Investigating unimodal isolated signer-independent sign language recognition
- Authors: Marais, Marc Jason
- Date: 2024-04-04
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/435343 , vital:73149
- Description: Sign language serves as the mode of communication for the Deaf and Hard of Hearing community, embodying a rich linguistic and cultural heritage. Recent Sign Language Recognition (SLR) system developments aim to facilitate seamless communication between the Deaf community and the broader society. However, most existing systems are limited by signer-dependent models, hindering their adaptability to diverse signing styles and signers, thus impeding their practical implementation in real-world scenarios. This research explores various unimodal approaches, both pose-based and vision-based, for isolated signer-independent SLR using RGB video input on the LSA64 and AUTSL datasets. The unimodal RGB-only input strategy provides a realistic SLR setting where alternative data sources are either unavailable or necessitate specialised equipment. Through systematic testing scenarios, isolated signer-independent SLR experiments are conducted on both datasets, primarily focusing on AUTSL – a signer-independent dataset. The vision-based R(2+1)D-18 model emerged as the top performer, achieving 90.64% accuracy on the unseen AUTSL dataset test split, closely followed by the pose-based Spatio- Temporal Graph Convolutional Network (ST-GCN) model with an accuracy of 89.95%. Furthermore, these models achieved comparable accuracies at a significantly lower computational demand. Notably, the pose-based approach demonstrates robust generalisation to substantial background and signer variation. Moreover, the pose-based approach demands significantly less computational power and training time than vision-based approaches. The proposed unimodal pose-based and vision-based systems were concluded to both be effective at classifying sign classes in the LSA64 and AUTSL datasets. , Thesis (MSc) -- Faculty of Science, Ichthyology and Fisheries Science, 2024
- Full Text:
- Date Issued: 2024-04-04
- Authors: Marais, Marc Jason
- Date: 2024-04-04
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/435343 , vital:73149
- Description: Sign language serves as the mode of communication for the Deaf and Hard of Hearing community, embodying a rich linguistic and cultural heritage. Recent Sign Language Recognition (SLR) system developments aim to facilitate seamless communication between the Deaf community and the broader society. However, most existing systems are limited by signer-dependent models, hindering their adaptability to diverse signing styles and signers, thus impeding their practical implementation in real-world scenarios. This research explores various unimodal approaches, both pose-based and vision-based, for isolated signer-independent SLR using RGB video input on the LSA64 and AUTSL datasets. The unimodal RGB-only input strategy provides a realistic SLR setting where alternative data sources are either unavailable or necessitate specialised equipment. Through systematic testing scenarios, isolated signer-independent SLR experiments are conducted on both datasets, primarily focusing on AUTSL – a signer-independent dataset. The vision-based R(2+1)D-18 model emerged as the top performer, achieving 90.64% accuracy on the unseen AUTSL dataset test split, closely followed by the pose-based Spatio- Temporal Graph Convolutional Network (ST-GCN) model with an accuracy of 89.95%. Furthermore, these models achieved comparable accuracies at a significantly lower computational demand. Notably, the pose-based approach demonstrates robust generalisation to substantial background and signer variation. Moreover, the pose-based approach demands significantly less computational power and training time than vision-based approaches. The proposed unimodal pose-based and vision-based systems were concluded to both be effective at classifying sign classes in the LSA64 and AUTSL datasets. , Thesis (MSc) -- Faculty of Science, Ichthyology and Fisheries Science, 2024
- Full Text:
- Date Issued: 2024-04-04
Natural Language Processing with machine learning for anomaly detection on system call logs
- Authors: Goosen, Christo
- Date: 2023-10-13
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/424699 , vital:72176
- Description: Host intrusion detection systems and machine learning have been studied for many years especially on datasets like KDD99. Current research and systems are focused on low training and processing complex problems such as system call returns, which lack the system call arguments and potential traces of exploits run against a system. With respect to malware and vulnerabilities, signatures are relied upon, and the potential for natural language processing of the resulting logs and system call traces needs further experimentation. This research looks at unstructured raw system call traces from x86_64 bit GNU Linux operating systems with natural language processing and supervised and unsupervised machine learning techniques to identify current and unseen threats. The research explores whether these tools are within the skill set of information security professionals, or require data science professionals. The research makes use of an academic and modern system call dataset from Leipzig University and applies two machine learning models based on decision trees. Random Forest as the supervised algorithm is compared to the unsupervised Isolation Forest algorithm for this research, with each experiment repeated after hyper-parameter tuning. The research finds conclusive evidence that the Isolation Forest Tree algorithm is effective, when paired with a Principal Component Analysis, in identifying anomalies in the modern Leipzig Intrusion Detection Data Set (LID-DS) dataset combined with samples of executed malware from the Virus Total Academic dataset. The base or default model parameters produce sub-optimal results, whereas using a hyper-parameter tuning technique increases the accuracy to within promising levels for anomaly and potential zero day detection. , Thesis (MSc) -- Faculty of Science, Computer Science, 2023
- Full Text:
- Date Issued: 2023-10-13
