Big Data最新文献

{"title":"Topic Classification of Tweets in the Broadcasting Domain using Machine Learning Methods","authors":"T. Sefara, Mapitsi Rangata","doi":"10.1109/icABCD59051.2023.10220553","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220553","url":null,"abstract":"Twitter is one of the microblogging sites with millions of daily users. Broadcast companies use Twitter to share short messages to engage or share opinions about a particular topic or product. With a large number of conversations available on Twitter, it is difficult to identify the category of topics in the broadcasting domain. This paper proposes the use of unsupervised learning to generate topics from unlabelled tweet data sets in the broadcasting domain using the latent Dirichlet allocation (LDA) method. Approximately six groups of topics were generated and each group was assigned a label or category. These labels were used to label the data by finding the dominating label in each tweet as the main category. Supervised learning was conducted to train six machine learning models which are multinomial logistic regression, XGBoost, decision trees, random forest, support vector machines, and multilayer perceptron (MLP). The models were able to learn from the data to predict the category of each tweet from the testing data. The models were evaluated using accuracy and the f1 score. Linear support vector machine and MLP obtained better classi-fication results compared to other trained models.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74084993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Management and Monitoring of Livestock in the Farm Using Deep Learning","authors":"Makhabane Molapo, Chunling Tu, Deao Du Plessis, Shengzhi Du","doi":"10.1109/icABCD59051.2023.10220556","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220556","url":null,"abstract":"Livestock management and monitoring system play a crucial role in farm operations. This paper proposes a system for the management and monitoring of livestock on a farm using deep learning techniques. Traditional methods of monitoring livestock involve manual observation, which can be time-consuming and unreliable. Various systems have been developed, however, there are still challenges existing in present livestock classification and counting, including occlusion, animal overlapping, shadow, etc. To improve all these challenges, this paper presents a monitoring system of livestock under different conditions by the end-to-end deep learning model of You Only Look Once version 5 (YOLOv5). The suggested model conducts feature extraction on the original image with the original YOLOv5 backbone network and detects livestock of different sizes for counting on each anchor frame. Additionally, this model identifies and tracks individual animals The Kaggle dataset collected in real-time containing different animals is used as YOLOv5 relies heavily on data augmentation to improve its detection and tracking performance and validate the proposed system. The scaling, resizing, and manipulation of the splitting dataset are done by the Roboflow application. Additionally, this paper seeks to demonstrate the latest research in utilizing Faster Regional convolutional neural networks (R-CNN) and compare its backbones with the original YOLOv5 backbone. The tensor board graphs from Colab show that this proposed system outperformed other R-CNN, achieving an accuracy of 93% on mAP@_0.5%, making it a promising option for intelligent farm monitoring and managing.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79674554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Using an Exploratory Analytical Approach to Distinguish the Habits of Graduating and Non-Graduating Students in a Virtual Learning Environment","authors":"Fati Tahiru, Steven. Parbanath","doi":"10.1109/icABCD59051.2023.10220542","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220542","url":null,"abstract":"Understanding student behaviour is crucial for creating personalised learning and other interventions. Educational stakeholders continue investigating diverse solutions to improve student learning behaviour in higher educational institutions. One solution that stands out is to gain insights and identify the trends and patterns in data about students learning behaviour for decision-making. Exploratory Data Analysis (EDA) is a method for analysing and summarising data in order to get insights and recognise patterns or trends about an entity. This study seeks to utilise Exploratory Data Analysis to analyse students' logs in the virtual learning environment to distinguish the characteristics/habits of students who graduate and students who do not graduate from higher educational institutions. The process flow for implementing EDA can act as a helpful guide for educational stakeholders. The study findings indicate that the revision trend of graduated students is much more frequent than that of non-graduated students. However, there were no differences in habits in the early access to the learning materials before the start of the program. Academic stakeholders can utilise the approach to enable them to make better decisions when assessing students' behaviour and trends in the virtual environment.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80670616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Blockchain Electoral Vote Counting Solutions: A Comparative Analysis of Methods, Constraints, and Approaches","authors":"Patrick Mwansa, Boniface Kabaso","doi":"10.1109/icABCD59051.2023.10220467","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220467","url":null,"abstract":"Blockchain technology in electronic voting has emerged as an alternative to other electronic and paper-based voting systems to minimize inconsistencies and redundancies. However, past experiences indicate limited success due to scalability, speed, and privacy issues. This systematic literature review examines the methods, constraints, and approaches in the existing literature on blockchain-based electoral vote-counting solutions. A thorough search of pertinent databases was performed, and selected studies were assessed based on predefined inclusion and exclusion criteria. The review's findings reveal that most existing solutions employ smart contracts and various cryptographic algorithms to create secure and transparent voting systems. However, the study also pinpoints areas that require improvement, such as scalability, privacy, and accessibility. The review recommends exploring different combinations of blockchain platforms, cryptographic algorithms, and programming languages to develop secure and transparent voting systems. Additionally, future research could investigate the potential benefits and challenges of incorporating Internet of Things (IoT) devices, consensus mechanisms, and other technologies into the voting process. The review concludes that more research is needed to enhance the security and transparency of blockchain-based voting systems.