Pub Date : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464568
Birol Kuyumcu, Selman Delil, Cüneyt Aksakalli
This paper describes our contribution to ICICS 2021 Mowjaz Multi-Topic Labelling Task. The purpose of the task is to classify Arabic articles based on their topics. Participating systems are expected to select one or more of the determined topics for each given article. In our system, we experiment with state-of-art pre-trained language models (GigaBERT-v4 and Arabic BERT) and a classical logistic regression to find the best effective model for the problem. We obtained the highest F1-score of 0.8563 with GigaBERT-v4 while Arabic-BERT and logistic regression reached 0.8442 and 0.8081 respectively. Our system ranked 2nd in the competition very close to the winner.
{"title":"Sefamerve R&D at ICICS 2021 Mowjaz Multi-Topic Labelling Task","authors":"Birol Kuyumcu, Selman Delil, Cüneyt Aksakalli","doi":"10.1109/ICICS52457.2021.9464568","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464568","url":null,"abstract":"This paper describes our contribution to ICICS 2021 Mowjaz Multi-Topic Labelling Task. The purpose of the task is to classify Arabic articles based on their topics. Participating systems are expected to select one or more of the determined topics for each given article. In our system, we experiment with state-of-art pre-trained language models (GigaBERT-v4 and Arabic BERT) and a classical logistic regression to find the best effective model for the problem. We obtained the highest F1-score of 0.8563 with GigaBERT-v4 while Arabic-BERT and logistic regression reached 0.8442 and 0.8081 respectively. Our system ranked 2nd in the competition very close to the winner.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"43 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116338539","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464616
Sofiane Hamrioui, Upasana Dohare, D. K. Lobiyal
Fairly and efficiently energy utilization among the devices and the delay spent by sending their data are a majors constraints within Internet of Things (IoT) environment. We focus the applicability of coalition game theory to stimulate cooperation among devices and analysing the performance of multipath routing to meet the quality of service (QoS). In this paper, we present a new routing algorithm, called RACD (Routing Algorithm based on Cooperation between Devices) in order to optimize the energy consumption and the link delay during the routing process. The principles that have been exploited by RACD are the multi-metrics routing and the coalitional game theory. RACD allow to selects a route with lower delay and minimum energy consumption among multiple coalitions. Others principles have been exploited too by our proposed algorithm, such as the payoff function of coalition, the core of the game and the Shapely. The performances of our proposed routing algorithm have been evaluated by simulations according to important metrics and the results obtained have been compared to two others routing solutions that are ELDR and MAODV.
在物联网(IoT)环境中，设备之间公平有效的能源利用和发送数据所花费的延迟是一个主要的制约因素。重点研究了联盟博弈论在激励设备间合作中的适用性，并分析了多径路由满足服务质量(QoS)的性能。为了优化路由过程中的能量消耗和链路延迟，本文提出了一种新的路由算法，称为RACD (routing algorithm based on cooperative between Devices)。RACD所利用的原则是多指标路由和联合博弈论。RACD允许在多个联盟中选择时延较低、能耗最小的路由。我们提出的算法也利用了其他原则，如联盟的收益函数、游戏的核心和Shapely。我们提出的路由算法的性能根据重要指标进行了仿真评估，并得到的结果与另外两种路由解决方案ELDR和MAODV进行了比较。
{"title":"A New Routing Algorithm based on Devices Cooperation for better QoS within IoT","authors":"Sofiane Hamrioui, Upasana Dohare, D. K. Lobiyal","doi":"10.1109/ICICS52457.2021.9464616","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464616","url":null,"abstract":"Fairly and efficiently energy utilization among the devices and the delay spent by sending their data are a majors constraints within Internet of Things (IoT) environment. We focus the applicability of coalition game theory to stimulate cooperation among devices and analysing the performance of multipath routing to meet the quality of service (QoS). In this paper, we present a new routing algorithm, called RACD (Routing Algorithm based on Cooperation between Devices) in order to optimize the energy consumption and the link delay during the routing process. The principles that have been exploited by RACD are the multi-metrics routing and the coalitional game theory. RACD allow to selects a route with lower delay and minimum energy consumption among multiple coalitions. Others principles have been exploited too by our proposed algorithm, such as the payoff function of coalition, the core of the game and the Shapely. The performances of our proposed routing algorithm have been evaluated by simulations according to important metrics and the results obtained have been compared to two others routing solutions that are ELDR and MAODV.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"131 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116754192","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464587
Siwar Kriaa, Yahia Chaabane
Advanced persistent threats targeting sensitive corporations, are becoming today stealthier and more complex, coordinating different attacks steps and lateral movements, and trying to stay undetected for long time. Classical security solutions that rely on signature-based detection can be easily thwarted by malware using obfuscation and encryption techniques. More recent solutions are using machine learning approaches for detecting outliers. Nevertheless, the majority of them reason on tabular unstructured data which can lead to missing obvious conclusions. We propose in this paper a novel approach that leverages a combination of both knowledge graphs and machine learning techniques to detect and predict attacks. Using Cyber Threat Intelligence (CTI), we built a knowledge graph that processes event logs in order to not only detect attack techniques, but also learn how to predict them.