- Authors: Goosen, Christo
- Date: 2023-10-13
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/424699 , vital:72176
- Description: Host intrusion detection systems and machine learning have been studied for many years especially on datasets like KDD99. Current research and systems are focused on low training and processing complex problems such as system call returns, which lack the system call arguments and potential traces of exploits run against a system. With respect to malware and vulnerabilities, signatures are relied upon, and the potential for natural language processing of the resulting logs and system call traces needs further experimentation. This research looks at unstructured raw system call traces from x86_64 bit GNU Linux operating systems with natural language processing and supervised and unsupervised machine learning techniques to identify current and unseen threats. The research explores whether these tools are within the skill set of information security professionals, or require data science professionals. The research makes use of an academic and modern system call dataset from Leipzig University and applies two machine learning models based on decision trees. Random Forest as the supervised algorithm is compared to the unsupervised Isolation Forest algorithm for this research, with each experiment repeated after hyper-parameter tuning. The research finds conclusive evidence that the Isolation Forest Tree algorithm is effective, when paired with a Principal Component Analysis, in identifying anomalies in the modern Leipzig Intrusion Detection Data Set (LID-DS) dataset combined with samples of executed malware from the Virus Total Academic dataset. The base or default model parameters produce sub-optimal results, whereas using a hyper-parameter tuning technique increases the accuracy to within promising levels for anomaly and potential zero day detection. , Thesis (MSc) -- Faculty of Science, Computer Science, 2023
- Full Text:
- Date Issued: 2023-10-13
Evaluation of the effectiveness of small aperture network telescopes as IBR data sources
- Authors: Chindipha, Stones Dalitso
- Date: 2023-03-31
- Subjects: Computer networks Monitoring , Computer networks Security measures , Computer bootstrapping , Time-series analysis , Regression analysis , Mathematical models
- Language: English
- Type: Academic theses , Doctoral theses , text
- Identifier: http://hdl.handle.net/10962/366264 , vital:65849 , DOI https://doi.org/10.21504/10962/366264
- Description: The use of network telescopes to collect unsolicited network traffic by monitoring unallocated address space has been in existence for over two decades. Past research has shown that there is a lot of activity happening in this unallocated space that needs monitoring as it carries threat intelligence data that has proven to be very useful in the security field. Prior to the emergence of the Internet of Things (IoT), commercialisation of IP addresses and widespread of mobile devices, there was a large pool of IPv4 addresses and thus reserving IPv4 addresses to be used for monitoring unsolicited activities going in the unallocated space was not a problem. Now, preservation of such IPv4 addresses just for monitoring is increasingly difficult as there is not enough free addresses in the IPv4 address space to be used for just monitoring. This is the case because such monitoring is seen as a ’non-productive’ use of the IP addresses. This research addresses the problem brought forth by this IPv4 address space exhaustion in relation to Internet Background Radiation (IBR) monitoring. In order to address the research questions, this research developed four mathematical models: Absolute Mean Accuracy Percentage Score (AMAPS), Symmetric Absolute Mean Accuracy Percentage Score (SAMAPS), Standardised Mean Absolute Error (SMAE), and Standardised Mean Absolute Scaled Error (SMASE). These models are used to evaluate the research objectives and quantify the variations that exist between different samples. The sample sizes represent different lens sizes of the telescopes. The study has brought to light a time series plot that shows the expected proportion of unique source IP addresses collected over time. The study also imputed data using the smaller /24 IPv4 net-block subnets to regenerate the missing data points using bootstrapping to create confidence intervals (CI). The findings from the simulated data supports the findings computed from the models. The CI offers a boost to decision making. Through a series of experiments with monthly and quarterly datasets, the study proposed a 95% - 99% confidence level to be used. It was known that large network telescopes collect more threat intelligence data than small-sized network telescopes, however, no study, to the best of our knowledge, has ever quantified such a knowledge gap. With the findings from the study, small-sized network telescope users can now use their network telescopes with full knowledge of gap that exists in the data collected between different network telescopes. , Thesis (PhD) -- Faculty of Science, Computer Science, 2023
- Full Text:
- Date Issued: 2023-03-31
- Authors: Chindipha, Stones Dalitso
- Date: 2023-03-31
- Subjects: Computer networks Monitoring , Computer networks Security measures , Computer bootstrapping , Time-series analysis , Regression analysis , Mathematical models
- Language: English
- Type: Academic theses , Doctoral theses , text
- Identifier: http://hdl.handle.net/10962/366264 , vital:65849 , DOI https://doi.org/10.21504/10962/366264
