functionalities of each OS guides users’ decisions about the OS to install on their computers. In view of this, the comparative analysis of different OS is needed to provide details on the similarities and difference in recent types of OS vis-à-vis their strengths and weaknesses. This paper focus on the comparative analysis of Windows, Unix, Linux, Mac, Android and iOS operating systems based on the OS features and their strengths and weaknesses. A qualitative analysis of six different operating systems and result showed that Windows 10 had 0.04 malware file present while Windows 7 machine was 0.08. Higher percentage of mobile malware target Androids than iOS. Windows 10, Linux, UNIX and Mac OS are more secured and reliable. Windows and Android are more popular, user-friendly, easy to use and allow more application program than Mac OS. Linux and Android are free while Windows is moderately costly and Mac OS is very costly. Except for Mac and iOS others allow compatibility. Windows 10 and Mac OS integrated firewall. Windows and Android tend to be the most widely used especially the newest versions. It is because they are affordable, secure, reliable, compatible and user friendly. This study helps to provide some guides to both end-users and developers guiding them in taking decisions about operating systems that are most suitable for them.


Journal ArticleDOI
TL;DR: The fundamental concepts of game theory are streamlined, an overview on the applications of game theoretical concepts in various microgrid optimization problems are presented, and some future opportunities that are expected to solve some of the technical challenges facing micro-grid technology are introduced.
Abstract: The technology of Smart Grid is believed to be the future of power system networks. Smart Grid (SG) gains its importance due to its proven ability to improve stability, efficiency and robustness of electrical power grids. SG system consists mainly of two components which are electrical distribution system and communication layer. In the electrical distribution system, the generated energy comes from a network of distributed energy resources \"DERs\", which is called microgrid. In most cases, these DERs are recommended to be renewable energy sources \"RESs\" to reduce emissions and harmful environmental effects. One of the main drawbacks of renewable energy sources is that their availability varies with time and so that the micro-grid technology faces various technical challenges which motivate many researchers to adopt techniques to overcome these challenges. In this regard and due to its capability of studying complex interactions between independent rational players, game theory is expected to have a great contribution in the phase of design and analysis of micro-grids. In this paper, the fundamental concepts of game theory are streamlined, an overview on the applications of game theoretical concepts in various microgrid optimization problems are presented, a novel classification of research points covered by researchers are provided. Finally, some future opportunities that are expected to solve some of the technical challenges facing micro-grid technology are introduced.


Journal ArticleDOI
TL;DR: Bangla news which has been collected from newspapers and gathered around to make a Bengali Corpus is collected and classified using baseline and deep learning models of Machine Learning.
Abstract: Today’s universe is the type of world where everyone thrives to live in virtual life. According to the perspective of the present time, the online news portal holds a major door to that gradually increasing greedy life. So around the globe, the various platform has been developed to fulfill the requirement of mankind. A heavy load of work has been carried out for making this platform autonomous in the English language. That’s why the machine learning approach is quite a fully developed field in English in news classification. But it can't be said the same for Bangla language. These put in the inspiration to do a research on this topic. So, here Bangla news which has been collected from newspapers and gathered around to make a Bengali Corpus. After preprocessing the news text, different sorts of procedures to classify the news text using baseline and deep learning models of Machine Learning are applied.

Journal ArticleDOI
TL;DR: The concepts of soft set theory and its relations are presented from trapezoidal to octagonal symptoms symbolically and Sanchez’s Approach in terms of Octagonal fuzzy number is used.
Abstract: This article presents the concepts of soft set theory and its relations. It the common observation that more symptoms give more accuracy to calculate the disease, so we enhanced the work (Jafar at el., 2019) from trapezoidal to octagonal symptoms symbolically. So, we have used Sanchez’s Approach in terms of Octagonal fuzzy number. In the end for the better understanding the theory an elaborative example using hypothetical data has also been presented.



