The classification performance of Bayesian Networks Classifiers: a case study of detecting Denial of Service (DoS) attacks in cloud computing environments
- Authors: Moyo, Lindani
- Date: 2015
- Language: English
- Type: Thesis , Masters , MSc (Computer Science)
- Identifier: vital:11405 , http://hdl.handle.net/10353/d1021327
- Description: In this research we propose a Bayesian networks approach as a promissory classification technique for detecting malicious traffic due to Denial of Service (DoS) attacks. Bayesian networks have been applied in numerous fields fraught with uncertainty and they have been proved to be successful. They have excelled tremendously in classification tasks i.e. text analysis, medical diagnoses and environmental modeling and management. The detection of DoS attacks has received tremendous attention in the field of network security. DoS attacks have proved to be detrimental and are the bane of cloud computing environments. Large business enterprises have been/or are still unwilling to outsource their businesses to the cloud due to the intrusive tendencies that the cloud platforms are prone too. To make use of Bayesian networks it is imperative to understand the ―ecosystem‖ of factors that are external to modeling the Bayesian algorithm itself. Understanding these factors have proven to result in comparable improvement in classification performance beyond the augmentation of the existing algorithms. Literature provides discussions pertaining to the factors that impact the classification capability, however it was noticed that the effects of the factors are not universal, they tend to be unique for each domain problem. This study investigates the effects of modeling parameters on the classification performance of Bayesian network classifiers in detecting DoS attacks in cloud platforms. We analyzed how structural complexity, training sample size, the choice of discretization method and lastly the score function both individually and collectively impact the performance of classifying between normal and DoS attacks on the cloud. To study the aforementioned factors, we conducted a series of experiments in detecting live DoS attacks launched against a deployed cloud and thereafter examined the classification performance in terms of accuracy of different classes of Bayesian networks. NSL-KDD dataset was used as our training set. We used ownCloud software to deploy our cloud platform. To launch DoS attacks, we used hping3 hacker friendly utility. A live packet capture was used as our test set. WEKA version 3.7.12 was used for our experiments. Our results show that the progression in model complexity improves the classification performance. This is attributed to the increase in the number of attribute correlations. Also the size of the training sample size proved to improve classification ability. Our findings noted that the choice of discretization algorithm does matter in the quest for optimal classification performance. Furthermore, our results indicate that the choice of scoring function does not affect the classification performance of Bayesian networks. Conclusions drawn from this research are prescriptive particularly for a novice machine learning researcher with valuable recommendations that ensure optimal classification performance of Bayesian networks classifiers.
- Full Text:
- Date Issued: 2015
- Authors: Moyo, Lindani
- Date: 2015
- Language: English
- Type: Thesis , Masters , MSc (Computer Science)
- Identifier: vital:11405 , http://hdl.handle.net/10353/d1021327
- Description: In this research we propose a Bayesian networks approach as a promissory classification technique for detecting malicious traffic due to Denial of Service (DoS) attacks. Bayesian networks have been applied in numerous fields fraught with uncertainty and they have been proved to be successful. They have excelled tremendously in classification tasks i.e. text analysis, medical diagnoses and environmental modeling and management. The detection of DoS attacks has received tremendous attention in the field of network security. DoS attacks have proved to be detrimental and are the bane of cloud computing environments. Large business enterprises have been/or are still unwilling to outsource their businesses to the cloud due to the intrusive tendencies that the cloud platforms are prone too. To make use of Bayesian networks it is imperative to understand the ―ecosystem‖ of factors that are external to modeling the Bayesian algorithm itself. Understanding these factors have proven to result in comparable improvement in classification performance beyond the augmentation of the existing algorithms. Literature provides discussions pertaining to the factors that impact the classification capability, however it was noticed that the effects of the factors are not universal, they tend to be unique for each domain problem. This study investigates the effects of modeling parameters on the classification performance of Bayesian network classifiers in detecting DoS attacks in cloud platforms. We analyzed how structural complexity, training sample size, the choice of discretization method and lastly the score function both individually and collectively impact the performance of classifying between normal and DoS attacks on the cloud. To study the aforementioned factors, we conducted a series of experiments in detecting live DoS attacks launched against a deployed cloud and thereafter examined the classification performance in terms of accuracy of different classes of Bayesian networks. NSL-KDD dataset was used as our training set. We used ownCloud software to deploy our cloud platform. To launch DoS attacks, we used hping3 hacker friendly utility. A live packet capture was used as our test set. WEKA version 3.7.12 was used for our experiments. Our results show that the progression in model complexity improves the classification performance. This is attributed to the increase in the number of attribute correlations. Also the size of the training sample size proved to improve classification ability. Our findings noted that the choice of discretization algorithm does matter in the quest for optimal classification performance. Furthermore, our results indicate that the choice of scoring function does not affect the classification performance of Bayesian networks. Conclusions drawn from this research are prescriptive particularly for a novice machine learning researcher with valuable recommendations that ensure optimal classification performance of Bayesian networks classifiers.