- Description: The use of network telescopes to collect unsolicited network traffic by monitoring unallocated address space has been in existence for over two decades. Past research has shown that there is a lot of activity happening in this unallocated space that needs monitoring as it carries threat intelligence data that has proven to be very useful in the security field. Prior to the emergence of the Internet of Things (IoT), commercialisation of IP addresses and widespread of mobile devices, there was a large pool of IPv4 addresses and thus reserving IPv4 addresses to be used for monitoring unsolicited activities going in the unallocated space was not a problem. Now, preservation of such IPv4 addresses just for monitoring is increasingly difficult as there is not enough free addresses in the IPv4 address space to be used for just monitoring. This is the case because such monitoring is seen as a ’non-productive’ use of the IP addresses. This research addresses the problem brought forth by this IPv4 address space exhaustion in relation to Internet Background Radiation (IBR) monitoring. In order to address the research questions, this research developed four mathematical models: Absolute Mean Accuracy Percentage Score (AMAPS), Symmetric Absolute Mean Accuracy Percentage Score (SAMAPS), Standardised Mean Absolute Error (SMAE), and Standardised Mean Absolute Scaled Error (SMASE). These models are used to evaluate the research objectives and quantify the variations that exist between different samples. The sample sizes represent different lens sizes of the telescopes. The study has brought to light a time series plot that shows the expected proportion of unique source IP addresses collected over time. The study also imputed data using the smaller /24 IPv4 net-block subnets to regenerate the missing data points using bootstrapping to create confidence intervals (CI). The findings from the simulated data supports the findings computed from the models. The CI offers a boost to decision making. Through a series of experiments with monthly and quarterly datasets, the study proposed a 95% - 99% confidence level to be used. It was known that large network telescopes collect more threat intelligence data than small-sized network telescopes, however, no study, to the best of our knowledge, has ever quantified such a knowledge gap. With the findings from the study, small-sized network telescope users can now use their network telescopes with full knowledge of gap that exists in the data collected between different network telescopes. , Thesis (PhD) -- Faculty of Science, Computer Science, 2023
- Full Text:
- Date Issued: 2023-03-31
A review of the Siyakhula Living Lab’s network solution for Internet in marginalized communities
- Muchatibaya, Hilbert Munashe
- Authors: Muchatibaya, Hilbert Munashe
- Date: 2022-10-14
- Subjects: Information and communication technologies for development , Information technology South Africa , Access network , User experience , Local area networks (Computer networks) South Africa
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/364943 , vital:65664
- Description: Changes within Information and Communication Technology (ICT) over the past decade required a review of the network layer component deployed in the Siyakhula Living Lab (SLL), a long-term joint venture between the Telkom Centres of Excellence hosted at University of Fort Hare and Rhodes University in South Africa. The SLL overall solution for the sustainable internet in poor communities consists of three main components – the computing infrastructure layer, the network layer, and the e-services layer. At the core of the network layer is the concept of BI, a high-speed local area network realized through easy-to deploy wireless technologies that establish point-to-multipoint connections among schools within a limited geographical area. Schools within the broadband island become then Digital Access Nodes (DANs), with computing infrastructure that provides access to the network. The review, reported in this thesis, aimed at determining whether the model for the network layer was still able to meet the needs of marginalized communities in South Africa, given the recent changes in ICT. The research work used the living lab methodology – a grassroots, user-driven approach that emphasizes co-creation between the beneficiaries and external entities (researchers, industry partners and the government) - to do viability tests on the solution for the network component. The viability tests included lab and field experiments, to produce the qualitative and quantitative data needed to propose an updated blueprint. The results of the review found that the network topology used in the SLL’s network, the BI, is still viable, while WiMAX is now outdated. Also, the in-network web cache, Squid, is no longer effective, given the switch to HTTPS and the pervasive presence of advertising. The solution to the first issue is outdoor Wi-Fi, a proven solution easily deployable in grass-roots fashion. The second issue can be mitigated by leveraging Squid’s ‘bumping’ and splicing features; deploying a browser extension to make picture download optional; and using Pihole, a DNS sinkhole. Hopefully, the revised solution could become a component of South African Government’s broadband plan, “SA Connect”. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Muchatibaya, Hilbert Munashe
- Date: 2022-10-14
- Subjects: Information and communication technologies for development , Information technology South Africa , Access network , User experience , Local area networks (Computer networks) South Africa
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/364943 , vital:65664
- Description: Changes within Information and Communication Technology (ICT) over the past decade required a review of the network layer component deployed in the Siyakhula Living Lab (SLL), a long-term joint venture between the Telkom Centres of Excellence hosted at University of Fort Hare and Rhodes University in South Africa. The SLL overall solution for the sustainable internet in poor communities consists of three main components – the computing infrastructure layer, the network layer, and the e-services layer. At the core of the network layer is the concept of BI, a high-speed local area network realized through easy-to deploy wireless technologies that establish point-to-multipoint connections among schools within a limited geographical area. Schools within the broadband island become then Digital Access Nodes (DANs), with computing infrastructure that provides access to the network. The review, reported in this thesis, aimed at determining whether the model for the network layer was still able to meet the needs of marginalized communities in South Africa, given the recent changes in ICT. The research work used the living lab methodology – a grassroots, user-driven approach that emphasizes