{"title":"SecKG: Leveraging attack detection and prediction using knowledge graphs","authors":"Siwar Kriaa, Yahia Chaabane","doi":"10.1109/ICICS52457.2021.9464587","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464587","url":null,"abstract":"Advanced persistent threats targeting sensitive corporations, are becoming today stealthier and more complex, coordinating different attacks steps and lateral movements, and trying to stay undetected for long time. Classical security solutions that rely on signature-based detection can be easily thwarted by malware using obfuscation and encryption techniques. More recent solutions are using machine learning approaches for detecting outliers. Nevertheless, the majority of them reason on tabular unstructured data which can lead to missing obvious conclusions. We propose in this paper a novel approach that leverages a combination of both knowledge graphs and machine learning techniques to detect and predict attacks. Using Cyber Threat Intelligence (CTI), we built a knowledge graph that processes event logs in order to not only detect attack techniques, but also learn how to predict them.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114538747","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464581
Sireen Abuqran
The number of text documents on the internet is rapidly increasing. As a result, there is a growing demand for methods that can automatically organize and identify electronic documents (instances). The multi-label classification task has been used in a variety of applications and is commonly used in real-world problems. It simultaneously assigns multiple labels to each text. In the Arabic language, there have been few and inadequate research studies on the multi-label text classification issue. In this paper, I proposed a deep learning model using bidirectional long short-term memory (BiLSTM) for multi-class topic classification using Mowjaz Multi-Topic Labelling Task dataset. The BiLSTM model consists of 4 layers only which can be considered as light weight model, these layers are input layer, bidirectional LSTM layer, and two dense layers. The results show that the model successfully to classify topics with F1-Socre of 0.8089 on the testing dataset.
{"title":"Arabic Multi-Topic Labelling using Bidirectional Long Short-Term Memory","authors":"Sireen Abuqran","doi":"10.1109/ICICS52457.2021.9464581","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464581","url":null,"abstract":"The number of text documents on the internet is rapidly increasing. As a result, there is a growing demand for methods that can automatically organize and identify electronic documents (instances). The multi-label classification task has been used in a variety of applications and is commonly used in real-world problems. It simultaneously assigns multiple labels to each text. In the Arabic language, there have been few and inadequate research studies on the multi-label text classification issue. In this paper, I proposed a deep learning model using bidirectional long short-term memory (BiLSTM) for multi-class topic classification using Mowjaz Multi-Topic Labelling Task dataset. The BiLSTM model consists of 4 layers only which can be considered as light weight model, these layers are input layer, bidirectional LSTM layer, and two dense layers. The results show that the model successfully to classify topics with F1-Socre of 0.8089 on the testing dataset.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114977027","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464541
Karan Singh, R. B. S., R. Shyamasundar
SELinux policies used in practice are generally large and complex. As a result, it is difficult for the policy writers to completely understand the policy and ensure that the policy meets the intended security goals. To remedy this, we have developed a tool called SEFlowViz that helps in visualizing the information flows of a policy and thereby helps in creating flow-secure policies. The tool uses the graph database Neo4j to visualize the policy. Along with visualization, the tool also supports extracting various information regarding the policy and its components through queries. Furthermore, the tool also supports the addition and deletion of rules which is useful in converting inconsistent policies into consistent policies.
{"title":"SEFlowViz: A Visualization Tool for SELinux Policy Analysis","authors":"Karan Singh, R. B. S., R. Shyamasundar","doi":"10.1109/ICICS52457.2021.9464541","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464541","url":null,"abstract":"SELinux policies used in practice are generally large and complex. As a result, it is difficult for the policy writers to completely understand the policy and ensure that the policy meets the intended security goals. To remedy this, we have developed a tool called SEFlowViz that helps in visualizing the information flows of a policy and thereby helps in creating flow-secure policies. The tool uses the graph database Neo4j to visualize the policy. Along with visualization, the tool also supports extracting various information regarding the policy and its components through queries. Furthermore, the tool also supports the addition and deletion of rules which is useful in converting inconsistent policies into consistent policies.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115088978","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464572
Stijn Pletinckx, G. Jansen, A. Brussen, R. van Wegberg
The COVID-19 pandemic introduced novel incentives for adversaries to exploit the state of turmoil. As we have witnessed with the increase in for instance phishing attacks and domain name registrations piggybacking the COVID-19 brand name. In this paper, we perform an analysis at Internet-scale of COVID-19 domain name registrations during the early stages of the virus’ spread, and investigate the rationales behind them. We leverage the DomainTools COVID-19 Threat List and additional measurements to analyze over 150,000 domains registered between January 1st 2020 and May 1st 2020. We identify two key rationales for covid-related domain registrations. Online marketing, by either redirecting traffic or hosting a commercial service on the domain, and domain parking, by registering domains containing popular COVID-19 keywords, presumably anticipating a profit when reselling the domain later on. We also highlight three public policy take-aways that can counteract this domain registration behavior.
