Pub Date : 2023-01-31DOI: 10.15282/ijsecs.9.1.2023.1.0105
Manuel O. Diaz Jr.
There is now an increasing number of sentiment analysis software-as-a-service (SA-SaaS) offerings in the market. Approaches to sentiment analysis and their implementation as SA-SaaS vary, and there really is no sure way of knowing what SA-SaaS uses which approach. For potential users, SA-SaaS products are black boxes. Black boxes, however, can be evaluated using a set of standard input and a comparison of the output. Using a test data set drawn from human annotated samples in existing studies covering sentiment polarity of news headlines, this study compares the performance of selected popular and free (or at least free-to-try) SA-SaaS in terms of the accuracy, precision, recall and specificity of the sentiment classification using the black box testing methodology. SentiStrength, developed at the University of Wolverhampton in the UK, emerged as consistent performer across all metrics.
现在市场上有越来越多的情绪分析软件即服务(SA-SaaS)产品。情感分析的方法及其作为SA-SaaS的实现各不相同，并且确实无法确定哪种SA-SaaS使用哪种方法。对于潜在用户来说，SA-SaaS产品是黑盒。然而，黑盒可以使用一组标准输入和输出的比较来评估。使用现有研究中包含新闻标题情感极性的人类注释样本的测试数据集，本研究使用黑盒测试方法，比较了选定的流行和免费(或至少免费试用)SA-SaaS在情感分类的准确性、精密度、召回率和特异性方面的表现。英国伍尔弗汉普顿大学(University of Wolverhampton)开发的SentiStrength在所有指标上都表现一致。
{"title":"A Domain-Specific Evaluation of the Performance of Selected Web-based Sentiment Analysis Platforms","authors":"Manuel O. Diaz Jr.","doi":"10.15282/ijsecs.9.1.2023.1.0105","DOIUrl":"https://doi.org/10.15282/ijsecs.9.1.2023.1.0105","url":null,"abstract":"There is now an increasing number of sentiment analysis software-as-a-service (SA-SaaS) offerings in the market. Approaches to sentiment analysis and their implementation as SA-SaaS vary, and there really is no sure way of knowing what SA-SaaS uses which approach. For potential users, SA-SaaS products are black boxes. Black boxes, however, can be evaluated using a set of standard input and a comparison of the output. Using a test data set drawn from human annotated samples in existing studies covering sentiment polarity of news headlines, this study compares the performance of selected popular and free (or at least free-to-try) SA-SaaS in terms of the accuracy, precision, recall and specificity of the sentiment classification using the black box testing methodology. SentiStrength, developed at the University of Wolverhampton in the UK, emerged as consistent performer across all metrics.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"4 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84280381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-01-24DOI: 10.15282/ijsecs.9.1.2023.7.0111
O. M. Olanrewaju, Adebayo Abdulhafeez Abdulwasiu, Nuhu Abdulhafiz
Wireless networks are becoming increasingly popular. Mobile ad hoc networks are one category among the different types of wireless networks that transmit packets from the sender node to the receiver node without the use of abase station or infrastructure, as the nodes serve as both hosts and routers. These networks are referred to as mobile because they are movable. MANET is an ad-hoc network that can change positions at any time, and nodes can join or leave at any moment, making it vulnerable to attacks such as Blackhole. Existing solutions, in some ways, led to more memory space consumption, while others led to an overhead. This research proposes an Enhanced On-demand Distance Vector (AODV) routing protocol to prevent Blackhole attacks on MANETs using Diffie Hellman and Message Digest 5 (DHMD), implemented using Network Simulator 2 (NS2). The performance of the proposed protocol was evaluated using the following parameters: Packet Delivery Ratio, throughput, End to End (E2E)Delay, and routing overhead. It was concluded that DHMD has reduced network over head as it resulted to 23% while AODV resulted at 38%and memory consumption for DHMD gave 0.52ms compared to AODV that gave 0.81msdue to Blackhole prevention. This research will help to mitigate the effect of blackhole attacks in a network and increase network performance by reducing overhead and memory consumption.