co-creation between the beneficiaries and external entities (researchers, industry partners and the government) - to do viability tests on the solution for the network component. The viability tests included lab and field experiments, to produce the qualitative and quantitative data needed to propose an updated blueprint. The results of the review found that the network topology used in the SLL’s network, the BI, is still viable, while WiMAX is now outdated. Also, the in-network web cache, Squid, is no longer effective, given the switch to HTTPS and the pervasive presence of advertising. The solution to the first issue is outdoor Wi-Fi, a proven solution easily deployable in grass-roots fashion. The second issue can be mitigated by leveraging Squid’s ‘bumping’ and splicing features; deploying a browser extension to make picture download optional; and using Pihole, a DNS sinkhole. Hopefully, the revised solution could become a component of South African Government’s broadband plan, “SA Connect”. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
A systematic methodology to evaluating optimised machine learning based network intrusion detection systems
- Authors: Chindove, Hatitye Ethridge
- Date: 2022-10-14
- Subjects: Intrusion detection systems (Computer security) , Machine learning , Computer networks Security measures , Principal components analysis
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/362774 , vital:65361
- Description: A network intrusion detection system (NIDS) is essential for mitigating computer network attacks in various scenarios. However, the increasing complexity of computer networks and attacks makes classifying unseen or novel network traffic challenging. Supervised machine learning techniques (ML) used in a NIDS can be affected by different scenarios. Thus, dataset recency, size, and applicability are essential factors when selecting and tuning a machine learning classifier. This thesis explores developing and optimising several supervised ML algorithms with relatively new datasets constructed to depict real-world scenarios. The methodology includes empirical analyses of systematic ML-based NIDS for a near real-world network system to improve intrusion detection. The thesis is experimental heavy for model assessment. Data preparation methods are explored, followed by feature engineering techniques. The model evaluation process involves three experiments testing against a validation, un-trained, and retrained set. They compare several traditional machine learning and deep learning classifiers to identify the best NIDS model. Results show that the focus on feature scaling, feature selection methods and ML algo- rithm hyper-parameter tuning per model is an essential optimisation component. Distance based ML algorithm performed much better with quantile transformation whilst the tree based algorithms performed better without scaling. Permutation importance performs as a feature selection method compared to feature extraction using Principal Component Analysis (PCA) when applied against all ML algorithms explored. Random forests, Sup- port Vector Machines and recurrent neural networks consistently achieved the best results with high macro f1-score results of 90% 81% and 73% for the CICIDS 2017 dataset; and 72% 68% and 73% against the CICIDS 2018 dataset. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Chindove, Hatitye Ethridge
- Date: 2022-10-14
- Subjects: Intrusion detection systems (Computer security) , Machine learning , Computer networks Security measures , Principal components analysis
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/362774 , vital:65361
- Description: A network intrusion detection system (NIDS) is essential for mitigating computer network attacks in various scenarios. However, the increasing complexity of computer networks and attacks makes classifying unseen or novel network traffic challenging. Supervised machine learning techniques (ML) used in a NIDS can be affected by different scenarios. Thus, dataset recency, size, and applicability are essential factors when selecting and tuning a machine learning classifier. This thesis explores developing and optimising several supervised ML algorithms with relatively new datasets constructed to depict real-world scenarios. The methodology includes empirical analyses of systematic ML-based NIDS for a near real-world network system to improve intrusion detection. The thesis is experimental heavy for model assessment. Data preparation methods are explored, followed by feature engineering techniques. The model evaluation process involves three experiments testing against a validation, un-trained, and retrained set. They compare several traditional machine learning and deep learning classifiers to identify the best NIDS model. Results show that the focus on feature scaling, feature selection methods and ML algo- rithm hyper-parameter tuning per model is an essential optimisation component. Distance based ML algorithm performed much better with quantile transformation whilst the tree based algorithms performed better without scaling. Permutation importance performs as a feature selection method compared to feature extraction using Principal Component Analysis (PCA) when applied against all ML algorithms explored. Random forests, Sup- port Vector Machines and recurrent neural networks consistently achieved the best results with high macro f1-score results of 90% 81% and 73% for the CICIDS 2017 dataset; and 72% 68% and 73% against the CICIDS 2018 dataset. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
An Investigation into Speaker and Headphone-Based Immersive Audio for VR and Digital Gaming Applications
- Authors: Marais, Kyle Donald
- Date: 2022-10-14
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/365246 , vital:65720
- Description: Thesis embargoed. Possible release date set for early 2024. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Marais, Kyle Donald
- Date: 2022-10-14
- Subjects: Uncatalogued
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/365246 , vital:65720
- Description: Thesis embargoed. Possible release date set for early 2024. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
Evolving IoT honeypots