Journal ArticleDOI
TL;DR: This paper is intended to determine the tangent, cosine and cotangent similarity measure for single valued Neutrosophic sets and will compare the accuracy of all above similarity measures and applied it in decision making problems such as selection of an academic programs.
Abstract: Similarity measures have wide range of applications in realworld such as patterns, face recognitions, codding etc. In this paper it is intended to determine the tangent, cosine and cotangent similarity measure for single valued Neutrosophic sets and will compare the accuracy of all above similarity measures and applied it in decision making problems such as selection of an academic programs.


Journal ArticleDOI
TL;DR: This study creates an advanced taxonomy of social engineering attacks with the aim of facilitating the development and implementation of better prevention measures, stressing the importance of organizational awareness.
Abstract: Rapid technological advancement has not only resulted in a change in the pace of economic development, but also led to increase in cyber-threats. A social engineering attack is one such threat where an attacker not only accesses critical information about a user through technology, but also through manipulation. Although the types of attacks are different i.e. social, physical, technical or socio-technical, the process is the same. This study creates an advanced taxonomy of social engineering attacks with the aim of facilitating the development and implementation of better prevention measures, stressing the importance of organizational awareness.

Journal ArticleDOI
TL;DR: The Smart water level management system uses an ultrasonic sensor to detect the water level and it is based on the sound made from flow of water it calculates the level of water in percentage and returns the value to the LCD display.
Abstract: In everyday life, million liters of water are getting wasted by overflowing and also heavy usage. To reduce this ,we have to implement some overflow control techniques to minimize the heavy wastage of water. For this purpose, we propose the Smart water level management system. This system uses an ultrasonic sensor to detect the water level and it is based on the sound made from flow of water it calculates the level of water in percentage and returns the value to the LCD display. It will calculate the level of water up to 100% by the intervals of 10%. After each interval it reaches the value is displayed on the LCD screen. This system is connected with relay switch which will automatically turn OFF and ON based on level of water. When the water level reaches 0% it will automatically turned ON and if it reaches 100% it will automatically turned OFF. We can also control the system online using a WiFi module which connects the system with web application we created. By also using PH sensor we can check the purity of the water. The result of this will be displayed on our web application.

Journal ArticleDOI
TL;DR: This article surveys denial-of-service (DoS) attacks that occur in the network layer of IoT systems and the impact on various aspects, along with several distinct DoS attack mitigation methods.
Abstract: The internet of things (IoT) has been gaining attention in the past decade, and this rapid growth is due to the many different advantages delivered towards achieving a smart world. However, security is one of the biggest challenges, as it builds upon the internet. This article surveys denial-of-service (DoS) attacks that occur in the network layer of IoT systems and the impact on various aspects. The Smurf and SYN flood attacks are briefly discussed along with several distinct DoS attack mitigation methods. Two DoS mitigation technologies implemented by IoT security companies are discussed as a case study.

Journal ArticleDOI
TL;DR: This work proposes a simple convolutional neural network model trained from scratch for discriminating benign and malignant breast cancer tumors in histopathological images and explores how optimizers aid in finding good sets of parameters that help minimize loss and increase overall classification accuracy.
Abstract: Conventional approaches to breast cancer diagnosis are associated with drawbacks that ultimately affect the quality of diagnosis and subsequent treatment, pushing for the need for automatic and precise classification of breast cancer tumors. The advent of deep learning methods has witnessed an increasing interest in their applications in many tasks. The specific case of using convolutional neural networks with transfer learning has witnessed tremendous successes in many classification tasks. Nonetheless, with transfer learning, the sheer number of parameters associated with deep networks coupled with the distance disparity between source data and target data leave networks prone to overfitting, particularly in the case of limited data. Also, negative transfer may occur in the situation where the source and target domains are not related. This work proposes a simple convolutional neural network model trained from scratch for discriminating benign and malignant breast cancer tumors in histopathological images. Four deep learning optimization algorithms are leveraged and explored to ascertain how optimizers aid in finding good sets of parameters that help minimize loss and increase overall classification accuracy. By adopting a polynomial learning rate decay scheduling and implementing several

Journal ArticleDOI
TL;DR: This paper intends to propose a solution for this problem “Smart Bin” which will focus on resolving this problem and will alarm and notify the authorized person through a software when the garbage bin is about to fill.
Abstract: In this paper, we describe the formatting guidelines for IJCA Journal Submission Nowadays, waste management has become a major problem in life cycle. It includes the collection, transport, treatment and disposal of waste, together with monitoring and regulation of the waste management process. The significant cause of waste management is brisk growth in the rate of urbanization and thus there is a need of proper planning. To avoid all such harmful scenarios and maintain public cleanliness and health, we intend to propose a solution for this problem “Smart Bin” which will focus on resolving this problem. This process will alarm and notify the authorized person through a software when the garbage bin is about to fill. This system pivots around the overflowing bins and keeping the areas clean.