{"title":"Enhanced On-demand Distance Vector Routing Protocol to prevent Blackhole Attack in MANET","authors":"O. M. Olanrewaju, Adebayo Abdulhafeez Abdulwasiu, Nuhu Abdulhafiz","doi":"10.15282/ijsecs.9.1.2023.7.0111","DOIUrl":"https://doi.org/10.15282/ijsecs.9.1.2023.7.0111","url":null,"abstract":"Wireless networks are becoming increasingly popular. Mobile ad hoc networks are one category among the different types of wireless networks that transmit packets from the sender node to the receiver node without the use of abase station or infrastructure, as the nodes serve as both hosts and routers. These networks are referred to as mobile because they are movable. MANET is an ad-hoc network that can change positions at any time, and nodes can join or leave at any moment, making it vulnerable to attacks such as Blackhole. Existing solutions, in some ways, led to more memory space consumption, while others led to an overhead. This research proposes an Enhanced On-demand Distance Vector (AODV) routing protocol to prevent Blackhole attacks on MANETs using Diffie Hellman and Message Digest 5 (DHMD), implemented using Network Simulator 2 (NS2). The performance of the proposed protocol was evaluated using the following parameters: Packet Delivery Ratio, throughput, End to End (E2E)Delay, and routing overhead. It was concluded that DHMD has reduced network over head as it resulted to 23% while AODV resulted at 38%and memory consumption for DHMD gave 0.52ms compared to AODV that gave 0.81msdue to Blackhole prevention. This research will help to mitigate the effect of blackhole attacks in a network and increase network performance by reducing overhead and memory consumption.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"40 3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86016290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.3.0100
Mohd Azuwan EfendyMail, Mohd Faizal Ab Razak, Munirah Ab. Rahman
Today's internet continues to move forward, and with it comes the development of many applications. Therefore, these applications are also directly accessible via the Internet, which makes it one of the important things these days. In addition to this, these applications are sometimes developed as software that can be installed on users computers, laptops and even smartphones, which often attracts many attackers to compromise their computers with malware that is unintentionally installed in the computer. Gadgets and even computer systems. computer background. Many solutions have been employed to detect if these malware are installed. This paper aims to evaluate and study the effectiveness of machine learning methods in detecting and classifying malware being installed. This paper employs heuristics and machine learning classifiers to identify malware attacks detected in each website or software application. The study compares 3 classifiers to find the best machine learning classifier for detecting malware attacks. Prove that the cloud sandbox can achieve a high detection accuracy of 99.8% true positive rate value when identifying malware attacks? Use website features. Results show that Cloud Sandbox is an effective classifier for detecting malware attacks.