- Authors: Genov, Todor Stanislavov
- Date: 2022-10-14
- Subjects: Internet of things , Malware (Computer software) , QEMU , Honeypot , Cowrie
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/362819 , vital:65365
- Description: The Internet of Things (IoT) is the emerging world where arbitrary objects from our everyday lives gain basic computational and networking capabilities to become part of the Internet. Researchers are estimating between 25 and 35 billion devices will be part of Internet by 2022. Unlike conventional computers where one hardware platform (Intel x86) and three operating systems (Windows, Linux and OS X) dominate the market, the IoT landscape is far more heterogeneous. To meet the growth demand the number of The System-on-Chip (SoC) manufacturers has seen a corresponding exponential growth making embedded platforms based on ARM, MIPS or SH4 processors abundant. The pursuit for market share is further leading to a price war and cost-cutting ultimately resulting in cheap systems with limited hardware resources and capabilities. The frugality of IoT hardware has a domino effect. Due to resource constraints vendors are packaging devices with custom, stripped-down Linux-based firmwares optimized for performing the device’s primary function. Device management, monitoring and security features are by and far absent from IoT devices. This created an asymmetry favouring attackers and disadvantaging defenders. This research sets out to reduce the opacity and identify a viable strategy, tactics and tooling for gaining insight into the IoT threat landscape by leveraging honeypots to build and deploy an evolving world-wide Observatory, based on cloud platforms, to help with studying attacker behaviour and collecting IoT malware samples. The research produces useful tools and techniques for identifying behavioural differences between Medium-Interaction honeypots and real devices by replaying interactive attacker sessions collected from the Honeypot Network. The behavioural delta is used to evolve the Honeypot Network and improve its collection capabilities. Positive results are obtained with respect to effectiveness of the above technique. Findings by other researchers in the field are also replicated. The complete dataset and source code used for this research is made publicly available on the Open Science Framework website at https://osf.io/vkcrn/. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Genov, Todor Stanislavov
- Date: 2022-10-14
- Subjects: Internet of things , Malware (Computer software) , QEMU , Honeypot , Cowrie
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/362819 , vital:65365
- Description: The Internet of Things (IoT) is the emerging world where arbitrary objects from our everyday lives gain basic computational and networking capabilities to become part of the Internet. Researchers are estimating between 25 and 35 billion devices will be part of Internet by 2022. Unlike conventional computers where one hardware platform (Intel x86) and three operating systems (Windows, Linux and OS X) dominate the market, the IoT landscape is far more heterogeneous. To meet the growth demand the number of The System-on-Chip (SoC) manufacturers has seen a corresponding exponential growth making embedded platforms based on ARM, MIPS or SH4 processors abundant. The pursuit for market share is further leading to a price war and cost-cutting ultimately resulting in cheap systems with limited hardware resources and capabilities. The frugality of IoT hardware has a domino effect. Due to resource constraints vendors are packaging devices with custom, stripped-down Linux-based firmwares optimized for performing the device’s primary function. Device management, monitoring and security features are by and far absent from IoT devices. This created an asymmetry favouring attackers and disadvantaging defenders. This research sets out to reduce the opacity and identify a viable strategy, tactics and tooling for gaining insight into the IoT threat landscape by leveraging honeypots to build and deploy an evolving world-wide Observatory, based on cloud platforms, to help with studying attacker behaviour and collecting IoT malware samples. The research produces useful tools and techniques for identifying behavioural differences between Medium-Interaction honeypots and real devices by replaying interactive attacker sessions collected from the Honeypot Network. The behavioural delta is used to evolve the Honeypot Network and improve its collection capabilities. Positive results are obtained with respect to effectiveness of the above technique. Findings by other researchers in the field are also replicated. The complete dataset and source code used for this research is made publicly available on the Open Science Framework website at https://osf.io/vkcrn/. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
Leveraging LTSP to deploy a sustainable e-infrastructure for poor communities in South Africa
- Authors: Zvidzayi, Tichaona Manyara
- Date: 2022-10-14
- Subjects: Linux Terminal Server Project , Network computers , Thin client , Fat client , Cyberinfrastructure , Poverty reduction
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/365577 , vital:65761
- Description: Poverty alleviation is one of the main challenges the South African government is facing. Information and knowledge are key strategic resources for both social and economic development, and nowadays they most often rely on Information and Communication Technologies (ICTs). Poor communities have limited or no access to functioning e-infrastructure, which underpins ICT. The Siyakhula Living Lab (SLL) is a joint project between the universities of Rhodes and Fort Hare that has been running for over 15 years now. The SLL solution is currently implemented in schools in the Eastern Cape’s Dwesa-Mbhashe municipality as well as schools in Makhanda (formerly Grahamstown). Over the years, a number of blueprints for the meaningful connection of poor communities was developed. The research reported in this thesis sought to review and improve the Siyakhula Living Lab (SLL) blueprint regarding fixed computing infrastructure (as opposed to networking and applications). The review confirmed the