Adol-Safety: A Framework for Empowering Parents to be Aware of Social Network Threats Affecting Adolescents
- Authors: Mjoli, Phumelela , Shibeshi, Z
- Date: 2020
- Subjects: Social networks Social media|
- Language: English
- Type: Thesis , Masters , MSc (Computer Science )
- Identifier: http://hdl.handle.net/10353/12016 , vital:39127
- Description: The use of social networks has grown so much that adolescents have become active members of various social networks such as Twitter, Facebook, and Instagram, etc. The gradual change in how people choose to communicate, socialize and share ideas today has influenced adolescents to an extent that they find themselves wanting to engage more on social networks than they really should due to peer pressure. Whenever a person joins social networks or browses the Internet, they by default are exposed and become vulnerable to many cyber threats. Cyber threats are driven by users that have negative intentions on the Internet or social networks. Adolescents are no exception to these cyber threats. The findings of this research reveal that threats such as cyberbullying, harassment, and online predators to name a few are often designed to abuse and affect adolescents). Therefore, this research aims to prevent such threats from prevailing by empowering parents to be aware of the threats that affect their adolescents in an online environment, which typically includes social networks. To achieve this, this research starts by investigating the cyber threats that affect adolescents and then explores ways that can be used to empower parents. A framework is developed to handle this. The framework includes strategies that parents can adopt and ways in which safety on social networks can be increased, as well as guidelines that can be followed in order to prevent cyber threats. The framework also aims to enhance a parent-child relationship that can help in preventing social network threats. Lastly, the framework is implemented as a knowledgesharing website that can be used by parents to receive and give an insight into social network threats that influence adolescents on social networks.
- Full Text:
- Date Issued: 2020
Developing a Machine Learning Algorithm for Outdoor Scene Image Segmentation
- Authors: Zangwa, Yamkela
- Date: 2020
- Subjects: Computational intelligence Computer science
- Language: English
- Type: Thesis , Masters , MSc (Computer Science )
- Identifier: http://hdl.handle.net/10353/12087 , vital:39150
- Description: Image segmentation is one of the major problems in image processing, computer vision and machine learning fields. The main reason for image segmentation existence is to reduce the gap between computer vision and human vision by training computers with different data. Outdoor image segmentation and classification has become very important in the field of computer vision with its applications in woodland-surveillance, defence and security. The task of assigning an input image to one class from a fixed set of categories seem to be a major problem in image segmentation. The main question that has been addressed in this research is how outdoor image classification algorithms can be improved using Region-based Convolutional Neural Network (R-CNN) architecture. There has been no one segmentation method that works best on any given problem. To determine the best segmentation method for a certain dataset, various tests have to be done in order to achieve the best performance. However deep learning models have often achieved increasing success due to the availability of massive datasets and the expanding model depth and parameterisation. In this research Convolutional Neural Network architecture is used in trying to improve the implementation of outdoor scene image segmentation algorithms, empirical research method was used to answer questions about existing image segmentation algorithms and the techniques used to achieve the best performance. Outdoor scene images were trained on a pre-trained region-based convolutional neural network with Visual Geometric Group-16 (VGG-16) architecture. A pre-trained R-CNN model was retrained on five different sample data, the samples had different sizes. Sample size increased from sample one to five, to increase the size on the last two samples the data was duplicated. 21 test images were used to evaluate all the models. Researchers has shown that deep learning methods perform better in image segmentation because of the increase and availability of datasets. The duplication of images did not yield the best results; however, the model performed well on the first three samples.