{"title":"Malware Detection System Using Cloud Sandbox, Machine Learning","authors":"Mohd Azuwan EfendyMail, Mohd Faizal Ab Razak, Munirah Ab. Rahman","doi":"10.15282/ijsecs.8.2.2022.3.0100","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.3.0100","url":null,"abstract":"Today's internet continues to move forward, and with it comes the development of many applications. Therefore, these applications are also directly accessible via the Internet, which makes it one of the important things these days. In addition to this, these applications are sometimes developed as software that can be installed on users computers, laptops and even smartphones, which often attracts many attackers to compromise their computers with malware that is unintentionally installed in the computer. Gadgets and even computer systems. computer background. Many solutions have been employed to detect if these malware are installed. This paper aims to evaluate and study the effectiveness of machine learning methods in detecting and classifying malware being installed. This paper employs heuristics and machine learning classifiers to identify malware attacks detected in each website or software application. The study compares 3 classifiers to find the best machine learning classifier for detecting malware attacks. Prove that the cloud sandbox can achieve a high detection accuracy of 99.8% true positive rate value when identifying malware attacks? Use website features. Results show that Cloud Sandbox is an effective classifier for detecting malware attacks.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"2017 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84103194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.2.0099
Fahmi Akmal Dzulkifli, M. Y. Mashor, H. Jaafar
Meningioma is a type of primary brain tumour where this tumour arises in the three thin layers of tissues, called meninges. Tumour grading is usually used to describe tumour cells' characteristics and behaviours and how they look under a microscope. There were many techniques used for determining the grade of the tumour. Ki67 was the most common proliferation marker used to measure cell proliferation activity. Currently, pathologists used the manual counting technique to count the Ki67 cells before determining tumour grading. However, this technique was time-consuming, tiring and the counting results are often not accurate. Besides that, manual counting has poor reproducibility and discordant between counting values’ among the pathologist. Therefore, this study aimed to develop a Computer-Aided Design (CAD) software that automatically counts the Ki67 cells for determining tumour grading. The purpose of developing this software is to alleviate pathologists’ workload associated with counting Ki67 cells and scoring the Ki67 index. The CAD software was developed through seven stages. Based on Pearson Correlation Coefficient results, there was a good positive correlation between the proposed technique with the manual counting technique in counting positive and negative Ki67 cells with a correlation of 0.99 and 0.72 respectively. The proposed CAD system also showed promising results in computing the Ki67 labeling index with a low percentage absolute error of 1.85%.
{"title":"A Computer-Aided Diagnosis (CAD) System for Automatic Counting of Ki67 Cells in Meningioma","authors":"Fahmi Akmal Dzulkifli, M. Y. Mashor, H. Jaafar","doi":"10.15282/ijsecs.8.2.2022.2.0099","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.2.0099","url":null,"abstract":"Meningioma is a type of primary brain tumour where this tumour arises in the three thin layers of tissues, called meninges. Tumour grading is usually used to describe tumour cells' characteristics and behaviours and how they look under a microscope. There were many techniques used for determining the grade of the tumour. Ki67 was the most common proliferation marker used to measure cell proliferation activity. Currently, pathologists used the manual counting technique to count the Ki67 cells before determining tumour grading. However, this technique was time-consuming, tiring and the counting results are often not accurate. Besides that, manual counting has poor reproducibility and discordant between counting values’ among the pathologist. Therefore, this study aimed to develop a Computer-Aided Design (CAD) software that automatically counts the Ki67 cells for determining tumour grading. The purpose of developing this software is to alleviate pathologists’ workload associated with counting Ki67 cells and scoring the Ki67 index. The CAD software was developed through seven stages. Based on Pearson Correlation Coefficient results, there was a good positive correlation between the proposed technique with the manual counting technique in counting positive and negative Ki67 cells with a correlation of 0.99 and 0.72 respectively. The proposed CAD system also showed promising results in computing the Ki67 labeling index with a low percentage absolute error of 1.85%.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"13 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85432595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.6.0103
F. Chete, O. Ikeh
Software Engineers, Computer Scientists, and Software Experts alike are faced to decide which programming language is best suited for a certain purpose as the use of programming languages grows. When we consider the various types of programming languages available today, such as Domain Specific Languages (DSL), General Purpose Languages (GPL), Functional Programming Languages (FPL), Imperative Programming Languages (IPL), amongst others, this becomes complicated. In this study, we introduce BeeX, an interpreted language, with the aim of showing the process and principles involved in language design and consider various choices faced by language designers of various programming languages. BeeX was created with simplicity in mind, thus the study focused on architectural design options. We look at the implementation standpoint and try to figure out what the basic building parts of most programming languages are, such as lexical analysis, syntax analysis, and evaluation phase. To achieve this, we created an interactive command interface that evaluated various BeeX language constructs(conditional logic statements, arithmetic expressions, loop constructs etc.) which allowed students to easily experiment with the proposed language. The results of the tests showed that students and programmers alike can use the BeeX programming language to create a variety of code structures that are simple to use.