viability of the GNU/Linux Terminal Server Project (LTSP) based computing infrastructure deployed in schools to serve the surrounding community. In 2019 LTSP was redesigned and rewritten to improve on the previous version. Amongst other improvements, LTSP19+ has a smaller memory footprint and supports a graphical way to prepare and maintain the client’s image using virtual machines. These improvements increase the potential life of ICT projects implementing the SLL solution, increasing the participation of members of the community (especially teachers) to the maintenance of the computing installations. The review recommends the switching from thin clients deployments to full ("thick") clients deployments, still booting from the network and mounting their file systems on a central server. The switch is motivated by reasons that go from cost-effectiveness to the ability to survive the sudden unavailability of the central server. From experience in the previous deployment, electrical power surge protection should be mandatory. Also, UPS to protect the file system of the central server should be configured to start the shutdown immediately on electrical power loss in order to protect the life of the UPS battery (and make it possible to use cheaper UPS that report only on network power loss). The research study contributed to one real-life computing infrastructure deployment in the Ntsika school in Makhanda and one re-deployment in the Ngwane school in the Dwesa-Mbhashe area. For about two years, the research also supported continuous maintenance for the Ntsika, Ngwane and Mpume schools. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Zvidzayi, Tichaona Manyara
- Date: 2022-10-14
- Subjects: Linux Terminal Server Project , Network computers , Thin client , Fat client , Cyberinfrastructure , Poverty reduction
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/365577 , vital:65761
- Description: Poverty alleviation is one of the main challenges the South African government is facing. Information and knowledge are key strategic resources for both social and economic development, and nowadays they most often rely on Information and Communication Technologies (ICTs). Poor communities have limited or no access to functioning e-infrastructure, which underpins ICT. The Siyakhula Living Lab (SLL) is a joint project between the universities of Rhodes and Fort Hare that has been running for over 15 years now. The SLL solution is currently implemented in schools in the Eastern Cape’s Dwesa-Mbhashe municipality as well as schools in Makhanda (formerly Grahamstown). Over the years, a number of blueprints for the meaningful connection of poor communities was developed. The research reported in this thesis sought to review and improve the Siyakhula Living Lab (SLL) blueprint regarding fixed computing infrastructure (as opposed to networking and applications). The review confirmed the viability of the GNU/Linux Terminal Server Project (LTSP) based computing infrastructure deployed in schools to serve the surrounding community. In 2019 LTSP was redesigned and rewritten to improve on the previous version. Amongst other improvements, LTSP19+ has a smaller memory footprint and supports a graphical way to prepare and maintain the client’s image using virtual machines. These improvements increase the potential life of ICT projects implementing the SLL solution, increasing the participation of members of the community (especially teachers) to the maintenance of the computing installations. The review recommends the switching from thin clients deployments to full ("thick") clients deployments, still booting from the network and mounting their file systems on a central server. The switch is motivated by reasons that go from cost-effectiveness to the ability to survive the sudden unavailability of the central server. From experience in the previous deployment, electrical power surge protection should be mandatory. Also, UPS to protect the file system of the central server should be configured to start the shutdown immediately on electrical power loss in order to protect the life of the UPS battery (and make it possible to use cheaper UPS that report only on network power loss). The research study contributed to one real-life computing infrastructure deployment in the Ntsika school in Makhanda and one re-deployment in the Ngwane school in the Dwesa-Mbhashe area. For about two years, the research also supported continuous maintenance for the Ntsika, Ngwane and Mpume schools. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
Simplified menu-driven data analysis tool with macro-like automation
- Authors: Kazembe, Luntha
- Date: 2022-10-14
- Subjects: Data analysis , Macro instructions (Electronic computers) , Quantitative research Software , Python (Computer program language) , Scripting languages (Computer science)
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/362905 , vital:65373
- Description: This study seeks to improve the data analysis process for individuals and small businesses with limited resources by developing a simplified data analysis software tool that allows users to carry out data analysis effectively and efficiently. Design considerations were identified to address limitations common in such environments, these included making the tool easy-to-use, requiring only a basic understanding of the data analysis process, designing the tool in manner that minimises computing resource requirements and user interaction and implementing it using Python which is open-source, effective and efficient in processing data. We develop a prototype simplified data analysis tool as a proof-of-concept. The tool has two components, namely, core elements which provide functionality for the data anal- ysis process including data collection, transformations, analysis and visualizations, and automation and performance enhancements to improve the data analysis process. The automation enhancements consist of the record and playback macro feature while the performance enhancements include multiprocessing and multi-threading abilities. The data analysis software was developed to analyse various alpha-numeric data formats by using a variety of statistical and mathematical