- Full Text:
- Date Issued: 2020
A Model for Intrusion Detection in IoT using Machine Learning
- Authors: Nkala, Junior Ruddy
- Date: 2019
- Subjects: Internet of things
- Language: English
- Type: Thesis , Masters , MSc (Computer Science )
- Identifier: http://hdl.handle.net/10353/17180 , vital:40863
- Description: The Internet of Things is an open and comprehensive global network of intelligent objects that have the capacity to auto-organize, share information, data and resources. There are currently over a billion devices connected to the Internet, and this number increases by the day. While these devices make our life easier, safer and healthier, they are expanding the number of attack targets vulnerable to cyber-attacks from potential hackers and malicious software. Therefore, protecting these devices from adversaries and unauthorized access and modification is very important. The purpose of this study is to develop a secure lightweight intrusion and anomaly detection model for IoT to help detect threats in the environment. We propose the use of data mining and machine learning algorithms as a classification technique for detecting abnormal or malicious traffic transmitted between devices due to potential attacks such as DoS, Man-In-Middle and Flooding attacks at the application level. This study makes use of two robust machine learning algorithms, namely the C4.5 Decision Trees and K-means clustering to develop an anomaly detection model. MATLAB Math Simulator was used for implementation. The study conducts a series of experiments in detecting abnormal data and normal data in a dataset that contains gas concentration readings from a number of sensors deployed in an Italian city over a year. Thereafter we examined the classification performance in terms of accuracy of our proposed anomaly detection model. Results drawn from the experiments conducted indicate that the size of the training sample improves classification ability of the proposed model. Our findings noted that the choice of discretization algorithm does matter in the quest for optimal classification performance. The proposed model proved accurate in detecting anomalies in IoT, and classifying between normal and abnormal data. The proposed model has a classification accuracy of 96.51% which proved to be higher compared to other algorithms such as the Naïve Bayes. The model proved to be lightweight and efficient in-terms of being faster at training and testing as compared to Artificial Neural Networks. The conclusions drawn from this research are a perspective from a novice machine learning researcher with valuable recommendations that ensure optimal classification of normal and abnormal IoT data.
- Full Text:
- Date Issued: 2019
Development of an Extensible Framework for Easy Implementation of Image Processing Applications on Android Operating System
- Authors: Gunu, Bulelani
- Date: 2019
- Subjects: Operating systems (Computers)
- Language: English
- Type: Thesis , Masters , MSc (Computer Science )
- Identifier: http://hdl.handle.net/10353/17201 , vital:40865
- Description: Image processing is a field that has been in existence for many years and it continues to grow with many other research areas adopting its use. One such research area is the area of mobile devices. Mobile devices have been equipped with image processing software and hardware so as to apply image processing features. While there are many applications of image processing and new applications have been developed, there are still many functionalities that these image processing software perform the same. The development of these software from scratch requires a lot of effort and can be time consuming. This becomes even worse for mobile device application developers, specifically Android developers, who have no knowledge of implementing image processing functionalities. This project offers a software framework which allows Android application developers to focus on their unique requirements while incorporating image processing features into their applications. The framework provides the common image processing functionalities and Android developers do not need to know the internal working of the framework in order to use it. This helps reduce application development time and effort. The framework also offers an extensibility feature which takes into consideration the future growth. This means that third party developers can keep the framework up to date with the technological advancements. The presented framework is shown to be requiring less technical expertise. Also, the way in which the system is design makes it easy to understand. This design can be adopted for other related projects that require extensible frameworks for the Android operating system.