Journal ArticleDOI
TL;DR: An effective and accurate thyroid disease prediction model is developed using an ensemble of Bagging with J45 and ensemble ofBagging with SimpleCart to extract useful information and diagnose diseases.
Abstract: Accurate diagnose of diseases prior to their treatment is a challenging task for the modern research, therefore it becomes necessary and important to use modern computing techniques to design an efficient and accurate prediction systems. Thyroid is one of the most common diseases found in human body with many side effects the accuracy for thyroid diagnosis system may be greatly improved by considering an ensemble algorithm technique. In this paper, an effective and accurate thyroid disease prediction model is developed using an ensemble of Bagging with J45 and ensemble of Bagging with SimpleCart to extract useful information and diagnose diseases. The performances of the two ensemble model were compared with single classifiers. The Bagging ensemble algorithm for thyroid prediction system promises excellent overall accuracy of 99.66% while other single selected classifiers like Bagging and SimpleCART has accuracy of 99.55% and J48 with accuracy of 99.60%. General Terms Machine Learning.

Journal ArticleDOI
TL;DR: This research focused on temporal evolution, leading authors, most cited papers, leading journals, competitions and collaboration networks, using a bibliometric approach.
Abstract: Person identification based on eye movements is getting more and more attention, as it is anti-spoofing resistant and can be useful for continuous authentication. Therefore, it is noteworthy for researchers to know who and what is relevant in the field, including authors, journals, conferences, and institutions. This paper presents a comprehensive quantitative overview of the field of eye movement biometrics using a bibliometric approach. All data and analyses are based on documents written in English published between 2004 and 2019. Scopus was used to perform information retrieval. This research focused on temporal evolution, leading authors, most cited papers, leading journals, competitions and collaboration networks.

Journal ArticleDOI
TL;DR: Big data security analysis with the help of different techniques used in network intrusion detection system is introduced and results obtained after using NS-3 based svm classifier using KDD Cup 99 Dataset showed the accuracy of 99 percent.
Abstract: This paper introduces Big data security analysis with the help of different techniques used in network intrusion detection system. The topic of how big data affects any intrusion detection system being used and how huge volume of the dataset, its specialized features that are heterogeneous in nature and what will happen if big data is processed at real time. Different attacks and intrusion detection methods such as intrusion detection and prevention systems (IDPS), signature-based detection (SD) and anomaly-based detection (AD) has been done. Challenges faced by intrusion detection systems (IDS), how they can be prevented and how machine learning, data mining techniques could be used in any general intrusion detection-based system has also been discussed. Also, how all the problem faced by IDPS can be solved by network simulator named NS-3.0. Its objectives, advantages, comparison with other networks and limitation have also been to be discussed. The recommendation is also given to improve faults. Also, results obtained after using NS-3 based svm classifier using KDD Cup 99 Dataset showed the accuracy of 99 percent.

Journal ArticleDOI
TL;DR: Various elements related to neural network model such as dataset, findings, calculative metrics and results are embraced for effortless interpretation of tabular correlation research.
Abstract: Researchers have shown more interest in soft biometrics area to fill the commination gaps between humans and machines with the growth of real-world application has increased day to day life. Soft-biometric consists of age, gender, ethnicity, height, facial measurements and etc. This paper contains a detail discussion about the contribution of the researchers in the area of gender classification and age estimation using neural networking. Most of the work is done using Convolutional neural networks and auto encoders. Various elements related to neural network model such as dataset, findings, calculative metrics and results are embraced for effortless interpretation of tabular correlation research. Finally, the authors summarize germane tasks for future various research aspects.