techniques. The record and playback macro feature enhances the data analysis process by saving users time and computing resources when analysing large volumes of data or carrying out repetitive data analysis tasks. The feature has two components namely, the record component that is used to record data analysis steps and the playback component used to execute recorded steps. The simplified data analysis tool has parallelization designed and implemented which allows users to carry out two or more analysis tasks at a time, this improves productivity as users can do other tasks while the tool is processing data using recorded steps in the background. The tool was created and subsequently tested using common analysis scenarios applied to network data, log data and stock data. Results show that decision-making requirements such as accurate information, can be satisfied using this analysis tool. Based on the functionality implemented, similar analysis functionality to that provided by Microsoft Excel is available, but in a simplified manner. Moreover, a more sophisticated macro functionality is provided for the execution of repetitive tasks using the recording feature. Overall, the study found that the simplified data analysis tool is functional, usable, scalable, efficient and can carry out multiple analysis tasks simultaneously. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Kazembe, Luntha
- Date: 2022-10-14
- Subjects: Data analysis , Macro instructions (Electronic computers) , Quantitative research Software , Python (Computer program language) , Scripting languages (Computer science)
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/362905 , vital:65373
- Description: This study seeks to improve the data analysis process for individuals and small businesses with limited resources by developing a simplified data analysis software tool that allows users to carry out data analysis effectively and efficiently. Design considerations were identified to address limitations common in such environments, these included making the tool easy-to-use, requiring only a basic understanding of the data analysis process, designing the tool in manner that minimises computing resource requirements and user interaction and implementing it using Python which is open-source, effective and efficient in processing data. We develop a prototype simplified data analysis tool as a proof-of-concept. The tool has two components, namely, core elements which provide functionality for the data anal- ysis process including data collection, transformations, analysis and visualizations, and automation and performance enhancements to improve the data analysis process. The automation enhancements consist of the record and playback macro feature while the performance enhancements include multiprocessing and multi-threading abilities. The data analysis software was developed to analyse various alpha-numeric data formats by using a variety of statistical and mathematical techniques. The record and playback macro feature enhances the data analysis process by saving users time and computing resources when analysing large volumes of data or carrying out repetitive data analysis tasks. The feature has two components namely, the record component that is used to record data analysis steps and the playback component used to execute recorded steps. The simplified data analysis tool has parallelization designed and implemented which allows users to carry out two or more analysis tasks at a time, this improves productivity as users can do other tasks while the tool is processing data using recorded steps in the background. The tool was created and subsequently tested using common analysis scenarios applied to network data, log data and stock data. Results show that decision-making requirements such as accurate information, can be satisfied using this analysis tool. Based on the functionality implemented, similar analysis functionality to that provided by Microsoft Excel is available, but in a simplified manner. Moreover, a more sophisticated macro functionality is provided for the execution of repetitive tasks using the recording feature. Overall, the study found that the simplified data analysis tool is functional, usable, scalable, efficient and can carry out multiple analysis tasks simultaneously. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
The implementation of a mobile application to decrease occupational sitting through goal setting and social comparison
- Authors: Tsaoane, Moipone Lipalesa
- Date: 2022-10-14
- Subjects: Sedentary behavior , Sitting position , Feedback , Mobile apps , Behavior modification , Agile software development
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/365544 , vital:65758
- Description: Background: Feedback proves to be a valuable tool in behaviour change as it is said to increase compliance and improve the effectiveness of interventions. Interventions that focus on decreasing sedentary behaviour as an independent factor from physical activity are necessary, especially for office workers who spend most of their day seated. There is insufficient knowledge regarding the effectiveness of feedback as a tool to decrease sedentary behaviour. This project implemented a tool that can be used to determine this. To take advantage of the cost-effectiveness and scalability of digital technologies, a mobile application was selected as the mode of delivery. Method: The application was designed as an intervention, using the Theoretical Domains Framework. It was then implemented into a fully functioning application through an agile development process, using Xam- arin.Forms framework. Due to challenges with this framework, a second application was developed using the React Native framework. Pilot studies were used for testing, with the final one consisting of Rhodes University employees. Results: The Xamarin.Forms application proved to be unfeasible; some users experienced fatal errors and crashes. The React Native application worked as desired and produced accurate and consistent step count readings, proving feasible from a functionality standpoint. The agile methodology enabled the developer to focus on implementing and testing one component at a time, which made the development process more manageable. Conclusion: Future work must conduct empirical studies to determine if feedback is an effective tool compared to a control group and which type of feedback (between goal-setting and social comparison) is most effective. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