- Full Text:
- Date Issued: 2019
A Cost-Efficient Energy Management Model for ICT4D Platforms in Low Resource Communities
- Authors: Mbotoloshi, Dumisani
- Date: 2018
- Subjects: Information technology Communication in community development|
- Language: English
- Type: Thesis , Masters , MSc (Computer Science )
- Identifier: http://hdl.handle.net/10353/14538 , vital:39999
- Description: A couple of Information and Communication Technology (ICT) researches have been conducted in the rural communities of South Africa (SA). This has encouraged the development of ICT services such as e-government, e-commerce, e-judiciary, e-health, e-agriculture and many others. Proper ICT implementation has led to societal development, economic empowerment and sustainable communities. However, this is achieved after intensive planning and stakeholders‟ involvement. It is clear that ICTs require energy to function. Most significantly is the fact that a bulk of ICTs in low resource communities (LRC) are mainly donations and less consideration if any is made on their energy consumption. LRC are regions with minimum access to energy, clean water, educational facilities, health facilities, government facilities, technology etc. There have been debates on whether in LRC ICTs are consuming energy. Most ICTs have been deployed without much consideration on the required energy consumption. On the other hand, less research has been done on the sustainability of ICTs, especially considering that all ICT devices and related platforms rely on the availability of energy. There are a lot of possible low cost energy solutions that are available, which could be used to support ICTs in LRC. However, there are so many ICT services, devices and platforms that are lying idle and not been fully utilized in such areas due to the energy related issues. In most cases, ICT solutions in LRC are introduced mainly with Computer Science and Information Systems researchers, without including other key stakeholders like energy experts. The research presents a Cost Effecient Energy Management Model that could be used to assist ICT for Development (ICT4D) platforms or projects in LRC. Both qualitative and quantitative research methods were used within the Siyakhula Living Lab (SLL) in Eastern Cape (EC) Province of SA. Design Science theory is considered for the designing of the model. The SLL is an ICT project which is undertaken within the Dwesa, Bulembu and Alice rural communities. In this dissertation, Dwesa rural community is the chosen case. The surrounding schools in the communities have been used as the ICT point of contact. Interviews, questionnaires, literature v review, action research, focus group, experiments, formulas and evaluation were the specific methods used to conduct this research. Results show that ICTs deployed in the SLLs are mainly donations which consume much energy. The main source of energy use within the SLLs is electricity. From the three schools that were selected where ICTs have been deployed it was found that the electricity bill has increased enormously. This has caused some schools not to open the computer laboratories, which are supposed to be used by the community members. It was also found that there is no energy management plan or solution in place for the SLL resources. The findings show that ICTs within the SLL are been affected by energy related subjects, though the community has been willing to pay for electricity. The Cost Efficient Energy Management Model is proposed at the end of the dissertation. The model is targeted for ICT service providers and has five (5) main pillars i.e stakeholders‟ engagement, energy and ICT infrastructure, new business models, monitoring and evaluation, and finally awareness and training. The dissertation identified all the ICT resources and services at three selected schools within the SLL and drew up the energy consumption for each school. All schools indicated that there has been an increase in the electricity bill after the introduction of the SLL project. There was a clear indication that no energy expert was involved in the establishment of the SLLs. The dissertation indicates that there is a need for renewable source of energy for ICTs to improve service delivery in LRC. ICT service providers should consider energy subjects when planning for development of such solutions in LRC. This research focuses on designing a low cost ICT energy model to inform on ICTs energy consumption in LRC. It further outlines the possible ICT energy solutions that could be used to benefit targeted communities.
- Full Text:
- Date Issued: 2018
Modelling false positive reduction in maritime object detection
- Authors: Nkele, Nosiphiwo
- Date: 20xx
- Subjects: Computer vision Neural networks (Computer science)
- Language: English
- Type: Thesis , Masters , MSc (Computer Science )
- Identifier: http://hdl.handle.net/10353/17168 , vital:40862
- Description: Target detection has become a very significant research area in computer vision with its applications in military, maritime surveillance, and defense and security. Maritime target detection during critical sea conditions produces a number of false positives when using the existing algorithms due to sea waves, dynamic nature of the ocean, camera motion, sea glint, sensor noise, sea spray, swell and the presence of birds. The main question that has been addressed in this research is how can object detection be improved in maritime environment by reducing false positives and promoting detection rate. Most of Previous work on object detection still fails to address the problem of false positives and false negatives due to background clutter. Most of the researchers tried to reduce false positives by applying filters but filtering degrades the quality of an image leading to more false alarms during detection. As much as radar technology has previously been the most utilized method, it still fails to detect very small objects and it may be applied in special circumstances. In trying to improve the implementation of target detection in maritime, empirical research method was proposed to answer questions about existing target detection algorithms and techniques used to reduce false positives in object detection. Visible images were retrained on a pre-trained Faster R-CNN with inception v2. The pre-trained model was retrained on five different sample data with increasing size, however for the last two samples the data was duplicated to increase size. For testing purposes 20 test images were utilized to evaluate all the models. The results of this study showed that the deep learning method used performed best in detecting maritime vessels and the increase of dataset improved detection performance and false positives were reduced. The duplication of images did not yield the best results; however, the results were promising for the first three models with increasing data.
- Full Text:
- Date Issued: 20xx