- Full Text:
- Date Issued: 2015
A feasibility study of wireless network technologies for rural broadband connectivity
- Authors: Twele, Nombulelo
- Date: 2013
- Subjects: Wireless communication systems -- South Africa -- Eastern Cape , Local area networks (Computer networks) -- South Africa -- Eastern Cape , Broadband communication systems -- South Africa -- Eastern Cape
- Language: English
- Type: Thesis , Masters , MSc (Computer Science)
- Identifier: vital:11396 , http://hdl.handle.net/10353/d1016167 , Wireless communication systems -- South Africa -- Eastern Cape , Local area networks (Computer networks) -- South Africa -- Eastern Cape , Broadband communication systems -- South Africa -- Eastern Cape
- Description: The adoption of wireless broadband technologies to provide network and Internet connectivity in rural communities has conveyed the possibility to overcome the challenges caused by marginalization and many other characteristics possessed by these rural communities. With their different capabilities, these technologies enable communication for rural communities internally within the community and externally on a global scale. Deployment of these technologies in rural areas requires consideration of different factors - these are in contrast, to those considered when deploying these technologies in non-rural, urban areas. Numerous research show consideration of facts for deployment of broadband technologies in urban/ non-rural environments and a little has been done in considering facts for deployment in rural environments. Hence this research aims to define guidelines for selection of broadband technologies and make recommendations on which technologies are suitable for deployment in rural communities, thereby considering facts that are true only within these rural communities. To achieve this, the research determines the metrics that are relevant and important to consider when deploying wireless broadband technology in rural communities of South Africa. It further undertakes a survey of wireless broadband technologies that are suitable for deployment in such areas. The study first profiles a list of wireless communication technologies, determines and documents characteristics of rural communities in Africa, determines metrics used to declare technologies feasible in rural areas. The metrics and rural characteristics are then used to identify technologies that are better suited than others. Informed by this initial profiling, one technology: mobile WiMAX is then selected for deployment and further evaluation. A technical review of mobile WiMAX is then carried out by deploying it at our research site in the rural, marginalized community of Dwesa (Eastern Cape, South Africa). The final section of this research provides recommendations that mobile WiMAX, LTE and Wi-Fi are the best suitable technologies for deployment in rural marginalized environments. This has been supported by extensive research and real life deployment of both Wi-Fi and mobile WiMAX. This research also recommends consideration of the following facts when seeking deployment of these technologies in rural communities: the geographical setting of the target terrain, the distances between sources and target customers and distances between target communities, weather conditions of the area, applications to be deployed over the network, social well-being of the community and their financial freedom as well.
- Full Text:
- Date Issued: 2013
- Authors: Twele, Nombulelo
- Date: 2013
- Subjects: Wireless communication systems -- South Africa -- Eastern Cape , Local area networks (Computer networks) -- South Africa -- Eastern Cape , Broadband communication systems -- South Africa -- Eastern Cape
- Language: English
- Type: Thesis , Masters , MSc (Computer Science)
- Identifier: vital:11396 , http://hdl.handle.net/10353/d1016167 , Wireless communication systems -- South Africa -- Eastern Cape , Local area networks (Computer networks) -- South Africa -- Eastern Cape , Broadband communication systems -- South Africa -- Eastern Cape
- Description: The adoption of wireless broadband technologies to provide network and Internet connectivity in rural communities has conveyed the possibility to overcome the challenges caused by marginalization and many other characteristics possessed by these rural communities. With their different capabilities, these technologies enable communication for rural communities internally within the community and externally on a global scale. Deployment of these technologies in rural areas requires consideration of different factors - these are in contrast, to those considered when deploying these technologies in non-rural, urban areas. Numerous research show consideration of facts for deployment of broadband technologies in urban/ non-rural environments and a little has been done in considering facts for deployment in rural environments. Hence this research aims to define guidelines for selection of broadband technologies and make recommendations on which technologies are suitable for deployment in rural communities, thereby considering facts that are true only within these rural communities. To achieve this, the research determines the metrics that are relevant and important to consider when deploying wireless broadband technology in rural communities of South Africa. It further undertakes a survey of wireless broadband technologies that are suitable for deployment in such areas. The study first profiles a list of wireless communication technologies, determines and documents characteristics of rural communities in Africa, determines metrics used to declare technologies feasible in rural areas. The metrics and rural characteristics are then used to identify technologies that are better suited than others. Informed by this initial profiling, one technology: mobile WiMAX is then selected for deployment and further evaluation. A technical review of mobile WiMAX is then carried out by deploying it at our research site in the rural, marginalized community of Dwesa (Eastern Cape, South Africa). The final section of this research provides recommendations that mobile WiMAX, LTE and Wi-Fi are the best suitable technologies for deployment in rural marginalized environments. This has been supported by extensive research and real life deployment of both Wi-Fi and mobile WiMAX. This research also recommends consideration of the following facts when seeking deployment of these technologies in rural communities: the geographical setting of the target terrain, the distances between sources and target customers and distances between target communities, weather conditions of the area, applications to be deployed over the network, social well-being of the community and their financial freedom as well.
- Full Text:
- Date Issued: 2013
- «
- ‹
- 1
- ›
- »