{"title":"Towards the Design and Implementation of a Programming Language (Beex)","authors":"F. Chete, O. Ikeh","doi":"10.15282/ijsecs.8.2.2022.6.0103","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.6.0103","url":null,"abstract":"Software Engineers, Computer Scientists, and Software Experts alike are faced to decide which programming language is best suited for a certain purpose as the use of programming languages grows. When we consider the various types of programming languages available today, such as Domain Specific Languages (DSL), General Purpose Languages (GPL), Functional Programming Languages (FPL), Imperative Programming Languages (IPL), amongst others, this becomes complicated. In this study, we introduce BeeX, an interpreted language, with the aim of showing the process and principles involved in language design and consider various choices faced by language designers of various programming languages. BeeX was created with simplicity in mind, thus the study focused on architectural design options. We look at the implementation standpoint and try to figure out what the basic building parts of most programming languages are, such as lexical analysis, syntax analysis, and evaluation phase. To achieve this, we created an interactive command interface that evaluated various BeeX language constructs(conditional logic statements, arithmetic expressions, loop constructs etc.) which allowed students to easily experiment with the proposed language. The results of the tests showed that students and programmers alike can use the BeeX programming language to create a variety of code structures that are simple to use.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84557121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.1.0098
Suit Yan Ng, R. Mohamed
An online voting system is an election system that manages the election process. This is a medium for the voters to cast their votes. It is also being used to calculate the votes collected from the voters to choose the representative for their own faculty. A typical voting system is based on a single attempt for each candidate being voted. The voting does not reflect the criteria implies to the characteristic of the candidate that going to be the student leader. To be a student leader, the student should fulfil the requirement such as good academic results, interpersonal skills with society, involving in activities of university and etc. Although the current voting system is able to maximize the participation of the voters, the voters may blindly vote the ballots casually due to they do not know the details of the candidates and the result is low quality and low public’s trust in the selected candidate. In this study, the aim is to develop an interactive online voting system that have ranking feature with MCDM method which allow online voting system to collect high-quality results from the voters. The Multiple Criteria Decision Making (MCDM) method is used in the voting system while choosing the candidate. MCDM can let the voters make decision making or selecting the candidate based on the criteria that suit the position. The study starts with the literature study on implementing MCDM for a voting system. Then, a survey will be made to get the users’ views on the with and without implementation of the MCDM method in an online voting system. The expected result of the study is to investigate the current implementation of MCDM as a tool for decision making, then identify the possibility of adopting MCDM for the online voting system while choosing the representative for faculty students’ society. As a conclusion from the survey from the users’ views, it shown that most of the users thinks that the system with the implementation with MCDM method is less time consuming and able to produce high quality result compare to the current online voting system. Most of the respondents also stated that they are more preferring to use the online voting system with MCDM method in the future.