- Authors: Tsaoane, Moipone Lipalesa
- Date: 2022-10-14
- Subjects: Sedentary behavior , Sitting position , Feedback , Mobile apps , Behavior modification , Agile software development
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/365544 , vital:65758
- Description: Background: Feedback proves to be a valuable tool in behaviour change as it is said to increase compliance and improve the effectiveness of interventions. Interventions that focus on decreasing sedentary behaviour as an independent factor from physical activity are necessary, especially for office workers who spend most of their day seated. There is insufficient knowledge regarding the effectiveness of feedback as a tool to decrease sedentary behaviour. This project implemented a tool that can be used to determine this. To take advantage of the cost-effectiveness and scalability of digital technologies, a mobile application was selected as the mode of delivery. Method: The application was designed as an intervention, using the Theoretical Domains Framework. It was then implemented into a fully functioning application through an agile development process, using Xam- arin.Forms framework. Due to challenges with this framework, a second application was developed using the React Native framework. Pilot studies were used for testing, with the final one consisting of Rhodes University employees. Results: The Xamarin.Forms application proved to be unfeasible; some users experienced fatal errors and crashes. The React Native application worked as desired and produced accurate and consistent step count readings, proving feasible from a functionality standpoint. The agile methodology enabled the developer to focus on implementing and testing one component at a time, which made the development process more manageable. Conclusion: Future work must conduct empirical studies to determine if feedback is an effective tool compared to a control group and which type of feedback (between goal-setting and social comparison) is most effective. , Thesis (MSc) -- Faculty of Science, Computer Science, 2022
- Full Text:
- Date Issued: 2022-10-14
Determination of speaker configuration for an immersive audio content creation system
- Authors: Lebusa, Motebang
- Date: 2020
- Subjects: Loudspeakers , Surround-sound systems , Algorithms , Coordinates
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/163375 , vital:41034
- Description: Various spatialisation algorithms require the knowledge of speaker locations to accurately localise sound in 3D environments. The rendering process uses speaker coordinates to feed into their algorithms so that they can render the immersive audio content as intended by an artist. The need to measure the loudspeaker coordinates becomes necessary, especially in environments where the speaker layouts change frequently. Manually measuring the coordinates, however, tends to be a laborious task that is prone to errors. This research provides an automated solution to the problem of speaker coordinates measurement. The solution system, SDIAS, is a client-server system that uses the capabilities provided by the Ethernet Audio Video Bridging standard to measure the 3D loudspeaker coordinates for immersive sound systems. SDIAS deploys commodity hardware and readily available software to implement the solution. A server sends a short tone to each speaker in the speaker configuration, at equal intervals. A microphone attached to a mobile device picks up these transmitted tones on the client side, from different locations. The transmission and reception times from both components of the system are used to measure the time of flight for each tone sent to a loudspeaker. These are then used to determine the 3D coordinates of each loudspeaker in the available layout. Tests were performed to determine the accuracy of the determination algorithm for SDIAS, and were compared to the manually measured coordinates. , Thesis (MSc) -- Faculty of Science, Computer Science, 2020
- Full Text:
- Date Issued: 2020
- Authors: Lebusa, Motebang
- Date: 2020
- Subjects: Loudspeakers , Surround-sound systems , Algorithms , Coordinates
- Language: English
- Type: Academic theses , Master's theses , text
- Identifier: http://hdl.handle.net/10962/163375 , vital:41034
- Description: Various spatialisation algorithms require the knowledge of speaker locations to accurately localise sound in 3D environments. The rendering process uses speaker coordinates to feed into their algorithms so that they can render the immersive audio content as intended by an artist. The need to measure the loudspeaker coordinates becomes necessary, especially in environments where the speaker layouts change frequently. Manually measuring the coordinates, however, tends to be a laborious task that is prone to errors. This research provides an automated solution to the problem of speaker coordinates measurement. The solution system, SDIAS, is a client-server system that uses the capabilities provided by the Ethernet Audio Video Bridging standard to measure the 3D loudspeaker coordinates for immersive sound systems. SDIAS deploys commodity hardware and readily available software to implement the solution. A server sends a short tone to each speaker in the speaker configuration, at equal intervals. A microphone attached to a mobile device picks up these transmitted tones on the client side, from different locations. The transmission and reception times from both components of the system are used to measure the time of flight for each tone sent to a loudspeaker. These are then used to determine the 3D coordinates of each loudspeaker in the available layout. Tests were performed to determine the accuracy of the determination algorithm for SDIAS, and were compared to the manually measured coordinates. , Thesis (MSc) -- Faculty of Science, Computer Science, 2020
- Full Text:
- Date Issued: 2020
- «
- ‹
- 1
- ›
- »