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77640701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Model for Cyber Threat Intelligence for Organisations","authors":"Z. C. Khan, Thulile Mkhwanazi, M. Masango","doi":"10.1109/icABCD59051.2023.10220503","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220503","url":null,"abstract":"As cyber attacks are increasing in South Africa, organisations need to strengthen cyber security controls. Cyber Threat Intelligence is an essential component of a Cybersecurity program but is often overlooked. It can assist to identify future and potential cyber threats. Organisations process large volumes of data containing Cyber Threat Intelligence, but this is often not collected, processed, or considered as Cyber Threat Intelligence. South African organizations will continue to feel the repercussions of cyber-attacks if actions are not taken. To bring clarity and allow South African organizations to leverage on Cyber Threat Intelligence, this work aims to categorize Cyber Threat Intelligence for organizations. Several characteristics of Cyber Threat Intelligence are discussed, and thereafter a model is presented. The applicability of this model is demonstrated by a short use-case.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73718574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Neighbourhood Centality Based Algorithms for Switch-to-Controller Allocation in SD-WANs","authors":"Isaiah O. Adebayo, M. Adigun, P. Mudali","doi":"10.1109/icABCD59051.2023.10220485","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220485","url":null,"abstract":"The advent of artificial intelligence and big data makes it nearly impossible for large scale networks to be managed manually. To this end, software-defined networking (SDN) was introduced to provide network operators with the infrastructure for achieving greater flexibility and fine-grained control over networks. However, a critical issue to consider when incorporating SDN technology over large-scale networks like wide area networks (WANs) is the allocation of switches to controllers. In this paper, we address the switch-to-controller allocation problem that considers the heterogeneity of controller capacities. Specifically, we propose two neighbourhood centrality-based algorithms for addressing the problem with the aim of minimizing switch-to-controller latency. We also introduce a weighted centrality function that enables fair distribution of load across capacitated controllers. The proposed algorithms utilize centrality-based measures and heuristics to determine the ideal switch-to-controller allocations that consider the propagating capacity of suitable controller nodes. We evaluate the performance of the proposed algorithms on the internet2 topology. The results show that considering the heterogeneity of controller capacities reduces load imbalance significantly. Moreover, by limiting the exploration of the local centrality for each node to a maximum of two-step neighbours the complexity of the proposed algorithm is reduced. Thus, making it suitable for implementation in real-world SD-WANs.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76735105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Systematic Literature Review on Machine Learning and Laboratory Techniques for the Diagnosis of African swine fever (ASF)","authors":"Steven Lububu, Boniface Kabaso","doi":"10.1109/icABCD59051.2023.10220551","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220551","url":null,"abstract":"African swine fever (ASF) is a virulent infectious disease of pigs. It can infect domestic and wild pigs, causing severe economic and production losses. The virus can be spread through live or dead pigs and through pork products. Since there is currently no vaccine or treatment method, it poses a major challenge and threat to the pig industry once it breaks out. The results of the investigation show that most existing solutions use laboratory tests to diagnose possible ASF cases. In addition, various machine learning (ML) techniques have been used in the past to diagnose ASF. However, historical review of recent years shows that laboratories have difficulty diagnosing ASF with the required accuracy due to a lack of correlation between causes and effects. Lack of accuracy and incorrect ASF diagnoses by laboratories have proven to be a major problem for pig welfare. Consequently, misdiagnosis of ASF disease can result in severe direct and indirect economic losses to farmers, especially farmers whose income is derived primarily from pig production. While several other researchers have proposed the use of ML for ASF diagnosis, the application of cause-effect relationships between specific viruses and symptoms for ASF diagnosis is still missing. In this systematic literature review, we examine the methods, limitations, and approaches in the existing literature from ML and laboratories for ASF diagnosis. In this review, we evaluate the performance of ML and laboratory techniques for ASF diagnosis. In addition, we compare the performance of the techniques of ML with other statistical approaches such as causal ML and computer vision for ASF diagnosis. In addition, the strengths and weaknesses of ML and laboratory techniques for ASF diagnosis were summarized. A thorough search of relevant databases was performed, and the selected studies were examined using predefined inclusion and exclusion criteria. Nevertheless, the study also indicates an area for improvement, such as the accuracy of ASF diagnosis. The study recommends the use of