Pub Date : 2022-11-09DOI: 10.1109/ZCICT55726.2022.10045846
Casper Chigwedere, Sam Takavarasha, B. Chisaka
This Regulatory environments are judged by their independence from both the state and from stakeholders. This paper investigates the gaps in the independence of the ICT regulatory systems in Zimbabwe using Heeks’ [1] Design Reality Gap (DRG) under an interpretivist paradigm. Data collection used in-depth interviews with purposively selected stakeholders and the analysed using NVivo (release 1.6.1 (1136)) for a thematic analysis based on DRG constructs against regulatory independence. The results show a gap of 9 emanating from a private sector which perceived it as non-independent and stateowned enterprises that saw a partially independent regulatory system. This lack of independence from the state was however believed to be the norm both in Zimbabwe and in the region. The study makes its contribution by applying the DRG for assessing and evaluating the regulatory independence gaps in developing countries like Zimbabwe.
{"title":"An Assessment of the Independence Gap in the Zimbabwean ICT Regulatory Framework","authors":"Casper Chigwedere, Sam Takavarasha, B. Chisaka","doi":"10.1109/ZCICT55726.2022.10045846","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10045846","url":null,"abstract":"This Regulatory environments are judged by their independence from both the state and from stakeholders. This paper investigates the gaps in the independence of the ICT regulatory systems in Zimbabwe using Heeks’ [1] Design Reality Gap (DRG) under an interpretivist paradigm. Data collection used in-depth interviews with purposively selected stakeholders and the analysed using NVivo (release 1.6.1 (1136)) for a thematic analysis based on DRG constructs against regulatory independence. The results show a gap of 9 emanating from a private sector which perceived it as non-independent and stateowned enterprises that saw a partially independent regulatory system. This lack of independence from the state was however believed to be the norm both in Zimbabwe and in the region. The study makes its contribution by applying the DRG for assessing and evaluating the regulatory independence gaps in developing countries like Zimbabwe.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132719532","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 : 2022-11-09DOI: 10.1109/ZCICT55726.2022.10046037
Fine Masimba, Kudakwashe Maguraushe
The world wide outbreak of the COVID-19 pandemic reconfigured various landscapes in higher education as various institutions, lecturers and students were forced to adopt e-learning. However, the successful adoption and acceptance of e-learning by both lecturers and students involved has not been discussed and measured. This study seeks to investigate the influence of both the lecturer and student self efficacy on behavoral intention to use e-learning during the pandemic in the context of the Technology Acceptance Model (TAM). Lecturer Self-Efficacy (LSE) and Student Self-Efficacy (SSE) were hypothesized to have a correlation with the perceived usefullness and perceived ease of use of e-learning systems as well as the attitude towards using those e-learning systems which resultantly influence the behavoural intention to use e-learning systems. A total of 362 questionnaires were received from both students and lecturers in Zimbabwe’s two universities and two polytechnics. Structural Equation Modelling was utilized to test the hypothesized conceptual model. Reliability and validity checks were done to the model instrument. Results indicated that both LSE and SSE have a positive influence on perceived ease of use but however, both LSE and SSE revealed a negative influence on perceived usefullness. Results also indicated that both LSE and SSE have a positive influence on attitude towards use of e-learning systems. The findings of the study contribute to the literature by highlighting the influence of LSE and SSE in the adoption and acceptance of e-learning systems in higher education.