{"title":"Cash for the Register? Capturing Rationales of Early COVID-19 Domain Registrations at Internet-scale","authors":"Stijn Pletinckx, G. Jansen, A. Brussen, R. van Wegberg","doi":"10.1109/ICICS52457.2021.9464572","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464572","url":null,"abstract":"The COVID-19 pandemic introduced novel incentives for adversaries to exploit the state of turmoil. As we have witnessed with the increase in for instance phishing attacks and domain name registrations piggybacking the COVID-19 brand name. In this paper, we perform an analysis at Internet-scale of COVID-19 domain name registrations during the early stages of the virus’ spread, and investigate the rationales behind them. We leverage the DomainTools COVID-19 Threat List and additional measurements to analyze over 150,000 domains registered between January 1st 2020 and May 1st 2020. We identify two key rationales for covid-related domain registrations. Online marketing, by either redirecting traffic or hosting a commercial service on the domain, and domain parking, by registering domains containing popular COVID-19 keywords, presumably anticipating a profit when reselling the domain later on. We also highlight three public policy take-aways that can counteract this domain registration behavior.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129451041","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464566
Rahaf M. AL Mgheed
Recently, with the existence of internet we have witnessed great development in computer systems. Artificial intelligence means to developing computer systems that able to perform intelligent tasks.[1] Machine learning is one method of make such systems. In this paper, using SVM classifier, I build a Multi-label text classification model for Arabic text. This model is basically used to classify articles on their topics. The results show that using SVM classifier on the dataset generated the best results with 82.2% accuracy. The model was build using Python.
{"title":"Scalable Arabic text Classification Using Machine Learning Model","authors":"Rahaf M. AL Mgheed","doi":"10.1109/ICICS52457.2021.9464566","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464566","url":null,"abstract":"Recently, with the existence of internet we have witnessed great development in computer systems. Artificial intelligence means to developing computer systems that able to perform intelligent tasks.[1] Machine learning is one method of make such systems. In this paper, using SVM classifier, I build a Multi-label text classification model for Arabic text. This model is basically used to classify articles on their topics. The results show that using SVM classifier on the dataset generated the best results with 82.2% accuracy. The model was build using Python.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122489179","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464546
I. Kich, El Bachir Ameur, Y. Taouil, Amine Benhfid
Nowadays, Convolutional Neural Networks (CNN) have allowed us to solve many problems, difficult to solve with classical methods, in different fields of applications. The use of this technique in the field of modern steganography has improved the performance of steganographic schemes in terms of concealability and invisibility. In this article, we propose a system based on CNN in order to hide a color image in another color image of the same size. The proposed system is based on Auto-Encoder network and U-net architecture. The network is subdivided into two sub-networks, the first is for the concealment of the image secret by the sender, the second is for its extraction by the receiver. The network is end to end trained to ensure the integrity of the concealment and extraction process. The tests were performed on challenging images dataset publicly available, such as ImageNet, LFW, PASCAL-VOC12. The results show that the proposed steganographic scheme can hide a color image in another one of the same sizes, i.e. a capacity of 24 bpp, with acceptable PSNR and SSIM values compared to other previous work.
{"title":"Image Steganography Scheme Using Dilated Convolutional Network","authors":"I. Kich, El Bachir Ameur, Y. Taouil, Amine Benhfid","doi":"10.1109/ICICS52457.2021.9464546","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464546","url":null,"abstract":"Nowadays, Convolutional Neural Networks (CNN) have allowed us to solve many problems, difficult to solve with classical methods, in different fields of applications. The use of this technique in the field of modern steganography has improved the performance of steganographic schemes in terms of concealability and invisibility. In this article, we propose a system based on CNN in order to hide a color image in another color image of the same size. The proposed system is based on Auto-Encoder network and U-net architecture. The network is subdivided into two sub-networks, the first is for the concealment of the image secret by the sender, the second is for its extraction by the receiver. The network is end to end trained to ensure the integrity of the concealment and extraction process. The tests were performed on challenging images dataset publicly available, such as ImageNet, LFW, PASCAL-VOC12. The results show that the proposed steganographic scheme can hide a color image in another one of the same sizes, i.e. a capacity of 24 bpp, with acceptable PSNR and SSIM values compared to other previous work.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120960027","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464543
Esraa Mggdadi, Ahmad Al-Aiad, M. Al-Ayyad, Alaa Darabseh
Alzheimer's is one of the diseases that are the most publicized type of dementia. Alzheimer's disease will be born every 3 second the world. Previous research shows that early prediction of AD in the medical field for reduced cost of treatment and time of it. To this end, construct an efficient prediction system for AD, which is the goal of this paper, often reduces time to treatment, medical errors, and overall healthcare cost. We used Deep Learning to predict and diagnose AD and for this reason using python code in Colaboratory as platform environments. In particular, we used 2D CNN and vgg16 to achieve the research goal, we used experiments conducted on MRI images from Kaggle dataset. Our experiment achieved accuracy of 67.5% for 2D CNN algorithm, while the vgg16 algorithm achieved accuracy of 70.3%. We conclude by showing that deep learning can improve the prediction AD and using algorithm vgg16 is better than 2D CNN.
