Repository logo
 

Faculty of Accounting and Informatics

Permanent URI for this communityhttp://ir-dev.dut.ac.za/handle/10321/1

Browse

Author: search.filters.author.Olugbara, Oludayo O.Has files: Yes

Search Results

Now showing 1 - 10 of 16
  • Thumbnail Image
    Item
    Performance of local averaging handover technique in long term evolution networks
    (SAIEE, 2015) Elujide, Israel Oludayo; Olugbara, Oludayo O.; Nepal, Thiruthlall; Owolawi, Pius Adewale
    In this paper, we investigate the performance of an alternative received signal filtering technique based on local averaging to improve the quality of handover decisions in Long Term Evolution (LTE) networks. The focus of LTE-Advance (LTE-A) networks is to provide enhanced capacity and reliability of radio access as well as broadband demand for mobile users. The necessity to maintain quality of service, especially for the delay sensitive data services and applications, has made mobility and handover decisions between the base stations in the LTE networks critical. Unfortunately, several handover decision algorithms in the LTE networks are based on the Reference Signal Received Power (RSRP) obtained as a linear averaging over the reference signals. The critical challenge with the linear averaging technique is that the limited reference signal available in the downlink packet introduces an estimation error. This estimation error is a result of the effects of linear averaging on propagation loss components in eliminating fast-fading from the received signals. Moreover, prompt and precise handover decisions cannot be based on inaccurate measurement. The standardized LTE layer 3 filtering technique is applied to the local averaged layer 1 signal to render it suitable for LTE handover decisions. The local averaging technique produces better handover than the linear averaging technique in terms of the reduced number of handover failures, improved high spectral efficiency and increased throughput, especially for cell-edge users with high speeds. The findings of this study suggest that the local averaging technique enhances mobility performance of LTE-Advance networks.
  • Thumbnail Image
    Item
    Lung cancer prediction using neural network ensemble with histogram of oriented gradient genomic features
    (Hindawi Publishing Corporation, 2015) Adetiba, Emmanuel; Olugbara, Oludayo O.
    This paper reports an experimental comparison of artificial neural network (ANN) and support vector machine (SVM) ensembles and their “nonensemble” variants for lung cancer prediction. These machine learning classifiers were trained to predict lung cancer using samples of patient nucleotides with mutations in the epidermal growth factor receptor, Kirsten rat sarcoma viral oncogene, and tumor suppressor p53 genomes collected as biomarkers from the IGDB.NSCLC corpus. The Voss DNA encoding was used to map the nucleotide sequences of mutated and normal genomes to obtain the equivalent numerical genomic sequences for training the selected classifiers. The histogram of oriented gradient (HOG) and local binary pattern (LBP) state-of-the-art feature extraction schemes were applied to extract representative genomic features from the encoded sequences of nucleotides. The ANN ensemble and HOG best fit the training dataset of this study with an accuracy of 95.90% and mean square error of 0.0159. The result of the ANN ensemble and HOG genomic features is promising for automated screening and early detection of lung cancer. This will hopefully assist pathologists in administering targeted molecular therapy and offering counsel to early stage lung cancer patients and persons in at risk populations.
  • Thumbnail Image
    Item
    Multiobjective optimization of crop-mix planning using generalized differential evolution algorithm
    (2015) Adekanmbi, Oluwole; Olugbara, Oludayo O.
    This paper presents a model for constrained multiobjective optimization of mixed-cropping planning. The decision challenges that are normally faced by farmers include what to plant, when to plant, where to plant and how much to plant in order to yield maximum output. Consequently, the central objective of this work is to concurrently maximize net profit, maximize crop production and minimize planting area. For this purpose, the generalized differential evolution 3 algorithm was explored to implement the mixed-cropping planning model, which was tested with data from the South African grain information service and the South African abstract of agricultural statistics. Simulation experiments were conducted using the non-dominated sorting genetic algorithm II to validate the performance of the generalized differential evolution 3 algorithm. The empirical findings of this study indicated that generalized differential evolution 3 algorithm is a feasible optimization tool for solving optimal mixed-cropping planning problems.
  • Thumbnail Image
    Item
    Pixel intensity clustering algorithm for multilevel image segmentation
    (Hindawi Publishing Corporation, 2015) Adetiba, Emmanuel; Oyewole, Stanley A.; Olugbara, Oludayo O.
    Image segmentation is an important problem that has received significant attention in the literature. Over the last few decades, a lot of algorithms were developed to solve image segmentation problem; prominent amongst these are the thresholding algorithms. However, the computational time complexity of thresholding exponentially increases with increasing number of desired thresholds. A wealth of alternative algorithms, notably those based on particle swarm optimization and evolutionary metaheuristics, were proposed to tackle the intrinsic challenges of thresholding. In codicil, clustering based algorithms were developed as multidimensional extensions of thresholding. While these algorithms have demonstrated successful results for fewer thresholds, their computational costs for a large number of thresholds are still a limiting factor. We propose a new clustering algorithm based on linear partitioning of the pixel intensity set and between-cluster variance criterion function for multilevel image segmentation. The results of testing the proposed algorithm on real images from Berkeley Segmentation Dataset and Benchmark show that the algorithm is comparable with state-of-the-art multilevel segmentation algorithms and consistently produces high quality results. The attractive properties of the algorithm are its simplicity, generalization to a large number of clusters, and computational cost effectiveness.
  • Thumbnail Image
    Item
    Constructing frugal sales system for small enterprises
    (AJIS, 2014-10) Olugbara, Oludayo O.; Ndhlovu, Brenda N.