{"title":"Understanding E-learning Adoption and Acceptance during the Covid 19 Pandemic: The Influence of Self Efficacy","authors":"Fine Masimba, Kudakwashe Maguraushe","doi":"10.1109/ZCICT55726.2022.10046037","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10046037","url":null,"abstract":"The world wide outbreak of the COVID-19 pandemic reconfigured various landscapes in higher education as various institutions, lecturers and students were forced to adopt e-learning. However, the successful adoption and acceptance of e-learning by both lecturers and students involved has not been discussed and measured. This study seeks to investigate the influence of both the lecturer and student self efficacy on behavoral intention to use e-learning during the pandemic in the context of the Technology Acceptance Model (TAM). Lecturer Self-Efficacy (LSE) and Student Self-Efficacy (SSE) were hypothesized to have a correlation with the perceived usefullness and perceived ease of use of e-learning systems as well as the attitude towards using those e-learning systems which resultantly influence the behavoural intention to use e-learning systems. A total of 362 questionnaires were received from both students and lecturers in Zimbabwe’s two universities and two polytechnics. Structural Equation Modelling was utilized to test the hypothesized conceptual model. Reliability and validity checks were done to the model instrument. Results indicated that both LSE and SSE have a positive influence on perceived ease of use but however, both LSE and SSE revealed a negative influence on perceived usefullness. Results also indicated that both LSE and SSE have a positive influence on attitude towards use of e-learning systems. The findings of the study contribute to the literature by highlighting the influence of LSE and SSE in the adoption and acceptance of e-learning systems in higher education.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121135139","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 : 2022-11-09DOI: 10.1109/zcict55726.2022.10045968
{"title":"ZCICT 2022 Cover Page","authors":"","doi":"10.1109/zcict55726.2022.10045968","DOIUrl":"https://doi.org/10.1109/zcict55726.2022.10045968","url":null,"abstract":"","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128956118","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}
The Internet of Things (IoT) is fast becoming the new normal in our everyday lives. The communication of connected devices without requiring human intervention has led to the advent of smart ecosystems or environments. Smart ecosystems are an environment where smart devices or ‘things” are trying to improve the quality of life for their inhabitants by determining the inhabitant’s intent without explicit input. This technological advancement brings with it security concerns concerning confidentiality, integrity, and availability as large data volumes are processed by smart devices. Mainstream security solutions may not work in IoT environments due to their unique nature whereby IoT has different protocols, and they have computational resource limitations. This project seeks to develop an intrusion detection system for IoT environments in an IoT network utilizing a machine learning technique whereby a user is alerted if an anomaly has been detected.
{"title":"Intrusion Detection System for IoT environments using Machine Learning Techniques","authors":"Shammah Chishakwe, Nesisa Moyo, Belinda Mutunhu Ndlovu, Sibusisiwe Dube","doi":"10.1109/ZCICT55726.2022.10045992","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10045992","url":null,"abstract":"The Internet of Things (IoT) is fast becoming the new normal in our everyday lives. The communication of connected devices without requiring human intervention has led to the advent of smart ecosystems or environments. Smart ecosystems are an environment where smart devices or ‘things” are trying to improve the quality of life for their inhabitants by determining the inhabitant’s intent without explicit input. This technological advancement brings with it security concerns concerning confidentiality, integrity, and availability as large data volumes are processed by smart devices. Mainstream security solutions may not work in IoT environments due to their unique nature whereby IoT has different protocols, and they have computational resource limitations. This project seeks to develop an intrusion detection system for IoT environments in an IoT network utilizing a machine learning technique whereby a user is alerted if an anomaly has been detected.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130702465","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 : 2022-11-09DOI: 10.1109/ZCICT55726.2022.10045969
Nosipho Mavuso, N. Jere
There is a current paradigm shift within institutions of higher learning in terms of how teaching and learning take place. This radical transition has been forced globally as a result of the Covid-19 pandemic, which saw the higher education sector implementing drastic changes in terms of remote learning. It seems there is a misunderstanding of the term blended learning based on the students’ and lecturers’ views from a selected South African university. A mixed method approach was used to collect both qualitative and quantitative data. An online questionnaire was distributed to learners to get quantitative data. On the other hand, lecturers were engaged through interviews within a workshop. An experimental approach using the university’s learning management system known as WiseUp was used. Both traditional and online modes of deliverables are common within a case. Findings show different understanding of the term blended learning and the differences in preferred activities within the blended mode. The paper provides some examples of good applications of blended learning. As higher education institutions try to embrace blended learning techniques, there is a need to engage and improve awareness of the meaning of this approach.
