Pusat Pengajian Sains Komputer - Tesis
Browse
Browsing Pusat Pengajian Sains Komputer - Tesis by Type "doctoral thesis"
Now showing 1 - 20 of 24
Results Per Page
Sort Options
- PublicationA Multi-objective Evolutionary Algorithm Based On Decomposition For Continuous Optimization Using A Step-function Technique(2022-06)Chuah, How SiangMulti-objective optimization is an area of study which solves complex real-world problem that involves two or three objectives. Multi-objective Evolutionary Algorithm based on Decomposition (MOEA/D) is one of the algorithms that utilize the concepts of decomposition and neighbourhood to solve multi-objective problems. One of the recent MOEA/D algorithms, i.e., Constant-distance based Neighbours for MOEA/D with Dynamic Weight Vector Adjustment (MOEA/D-AWACD), integrates the concept of a constant-distance neighbourhood and a dynamic weight vector design. This combination creates a flexible neighbourhood that can adapt to the weight vectors changes. However, MOEA/D-AWACD’s performance is dependent on a constant-distance parameter,
- PublicationAn Efficient Distributed Slotted Multi-Hop Wireless Mac Protocol For Internet Of Things(2021-11)Mohammad Ali SarvghadiIn Wireless Sensor Networks (WSNs), the Media Access Control (MAC) layer is responsible for coordinating Sensor Nodes (SNs) as their access to the media. Real time data flow needs precise scheduling for sensor nodes to access the medium in order to send the data on time. Time slotted transmission scheduling is needed to access the media without collision. To achieve this, sensor nodes need to be synchronized and know about each other’s notion of time. However, time synchronization requires message exchanges, which incur high overhead. The proposed MAC protocol, which consists of algorithms that minimizes collision and detects hidden terminals, eliminates the message overheads of global time synchronization
- PublicationClassification Of Cpg Island And Promoter Regions Using Rare K-Mer Motifs(2015-07)Mohamed Hashim,Ezzeddin KamilEmpirical analysis on DNA k-mers is proven to be an effective means to discover functional elements in the human genomes. Among the empirical works, “rare k-mer" (RKM) is a very interesting subject to be studied due to their unique sequence properties. RKMs were referred as DNA k-mers (of k=7 to 11) that have a low frequency in k-mer frequency distributions of mammalian genomes, yet there are large variations of their mass at the lower k-mer spectra. Our first objective is to discover RKM motifs in the human genome; to identify potential RKM computational applications in biology; and to infer representations for the identified applications. In short, the first goal was achieved by using comparative strategy and several bioinformatic tools (of UCSC browsers, EpiGRAPH, and Galaxy) which correlated RKMs with several genomic features of CpG Islands (CGIs), promoter, 5’ Un- Translated Regions (5’UTRs), and open chromatin regions; and by using intrinsic string mining approaches which identified several unique RKM topological, compositional, and clustering properties in the correlated genomic features.
- PublicationCommunication Efficient Decentralized Collaborative Learning Of Heterogeneous Deep Learning Models Using Distillation And Incentive-Based Techniques(2023-07)Zahid, IqbalSmart devices, collectively, have very valuable and real-time data which can be used to train very efficient deep learning models for AI applications. However, due to the sensitive nature of this data, people are more concerned about the privacy of their data and not willing to share it. Therefore, there is a need to learn from this valuable data in a decentralized fashion by withholding data localized on these intended devices and efficiently performing necessary computation on these devices by exploiting their computational resources. Statistical heterogeneity and fully model heterogeneity are among the key challenges in applying the Decentralized Learning (DL) approaches in real scenarios. Typically, all existing DL techniques assume that all devices would have homogeneous model architecture. However, in real applications of DL, due to different computational resources and distinct business needs of devices, it is intuitive that they may have completely different model architectures. Very limited work has been performed to address fully model heterogeneity problem. In the same way, some work has been performed to address the statistical heterogeneity however mostly is hard to apply in real scenarios or is only for limited use cases.
- PublicationConcept And Relation Extraction Framework For Ontology Learning(2023-08)Al-Aswadi, Fatima Nadeem SalemExtracting valuable knowledge and representing it in a machine-understandable form is considered one of the main challenges of semantic web and knowledge engineering fields. The explosive growth of textual data is coupled with the increasing demand for ontologies. Ontology Learning (OL) from text is a process that aims to automatically or semi-automatically extract and represent the knowledge from text into the machine-understandable form. Ontology is a core scheme representing knowledge as a set of concepts and their relationships within a domain. Extracting the concepts and their relations is the backbone for an OL system. The existing OL systems have many limitations and drawbacks, such as not efficient for extracting relevant concepts especially, for large-length dataset; depending on a large amount of predefined patterns to extract relations, and extracting very few types of relations. In this thesis, a framework called Concept and Relation Extraction Framework (CREF) is proposed. It consists of four main stages: enhancing pre-processing method, developing methodology for the concept extraction task, proposing a new text representation approach for relations, and improving relation extraction method. The first stage involves proposing a new Concept Extraction stopwords (CE-stopwords) for scientific publications while the second stage involves introducing a new Domain Time Relevance (DTR) metric and proposing a Developed Concept Extraction Method based on DTR called (DTR-DCEM).