{"title":"Prediction Alzheimer's disease from MRI images using deep learning","authors":"Esraa Mggdadi, Ahmad Al-Aiad, M. Al-Ayyad, Alaa Darabseh","doi":"10.1109/ICICS52457.2021.9464543","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464543","url":null,"abstract":"Alzheimer's is one of the diseases that are the most publicized type of dementia. Alzheimer's disease will be born every 3 second the world. Previous research shows that early prediction of AD in the medical field for reduced cost of treatment and time of it. To this end, construct an efficient prediction system for AD, which is the goal of this paper, often reduces time to treatment, medical errors, and overall healthcare cost. We used Deep Learning to predict and diagnose AD and for this reason using python code in Colaboratory as platform environments. In particular, we used 2D CNN and vgg16 to achieve the research goal, we used experiments conducted on MRI images from Kaggle dataset. Our experiment achieved accuracy of 67.5% for 2D CNN algorithm, while the vgg16 algorithm achieved accuracy of 70.3%. We conclude by showing that deep learning can improve the prediction AD and using algorithm vgg16 is better than 2D CNN.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"46 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132134717","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 : 2021-05-24DOI: 10.1109/ICICS52457.2021.9464590
Qusai Ismail, Eslam Al-Sobh, Sarah Al-Omari, Tuqa M. Bani Yaseen, Malak Abdullah
The financial economy in Africa faces significant challenges that affect development and livelihood. One of these challenges is holding a bank account in Africa, indicating the person’s stable economic status. There is a need to solve bank problems in Africa and find solutions to the banking problems. Studies on this topic consider the enormous number of people who do not have a bank account compared to those who have and how this contributes to the decline of Africa’s economy. Therefore, in this research, we have implemented effective mechanisms using machine learning techniques to predict who owns a bank account and who is not in African banks. We used different machine learning algorithms, such as SVM, Naive Bays, Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, Bagging, AdaBoosting, Voting Ensemble, KNN, Stack, and XGBoosting Classifiers. We have experimented with these techniques on a public dataset obtained from African banks (publically available on Zindi) to predict whether a person has a bank account or not. We used the Holdout cross-validation method to split the training dataset randomly to train and validation. The results show that the XGBoost model has a superior accuracy score of 89.23%. This paper provides a comprehensive comparison for all mentioned models, which we used to perform our study.
{"title":"Using Machine Learning Algorithms to Predict the State of Financial Inclusion in Africa","authors":"Qusai Ismail, Eslam Al-Sobh, Sarah Al-Omari, Tuqa M. Bani Yaseen, Malak Abdullah","doi":"10.1109/ICICS52457.2021.9464590","DOIUrl":"https://doi.org/10.1109/ICICS52457.2021.9464590","url":null,"abstract":"The financial economy in Africa faces significant challenges that affect development and livelihood. One of these challenges is holding a bank account in Africa, indicating the person’s stable economic status. There is a need to solve bank problems in Africa and find solutions to the banking problems. Studies on this topic consider the enormous number of people who do not have a bank account compared to those who have and how this contributes to the decline of Africa’s economy. Therefore, in this research, we have implemented effective mechanisms using machine learning techniques to predict who owns a bank account and who is not in African banks. We used different machine learning algorithms, such as SVM, Naive Bays, Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, Bagging, AdaBoosting, Voting Ensemble, KNN, Stack, and XGBoosting Classifiers. We have experimented with these techniques on a public dataset obtained from African banks (publically available on Zindi) to predict whether a person has a bank account or not. We used the Holdout cross-validation method to split the training dataset randomly to train and validation. The results show that the XGBoost model has a superior accuracy score of 89.23%. This paper provides a comprehensive comparison for all mentioned models, which we used to perform our study.","PeriodicalId":421803,"journal":{"name":"2021 12th International Conference on Information and Communication Systems (ICICS)","volume":"2 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131141758","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：