{"title":"Blended Learning Approach: Students’ versus Lecturers’ Views from a South African Rural University","authors":"Nosipho Mavuso, N. Jere","doi":"10.1109/ZCICT55726.2022.10045969","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10045969","url":null,"abstract":"There is a current paradigm shift within institutions of higher learning in terms of how teaching and learning take place. This radical transition has been forced globally as a result of the Covid-19 pandemic, which saw the higher education sector implementing drastic changes in terms of remote learning. It seems there is a misunderstanding of the term blended learning based on the students’ and lecturers’ views from a selected South African university. A mixed method approach was used to collect both qualitative and quantitative data. An online questionnaire was distributed to learners to get quantitative data. On the other hand, lecturers were engaged through interviews within a workshop. An experimental approach using the university’s learning management system known as WiseUp was used. Both traditional and online modes of deliverables are common within a case. Findings show different understanding of the term blended learning and the differences in preferred activities within the blended mode. The paper provides some examples of good applications of blended learning. As higher education institutions try to embrace blended learning techniques, there is a need to engage and improve awareness of the meaning of this approach.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134377409","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 : 2022-11-09DOI: 10.1109/ZCICT55726.2022.10046005
S. Dube, Admire Bhuru
Computer vision has recently been dominated by Convolutional Neural Networks (CNNs), these are a kind of Artificial Neural Networks (ANNs) mostly employed for image classification and object detection. Identifying a snake species is important when interacting with the species as well as when treating injuries due to envenoming. This task however proves to be a hurdle for the general public. This paper, therefore, sought to solve the problem of misidentification of snake species which often leads to envenoming, and mishandling of snake species by harnessing the power of CNNs together with the portability of mobile devices in developing a mobile application that identifies snake species from images almost in real-time. In implementing this system, the CNN model was trained in Google Collab on a custom-tailored dataset. The images in the dataset were sourced from the internet, and were divided into eight classes which represented eight different snake species. The images were annotated using MakeSense.ai, an online data annotation tool. After annotation the images were piped into the YOLOv5 CNN model on Google Collab for model training. The training process yielded an accuracy of 71% for all the eight classes. After training, the model was converted to a Tensorflow Lite model and exported to Android Studio IDE wherein the rest of the application was developed using Java programming language.
卷积神经网络(Convolutional Neural Networks, cnn)是近年来计算机视觉领域的主流，它是一种主要用于图像分类和目标检测的人工神经网络。识别蛇的种类是很重要的，当与该物种互动时，以及在治疗因中毒而受伤时。然而，事实证明，这项任务对公众来说是一个障碍。因此，本文试图通过利用cnn的力量和移动设备的可移植性，开发一种几乎实时地从图像中识别蛇种的移动应用程序，来解决蛇种的错误识别问题，这种问题经常导致蛇种的出现和处理不当。在实现该系统的过程中，CNN模型在谷歌Collab中进行了定制数据集的训练。数据集中的图像来自互联网，并被分为8类，代表8种不同的蛇种。使用MakeSense对图像进行注释。一个在线数据注释工具。注释后，将图像导入谷歌Collab上的YOLOv5 CNN模型中进行模型训练。训练过程对所有8个类别产生了71%的准确率。训练结束后，将模型转换为Tensorflow Lite模型并导出到Android Studio IDE，其中应用程序的其余部分使用Java编程语言开发。
{"title":"Snake Identification System Using Convolutional Neural Networks","authors":"S. Dube, Admire Bhuru","doi":"10.1109/ZCICT55726.2022.10046005","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10046005","url":null,"abstract":"Computer vision has recently been dominated by Convolutional Neural Networks (CNNs), these are a kind of Artificial Neural Networks (ANNs) mostly employed for image classification and object detection. Identifying a snake species is important when interacting with the species as well as when treating injuries due to envenoming. This task however proves to be a hurdle for the general public. This paper, therefore, sought to solve the problem of misidentification of snake species which often leads to envenoming, and mishandling of snake species by harnessing the power of CNNs together with the portability of mobile devices in developing a mobile application that identifies snake species from images almost in real-time. In implementing this system, the CNN model was trained in Google Collab on a custom-tailored dataset. The images in the dataset were sourced from the internet, and were divided into eight classes which represented eight different snake species. The images were annotated using MakeSense.ai, an online data annotation tool. After annotation the images were piped into the YOLOv5 CNN model on Google Collab for model training. The training process yielded an accuracy of 71% for all the eight classes. After training, the model was converted to a Tensorflow Lite model and exported to Android Studio IDE wherein the rest of the application was developed using Java programming language.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134419678","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 : 2022-11-09DOI: 10.1109/ZCICT55726.2022.10046032
B. Nyambo, G. Janssens, Hillary Marufu, M. Munyaradzi, Bernard Mapako, Kaitano Dzinavatonga
Jitter in multimedia traffic is mainly introduced by variations in network characteristics. Ifjitter is so significant in the application that is receiving, this can result in a degraded performance in real-time multimedia communications applications. Jitter causes inaudible audio or unclear video which can be undesirable and uncomfortable to the user. Buffers are often employed to temporarily to store arriving packets before playing them at equal intervals to curb and minimize jitter. Jitter for voice packets should not exceed 20-50 milliseconds within a given stream. This paper proposes a method to reducejitter in intermediate nodes is proposed, such that when the packets arrive at the receiving end, there would be little or nojitter to process in the de-jitter buffer. We analyze M/G/1 nonpreemptive and pre-emptive resume priority queuing in the intermediate node and analyze how these models will affect delay and jitter. We found out that preemptive priority queuing performs better but almost the same to non-preemptive priority queuing. Both Priority based queues performed much better than the First in First out queue.