{"title":"Feasibility Study on using MCDM for E-Voting","authors":"Suit Yan Ng, R. Mohamed","doi":"10.15282/ijsecs.8.2.2022.1.0098","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.1.0098","url":null,"abstract":"An online voting system is an election system that manages the election process. This is a medium for the voters to cast their votes. It is also being used to calculate the votes collected from the voters to choose the representative for their own faculty. A typical voting system is based on a single attempt for each candidate being voted. The voting does not reflect the criteria implies to the characteristic of the candidate that going to be the student leader. To be a student leader, the student should fulfil the requirement such as good academic results, interpersonal skills with society, involving in activities of university and etc. Although the current voting system is able to maximize the participation of the voters, the voters may blindly vote the ballots casually due to they do not know the details of the candidates and the result is low quality and low public’s trust in the selected candidate. In this study, the aim is to develop an interactive online voting system that have ranking feature with MCDM method which allow online voting system to collect high-quality results from the voters. The Multiple Criteria Decision Making (MCDM) method is used in the voting system while choosing the candidate. MCDM can let the voters make decision making or selecting the candidate based on the criteria that suit the position. The study starts with the literature study on implementing MCDM for a voting system. Then, a survey will be made to get the users’ views on the with and without implementation of the MCDM method in an online voting system. The expected result of the study is to investigate the current implementation of MCDM as a tool for decision making, then identify the possibility of adopting MCDM for the online voting system while choosing the representative for faculty students’ society. As a conclusion from the survey from the users’ views, it shown that most of the users thinks that the system with the implementation with MCDM method is less time consuming and able to produce high quality result compare to the current online voting system. Most of the respondents also stated that they are more preferring to use the online voting system with MCDM method in the future.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"112 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72587934","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.7.0104
Fakihotun Titiani, D. Riana
Artificial intelligence, commonly known as AI, has greatly influenced marketing strategies, including business models, sales processes, customer service options, and customer behaviour in receiving marketing campaigns. In a marketing campaign, all customers are targeted by advertising, including those who will not respond positively to the marketing campaign and reject the offer. This means that the company is inefficient; the marketing campaign is ineffective because customers are not segmented and targeted. As a result, costs increase and the company's profit decreases. Thence, this leads to the failure of the company's marketing campaigns. The purpose of this study is to experiment with using Ensemble Learning and tuning on the Marketing Campaign dataset by providing the classification methods. That classification method is called the Light Gradient Boosting Machine (LightGBM), Gradient Boosting Classifier (GBC), and AdaBoost Classifier (ADA), which have never been used in the classification of the Marketing Campaign dataset. The study results in the highest model with an accuracy value of 98.64%, AUC 0.9994, recall 95.77%, precision 95.77%, F1-score 95.77%, and kappa 94.98% when using the LightGBM for marketing campaign predictions
{"title":"Ensemble Learning for the Prediction of Marketing Campaign Acceptance","authors":"Fakihotun Titiani, D. Riana","doi":"10.15282/ijsecs.8.2.2022.7.0104","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.7.0104","url":null,"abstract":"Artificial intelligence, commonly known as AI, has greatly influenced marketing strategies, including business models, sales processes, customer service options, and customer behaviour in receiving marketing campaigns. In a marketing campaign, all customers are targeted by advertising, including those who will not respond positively to the marketing campaign and reject the offer. This means that the company is inefficient; the marketing campaign is ineffective because customers are not segmented and targeted. As a result, costs increase and the company's profit decreases. Thence, this leads to the failure of the company's marketing campaigns. The purpose of this study is to experiment with using Ensemble Learning and tuning on the Marketing Campaign dataset by providing the classification methods. That classification method is called the Light Gradient Boosting Machine (LightGBM), Gradient Boosting Classifier (GBC), and AdaBoost Classifier (ADA), which have never been used in the classification of the Marketing Campaign dataset. The study results in the highest model with an accuracy value of 98.64%, AUC 0.9994, recall 95.77%, precision 95.77%, F1-score 95.77%, and kappa 94.98% when using the LightGBM for marketing campaign predictions","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"27 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82311543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.4.0101
D. Riana
Predicted click-through rate is one of the most frequently used criteria to determine the effectiveness of an ads. In advertising production, click-through predictions are very influential for the company that places the ads. Click-through rates need to be predicted accurately because accurate prediction results determine whether the click-through rate is exactly clicked or not by the viewing consumer. Predicted click-through can be done on advertising and social network datasets. The use of these two datasets is intended to make the comparison results more convincing from the proposed method. The purpose of this study is to compare two advertising and social network datasets, by proposing the application of the Deep Neural Network (DNN) model by testing hyperparameter variations to find a better architecture in predicting whether or not users click on an advertisement. The hyperparameter variations include 3 variations of the hidden layer, 2 variations of the activation function, namely ReLuand Sigmoid, 3 variations of the optimization (RMSprop, Adam, and Adagrad),and 3 variations of the learning rate (0.1, 0.01, and 0.001). The results of experiments conducted with the advertising parameter dataset with hidden layer of 3, learning rate of 0.01,and Adam optimization with an accuracy value of 99.90%, AUC of 99.90% and Precision-Recallof99.89% while the data for social network ads parameters with hidden layer of 5, learning rate of 0.1 and Adam optimization with accuracy of 92.25%, AUC of 92.72%,andPrecision-Recallof 89.70%.