Causal Reasoning with ML to develop a causal ML model capable of establishing relationships between viruses and symptoms to improve the accuracy of the ASF disease. The application of causal ML is presented as an alternative solution for laboratory diagnosis of ASF, which contributes to the field of the study. In addition, further research could investigate the possible characteristics of ASF, including virus variants originating from the ASF family. The review could provide essential information on ASF datasets based on the interpretation of results obtained from the use of appropriate samples and validated tests in combination with the information from laboratory tests of ASF disease epidemiology, scenario, clinical signs, and lesions produced by different virulence. This review concludes that more studies are needed for improving the accuracy and implementation of the causal ML model for ASF diagnosis in real-time surveill","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81080721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Factors Affecting the use of Smartphones for Learning: A Proposed Model","authors":"Sithembiso Dyubele, S. Soobramoney, D. Heukelman","doi":"10.1109/icABCD59051.2023.10220478","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220478","url":null,"abstract":"Increased functionalities of smartphones, such as providing easy access to the internet, have offered multiple learning opportunities, especially in a world surrounded by unprecedented periods like COVID'19. Despite the benefits of smartphones mentioned above, academics still have significant concerns about the effective utilisation of these technological devices by students for learning purposes. This paper aims to examine the factors affecting the use of smartphones for learning. The study utilised a quantitative method to pursue its aim and objectives. Data were gathered from 80 academic staff members from five Departments under the Faculty of Accounting & Informatics. A stratified sampling approach was applied to ensure a more realistic and accurate estimation of the population had been used. After applying the above approach, a simple random sampling method was used for this population according to the number of academic staff members in the above-mentioned departments. The data were analysed to ensure reliability and validity, and descriptive statistics were applied, and correlations identified to develop the proposed model. The outcomes indicate that academic staff members believe that Attitudes towards Smartphones, Facilitating Conditions, Perceived Ease of Use, Perceived Usefulness, and Performance Expectations significantly impact the use of smartphones for learning. This study was limited to academic staff from five departments of a single faculty at a South African University of Technology.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79004119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cybersecurity Practices of Rural Underserved Communities in Africa: A Case Study from Northern Namibia","authors":"G. Nhinda, Fungai Bhunu Shava","doi":"10.1109/icABCD59051.2023.10220449","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220449","url":null,"abstract":"Globally, Information Communication Technology (ICT) device usage has seen a steep rise over the last few years. This also holds in developing countries, which have embarked on connecting the unconnected or previously disadvantaged parts of their populations. This connectivity enables people to interact with cyberspace, which brings opportunities and challenges. Opportunities such as the ability to conduct business online, attend online education, and perform online banking activities. Challenges experienced are the cost of Internet access and more worrying cyber-risks and potential for exploitation. There remain pockets of communities that experience sporadic connectivity to cyberspace, these communities tend to be more susceptible to cyber-attacks due to issues of lack/limited awareness of cyber secure practices, an existent culture that might be exploited by cybercriminals, and overall, a lackluster approach to their cyber-hygiene. We present a qualitative study conducted in rural Northern Namibia. Our findings indicate that both secure and insecure cybersecurity practices exist. However, through the Ubuntu and Uushiindaism Afrocentric lenses, practices such as sharing mobile devices without passwords among the community mirror community unity. Practices such as this in mainstream research can be considered insecure. We also propose interrogating “common” secure cybersecurity practices in their universality of applicability.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72963314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"VTCGAN: A Proposed Multimodal Approach to Financial Time Series and Chart Pattern Generation for Algorithmic Trading","authors":"Joseph Tafataona Mtetwa, K. Ogudo, S. Pudaruth","doi":"10.1109/icABCD59051.2023.10220544","DOIUrl":"https://doi.org/10.1109/icABCD59051.2023.10220544","url":null,"abstract":"This paper presents a novel coupled Generative Adversarial Network (GAN) for the optimization of algorithmic trading techniques, termed Visio- Temporal Conditional Generative Adversarial Network (VTCGAN). The termed Visio- Temporal Conditional Generative Adversarial Network combines an Image Generative Adversarial Network and a Multivariate Time Series Generative Adversarial Network, offering an innovative approach for producing realistic and high-quality financial time series and chart patterns. By utilizing the generated synthetic data, the resilience and flexibility of algorithmic trading models can be enhanced, leading to improved decision-making and decreased risk exposure. Although empirical analyses have not yet been conducted, the termed Visio- Temporal Conditional Generative Adversarial Network shows promise as a valuable tool for optimizing algorithmic trading techniques, potentially leading to better performance and generalizability when applied to actual financial records.","PeriodicalId":51314,"journal":{"name":"Big Data","volume":null,"pages":null},"PeriodicalIF":4.6,"publicationDate":"2023-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84437815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}