    In the current study, the authors report on the application of the design science methodology to construct, utilize, and evaluate a frugal information system that uses mobile devices and cloud computing resources for documenting daily sales transactions of very small enterprises (VSEs). Small enterprises play significant roles in the socioeconomic landscape of a community by providing employment opportunities and contributing to the gross domestic product. However, VSEs have very little access to innovative information technologies that could help them manage their challenges that are restricting their effective growth, sustainability, and participation in a knowledge economy. The results of a field-evaluation experiment, involving 22 VSE entrepreneurs using a newly constructed system, MobiSales, disclosed that user behavior, which demonstrates confidence, excitement, enthusiasm, energy, and trust varied when employing a mobile electronic device for social interactions, as compared to using it for business transactions.
  • Thumbnail Image
    Item
    Experimental comparison of support vector machines with random forests for hyperspectral image land cover classification
    (Indian Academy of Sciences, 2014-06-12) Marwala, T.; Abe, B. T.; Olugbara, Oludayo O.
    The performances of regular support vector machines and random forests are experimentally com-pared for hyperspectral imaging land cover classification. Special characteristics of hyperspectral imaging dataset present diverse processing problems to be resolved under robust mathematical formalisms such as image classification. As a result, pixel purity index algorithm is used to obtain endmember spectral responses from Indiana pine hyperspectral image dataset. The generalized reduced gradient optimiza-tion algorithm is thereafter executed on the research data to estimate fractional abundances in the hyperspectral image and thereby obtain the numeric values for land cover classification. The Waikato environment for knowledge analysis (WEKA) data mining framework is selected as a tool to carry out the classification process by using support vector machines and random forests classifiers. Results show that performance of support vector machines is comparable to that of random forests. This study makes a positive contribution to the problem of land cover classification by exploring generalized reduced gra-dient method, support vector machines, and random forests to improve producer accuracy and overall classification accuracy. The performance comparison of these classifiers is valuable for a decision maker to consider tradeoffs in method accuracy versus method complexity.
  • Thumbnail Image
    Item
    Identifying critical success factors: the case of ERP systems in higher education
    (AJIS, 2014-07) Olugbara, Oludayo O.; Kalema, Billy Mathias; Kekwaletswe, Ray M.
    This paper reports on a study that uses a combination of techniques to formally characterize and determine the critical success factors influencing the effective usage of enterprise resource planning (ERP) systems, with special reference to higher education institutions. The thirty-seven ERP success factors identified from the literature are classified into: Critical, Active, Reactive and Inert categories. The classification of decision factors can generally support organizations to explore their current challenges and to adequately prepare decisions in a more participatory way for future endeavors. This study suggests a possible alternative that decision makers should take when a factor or a set of factors dominates during the implementation of ERP systems.
  • Thumbnail Image
    Item
    A hippocratic privacy protection framework for relational databases
    (IEEE, 2012-09) Oberholzer, Hendrik H.J.G.; Ojo, Sunday O.; Olugbara, Oludayo O.
    Individuals are not comfortable when disclosing their personal information to corporate organisations and are becoming increasingly concerned. Decision criteria needed for privacy protection are more complex than those that apply to access control when managing security. A typical problem in this context concerns giving individuals better control over their personal information, while at the same time allowing the organisation to process its transactions on the same personalised information. To address this difficulty, we consider extending the Hippocratic principles and model them in our Hippocratic Privacy Protection (HPP) framework that is based on the concept of privacy contracting. A prototype of the proposed HPP framework was constructed to serve as a proof of concept in order to demonstrate the developed HPP framework as an applicable and efficacious model for solving privacy problems. Based on this prototype, we afford individuals more control over their personal information. The prototype that we developed is validated against a proposed PET evaluation framework.
  • Thumbnail Image
    Item
    Kernel density feature points estimator for content-based image retrieval
    (AIRCC, 2012-02) Zuva, Tranos; Olugbara, Oludayo O.; Ojo, Sunday O.; Ngwira, Seleman M.
    Research is taking place to find effective algorithms for content-based image representation and description. There is a substantial amount of algorithms available that use visual features (color, shape, texture). Shape feature has attracted much attention from researchers that there are many shape representation and description algorithms in literature. These shape image representation and description algorithms are usually not application independent or robust, making them undesirable for generic shape description. This paper presents an object shape representation using Kernel Density Feature Points Estimator (KDFPE). In this method, the density of feature points within defined rings around the centroid of the image is obtained. The KDFPE is then applied to the vector of the image. KDFPE is invariant to translation, scale and rotation. This method of image representation shows improved retrieval rate when compared to Density Histogram Feature Points (DHFP) method. Analytic analysis is done to justify our method, which was compared with the DHFP to prove its robustness.
  • Thumbnail Image
    Item
    Introducing an adaptive kernel density feature points estimator for image representation
    (IJITCS, 2012-06) Zuva, Tranos; Olugbara, Oludayo O.; Ojo, Sunday O.; Ngwira, Seleman M.
    This paper provides an image shape representation technique known as Adaptive Kernel Density Feature Points Estimator (AKDFPE). In this method, the density of feature points within defined rings (bandwidth) around the centroid of the image is obtained in the form of a vector. The AKDFPE is then applied to the vector of the image. AKDFPE is invariant to translation, scale and rotation. This method of image representation shows improved retrieval rate when compared to Kernel Density Feature Points Estimator (KDFPE) method. Analytic analysis is done to justify our method, which was compared with the KDFPE to prove its robustness.