{"title":"Queue Modelling and Jitter Control in Mobile Ad Hoc Networks","authors":"B. Nyambo, G. Janssens, Hillary Marufu, M. Munyaradzi, Bernard Mapako, Kaitano Dzinavatonga","doi":"10.1109/ZCICT55726.2022.10046032","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10046032","url":null,"abstract":"Jitter in multimedia traffic is mainly introduced by variations in network characteristics. Ifjitter is so significant in the application that is receiving, this can result in a degraded performance in real-time multimedia communications applications. Jitter causes inaudible audio or unclear video which can be undesirable and uncomfortable to the user. Buffers are often employed to temporarily to store arriving packets before playing them at equal intervals to curb and minimize jitter. Jitter for voice packets should not exceed 20-50 milliseconds within a given stream. This paper proposes a method to reducejitter in intermediate nodes is proposed, such that when the packets arrive at the receiving end, there would be little or nojitter to process in the de-jitter buffer. We analyze M/G/1 nonpreemptive and pre-emptive resume priority queuing in the intermediate node and analyze how these models will affect delay and jitter. We found out that preemptive priority queuing performs better but almost the same to non-preemptive priority queuing. Both Priority based queues performed much better than the First in First out queue.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"253 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115415229","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 : 2022-11-09DOI: 10.1109/ZCICT55726.2022.10046006
Wellington Tatenda Gwavava, David I. Fadaraliki, Prudence Kadebu
Virtual Technologies can provide information visualisation and manipulation, and coupling such technology with virtual mapping can provide a promotional tool for geographic locations. Interest levels of particular historical and tourist sites are low as they lack the necessary promotion to become viable destinations, thus lowering employment and revenue potential. A high number of promotional tools exist and have been employed to create awareness of tourists, however, most have not quite grasped the imagination or attention of potential tourists. This paper aims to discuss the benefits of Virtual Reality to tourism promotion and go on to analyse and formulate a model framework for how to employ virtual technologies for geographic information through Webbased services that network millions of people together can promote tourism in Zimbabwe.
{"title":"Virtual Technologies for Tourism Promotion in Zimbabwe","authors":"Wellington Tatenda Gwavava, David I. Fadaraliki, Prudence Kadebu","doi":"10.1109/ZCICT55726.2022.10046006","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10046006","url":null,"abstract":"Virtual Technologies can provide information visualisation and manipulation, and coupling such technology with virtual mapping can provide a promotional tool for geographic locations. Interest levels of particular historical and tourist sites are low as they lack the necessary promotion to become viable destinations, thus lowering employment and revenue potential. A high number of promotional tools exist and have been employed to create awareness of tourists, however, most have not quite grasped the imagination or attention of potential tourists. This paper aims to discuss the benefits of Virtual Reality to tourism promotion and go on to analyse and formulate a model framework for how to employ virtual technologies for geographic information through Webbased services that network millions of people together can promote tourism in Zimbabwe.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"244 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115559497","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}
The thyroid gland’s edge experiences an abnormal proliferation of thyroid tissue, which causes thyroid illness. The two primary types of thyroid disorders are hypothyroidism and hyperthyroidism which typically result when this gland releases excessive amounts of hormones. To identify and diagnose thyroid disease, this study suggests employing effective classifiers and feature selection strategies that consider accuracy and other performance evaluation measures. This study offers a thorough examination of various classifiers that includes the support vector machine, logistic regression, and extreme gradient boosting algorithms. The algorithms use three feature removal strategies that is recursive feature elimination, Pearson’s correlation and chi-squared statistics. To determine thyroid illness, thyroid data from the Kaggle datasets were used. Numerous aspects of the experiment have been evaluated and analyzed, including accuracy, precision, and the receiver operating curve’s area under the curve. The outcome showed that classifiers that use feature selection have a greater overall accuracy(Xtreme Gradient Boost 98%and support vector machine 95%) compared to without feature selection technique (support vector machine 89%). Logistics regression performed better without at 95% than 94% with feature selection.