{"title":"Deep Neural Network for Click-Through Rate Prediction","authors":"D. Riana","doi":"10.15282/ijsecs.8.2.2022.4.0101","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.4.0101","url":null,"abstract":"Predicted click-through rate is one of the most frequently used criteria to determine the effectiveness of an ads. In advertising production, click-through predictions are very influential for the company that places the ads. Click-through rates need to be predicted accurately because accurate prediction results determine whether the click-through rate is exactly clicked or not by the viewing consumer. Predicted click-through can be done on advertising and social network datasets. The use of these two datasets is intended to make the comparison results more convincing from the proposed method. The purpose of this study is to compare two advertising and social network datasets, by proposing the application of the Deep Neural Network (DNN) model by testing hyperparameter variations to find a better architecture in predicting whether or not users click on an advertisement. The hyperparameter variations include 3 variations of the hidden layer, 2 variations of the activation function, namely ReLuand Sigmoid, 3 variations of the optimization (RMSprop, Adam, and Adagrad),and 3 variations of the learning rate (0.1, 0.01, and 0.001). The results of experiments conducted with the advertising parameter dataset with hidden layer of 3, learning rate of 0.01,and Adam optimization with an accuracy value of 99.90%, AUC of 99.90% and Precision-Recallof99.89% while the data for social network ads parameters with hidden layer of 5, learning rate of 0.1 and Adam optimization with accuracy of 92.25%, AUC of 92.72%,andPrecision-Recallof 89.70%.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75519348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-01DOI: 10.15282/ijsecs.8.2.2022.5.0102
A. Sallam, Taha H. Rassem, Hanadi Abdu, Haneen S. Abdulkareem, Nada Saif, Samia Abdullah
On the Ethereum network, users communicate with one another through a variety of different accounts. Pseudo-anonymity was enforced over the network to provide the highest level of privacy. By using accounts that engage in fraudulent activity across the network, such privacy may be exploited. Like other cryptocurrencies, Ethereum blockchain may exploited with several fraudulent activities such as Ponzi schemes, phishing, or Initial Coin Offering (ICO) exits, etc. However, the identification of parameters with abnormal account characteristics is not an easy task and requires an intelligent approach to distinguish between normal and fraudulent activities. Therefore, this paper has attempted to solve this a problem by using machine learning techniques to introduce a robust approach that can detect fraudulent accounts on Ethereum. We have used a K-Nearest Neighbor, Random Forest and XGBoost over a collected dataset of 4,681 instances along with 2,179 fraudulent accounts associated and 2,502 regular accounts. The XGBoost, RF, and KNN techniques achieved average accuracies of 96.80 %, 94.8 8%, and 87.85% and an average AUC of 0.995, 0.99 and 0.93, respectively.
{"title":"Fraudulent Account Detection in the Ethereum’s Network Using Various Machine Learning Techniques","authors":"A. Sallam, Taha H. Rassem, Hanadi Abdu, Haneen S. Abdulkareem, Nada Saif, Samia Abdullah","doi":"10.15282/ijsecs.8.2.2022.5.0102","DOIUrl":"https://doi.org/10.15282/ijsecs.8.2.2022.5.0102","url":null,"abstract":"On the Ethereum network, users communicate with one another through a variety of different accounts. Pseudo-anonymity was enforced over the network to provide the highest level of privacy. By using accounts that engage in fraudulent activity across the network, such privacy may be exploited. Like other cryptocurrencies, Ethereum blockchain may exploited with several fraudulent activities such as Ponzi schemes, phishing, or Initial Coin Offering (ICO) exits, etc. However, the identification of parameters with abnormal account characteristics is not an easy task and requires an intelligent approach to distinguish between normal and fraudulent activities. Therefore, this paper has attempted to solve this a problem by using machine learning techniques to introduce a robust approach that can detect fraudulent accounts on Ethereum. We have used a K-Nearest Neighbor, Random Forest and XGBoost over a collected dataset of 4,681 instances along with 2,179 fraudulent accounts associated and 2,502 regular accounts. The XGBoost, RF, and KNN techniques achieved average accuracies of 96.80 %, 94.8 8%, and 87.85% and an average AUC of 0.995, 0.99 and 0.93, respectively.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"66 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90266920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-10DOI: 10.15282/ijsecs.8.1.2022.7.0097
Himanshi Chaudhary, Himanshu Chaudhary, A. Sharma
Cloud computing is a mode to increase competence and capabilities devoid of investing in any infrastructure. It seems that in cloud computing environment the major problem that ensure the secure communication and protect responsive data in open networks from unauthorised access. These days it seems the headlines are jam-packed with stories about security breaches to these services; that result in the leak of a large amount of private data of the users. As cloud computing can offer new computing benefits, but it faces soaring risks, specifically on the security side where DOS attacks can make cloud services unavailable. This paper aims to turn up an effective method of detecting DOS attacks with an optimized Genetic algorithm and extended version of Diffie-hellman algorithm. To prevent data loss or corruption caused by the insiders in the cloud, Optimized Genetic Algorithm (OGA) is utilized, which effectively recovers the data and retrieve it if the missed data without loss. It is then followed with the decryption process as if requested by the user. An optimized path assortment for information broadcast proves to be an effective method in the cloud computing atmosphere. The proposed framework ensures certification and paves way for secure data access in an unauthorized network, with improved performance. It successfully assure the high level of protection of the transmission and data transmitted. And concurrently reduce the communication complexities.To reduce time complexity and detect the attackers by mutual secret key that is brought on using extended version of Diffie-hellman to endorse available key generation.
{"title":"OPTIMIZED GENETIC ALGORITHM AND EXTENDED DIFFIE HELLMAN AS AN EFFECTUAL APPROACH FOR DOS-ATTACK DETECTION IN CLOUD","authors":"Himanshi Chaudhary, Himanshu Chaudhary, A. Sharma","doi":"10.15282/ijsecs.8.1.2022.7.0097","DOIUrl":"https://doi.org/10.15282/ijsecs.8.1.2022.7.0097","url":null,"abstract":" Cloud computing is a mode to increase competence and capabilities devoid of investing in any infrastructure. It seems that in cloud computing environment the major problem that ensure the secure communication and protect responsive data in open networks from unauthorised access. These days it seems the headlines are jam-packed with stories about security breaches to these services; that result in the leak of a large amount of private data of the users. As cloud computing can offer new computing benefits, but it faces soaring risks, specifically on the security side where DOS attacks can make cloud services unavailable. This paper aims to turn up an effective method of detecting DOS attacks with an optimized Genetic algorithm and extended version of Diffie-hellman algorithm. To prevent data loss or corruption caused by the insiders in the cloud, Optimized Genetic Algorithm (OGA) is utilized, which effectively recovers the data and retrieve it if the missed data without loss. It is then followed with the decryption process as if requested by the user. An optimized path assortment for information broadcast proves to be an effective method in the cloud computing atmosphere. The proposed framework ensures certification and paves way for secure data access in an unauthorized network, with improved performance. It successfully assure the high level of protection of the transmission and data transmitted. And concurrently reduce the communication complexities.To reduce time complexity and detect the attackers by mutual secret key that is brought on using extended version of Diffie-hellman to endorse available key generation.","PeriodicalId":31240,"journal":{"name":"International Journal of Software Engineering and Computer Systems","volume":"103 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2022-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79262572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：