{"title":"A comparative analysis of the effectiveness of feature engineering techniques on thyroid disease prediction","authors":"Aidah Mashonga, Leslie KudzaiNyandoro, Kudakwashe Zvarevashe","doi":"10.1109/ZCICT55726.2022.10045927","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10045927","url":null,"abstract":"The thyroid gland’s edge experiences an abnormal proliferation of thyroid tissue, which causes thyroid illness. The two primary types of thyroid disorders are hypothyroidism and hyperthyroidism which typically result when this gland releases excessive amounts of hormones. To identify and diagnose thyroid disease, this study suggests employing effective classifiers and feature selection strategies that consider accuracy and other performance evaluation measures. This study offers a thorough examination of various classifiers that includes the support vector machine, logistic regression, and extreme gradient boosting algorithms. The algorithms use three feature removal strategies that is recursive feature elimination, Pearson’s correlation and chi-squared statistics. To determine thyroid illness, thyroid data from the Kaggle datasets were used. Numerous aspects of the experiment have been evaluated and analyzed, including accuracy, precision, and the receiver operating curve’s area under the curve. The outcome showed that classifiers that use feature selection have a greater overall accuracy(Xtreme Gradient Boost 98%and support vector machine 95%) compared to without feature selection technique (support vector machine 89%). Logistics regression performed better without at 95% than 94% with feature selection.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"362 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122058587","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}
With the advent of websites and social media technologies, the internet has been a great method of transmitting news and information around the globe. However, due to lack of editorial scrutiny and monitoring by the media authorities, the news distribution has been seriously abused leading to the fast spreading of fake news. Fake news involves misleading broadcasting of information from sources that target to intentionally manipulate the way how people view some events or statements. This study seeks to develop a fake news detection model which will be able to analyze the title and text information attached to the news articles. From the researches done by previous scholars, it showed that the models did not perform very well because of the lack of sufficient feature extraction and fine tuning of the classification of the text. This research will therefore, articulate the gap by employing 5L-CNN deep learning model which will use inbuilt tokenizers for word embedding and will show better accuracy compared to the traditional machine learning models. In this paper, we compare machine learning algorithms (Decision trees, Random Forest, Logistic Regression & Naive Bayes) and deep learning algorithms (RNN, 5L-CNN & LSTM) to classify the authenticity of news articles. The model developed in this paper attained the best accuracy with 5L-CNN which had a result of 99.99%.
{"title":"Fake News Detection using 5L-CNN","authors":"Demo Rangarirai Collen, Leslie Kudzai Nyandoro, Kudakwashe Zvarevashe","doi":"10.1109/ZCICT55726.2022.10045981","DOIUrl":"https://doi.org/10.1109/ZCICT55726.2022.10045981","url":null,"abstract":"With the advent of websites and social media technologies, the internet has been a great method of transmitting news and information around the globe. However, due to lack of editorial scrutiny and monitoring by the media authorities, the news distribution has been seriously abused leading to the fast spreading of fake news. Fake news involves misleading broadcasting of information from sources that target to intentionally manipulate the way how people view some events or statements. This study seeks to develop a fake news detection model which will be able to analyze the title and text information attached to the news articles. From the researches done by previous scholars, it showed that the models did not perform very well because of the lack of sufficient feature extraction and fine tuning of the classification of the text. This research will therefore, articulate the gap by employing 5L-CNN deep learning model which will use inbuilt tokenizers for word embedding and will show better accuracy compared to the traditional machine learning models. In this paper, we compare machine learning algorithms (Decision trees, Random Forest, Logistic Regression & Naive Bayes) and deep learning algorithms (RNN, 5L-CNN & LSTM) to classify the authenticity of news articles. The model developed in this paper attained the best accuracy with 5L-CNN which had a result of 99.99%.","PeriodicalId":125540,"journal":{"name":"2022 1st Zimbabwe Conference of Information and Communication Technologies (ZCICT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129194035","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：