- PublicationData Integrity For Cloud Computing With Homomorphic Encryption(2022-08)Awadallah, Ruba N SCloud computing is a new computing model in which resources are provided as general utilities that users can lease through the Internet on-demand fashion. How- ever, this technology has numerous data security concerns. Asymmetric Homomorphic cryptosystem has been acknowledged as one of the potential solutions for achieving secure cloud computing since it can provide data privacy and confidentiality. Homo- morphic Encryption (HE) is a cryptosystem that allows cloud computing to operate computations on encrypted data. However, HE schemes are non-compliance to indis- tinguishability under adaptive chosen-ciphertext attack (IND-CCA2) because of their malleable nature. Moreover, the client (data owner) cannot prove if the data has been manipulated by the Cloud Service Provider (CSP) since CSP has absolute authority over the client’s data. The client is also unable to trace the processes that are being applied to the data once the data is outsourced. The client can also not verify the CSP’s reliability in applying the required operations on the required data. Therefore, implementing HE alone is not sufficient against various data integrity attacks. Two dif- ferent verifiable cloud computing schemes are being proposed in this work to address these questions. The first verifiable cloud computing scheme uses modular arithmetic to produce verified data to verify the CSP’s HE computations over a finite integer field. The performance of the proposed scheme varied based on the underlying cryp- tosystems used. However, based on the tested cryptosystems, the scheme has 1.5% storage overhead and a computational overhead that can be configured to work below 1%.
- PublicationEnhanced Automated Framework For Cattle Tracking And Classification(2022-09)Williams, Bello RotimiEmploying computer vision-based methods in monitoring individual cows has become what researchers are striving for. Computer vision-based methods could be used to monitor each individual cows. The accuracy of the existing methods and frameworks is below expectation in handling these tasks. Moreover, they can still be improved to achieve better and more accurate results. The goal of this research is to provide a framework for better cattle tracking and classification systems. An enhanced object tracking algorithm (PFtmM) that integrates enhanced particle filter algorithm (PFtm) with mean-shift tracker (M) is proposed and deployed as first step to address the problems arise due to occurrence of occlusion and non-linear movement of cow objects in video. The integration of particle filter with mean-shift tracker considers the following techniques: (1) temporary memory for keeping tracks of occluded cow objects; (2) supplementing each algorithm’s weakness by the strength of the other for tracking non-linear movement. Strength of particle filter (PF) is its non-linearity property which it uses to track object’s non-linear movement but, with high computational time and search range as its weakness. Temporary memory (tm) strength is its ability to track full occlusion with reduced computational time and search range. Mean-shift strength is its sensitivity to object’s movement and colour distribution by using similarity function but, with inability to track object’s non-linear movement and full occlusion as its weakness.
- PublicationEnhanced Bluetooth Low Energy 5 Aodv-Based Mesh Communication Protocol With Multipath Support(2023-04)Muhammad Rizwan GhoriWireless Ad-Hoc Networks (WAHN) are growing more widespread due to improvements in Internet of Things (IoT) technologies. Several WAHN technologies, including ZigBee, Z-Wave, Threads, Bluetooth Low Energy (BLE) are available. Despite progress on IoT, network-based routing continues to be a critical problem that has to be tackled. In view of the above, this study focuses on a BLE-based mesh network.
- PublicationEnhanced Conditional Generative Adversarial Network For Handling Subject Variability In Human Activity Recognition(2023-07)Jimale, Ali OlowWhile splitting datasets, researchers assume that training set is exchangeable with test set and expect good classification performance. This assumption is invalid due to subject variability due to age differences. Classification models trained on activity data from one particular age group such as adults cannot generalize to activity data collected from a different age group such as elderly. Subject variability in the context of age is a valid problem that degrades the performance of activity recognition, but remains unexplored. Existing studies that investigated subject variability in activity recognition overlooked this problem and only focused on contextual subject variability, and intra-subject variability. This study investigates the effects of subject variability on the performance of sensor-based activity recognition. Elderly and adult datasets were used to evaluate the assessment techniques. Experiments on adult dataset only, experiments on elderly dataset only, and experiments on adult (as training) and elderly (as test) datasets were conducted using machine learning and deep learning. The results show a significant performance drop in activity recognition on different subjects with different age groups. On average, the drop in recognition accuracy is 9.75% and 12% for machine learning and deep learning models respectively. Conditional Generative Adversarial Network (CGAN) is an ideal solution to address subject variability.
- PublicationEnhanced Heterogeneous Stacked Ensemble Machine Learning Model For Detecting Nigerian Politically Motivated Cyberhate(2023-03)Sallau, Mullah NanlirHate speech is a universal problem from time immemorial. The high adoption of social media (SM) has made it a problem of gigantic proportions during elections in Nigeria. The anonymity enjoyed by the users is the main reason for the high volume of cyber hate in Nigeria's social media space. Politicians usually circulate different politically motivated hate messages on social media during elections. Though, different artificial intelligence (AI) approaches such as machine learning models have been developed to address the problem with reasonable success. Nonetheless, the problem persists and leads to a high rate of cyberhate crime in Nigeria. The main problem is the lack of research to build models to address peculiarities in Nigeria. These problems made existing models incapacitated in Nigeria's cyberspace. To solve the identified research gaps from the vantage point of a machine learning researcher, the problem was modelled as a text classification task. To achieve the main objective, the study proposed to enhance a technique called the stacking ensemble method. The proposed method is called the heterogeneous stacked ensemble (HSE).
- PublicationEnhanced Se-Resnet 101 For Food Image Segmentation(2022-12)Suhaila Farhan Ahmad AbuowaidaIn recent years, deep learning has demonstrated its usefulness and capability in computer vision due to its high accuracy and acceptability. This thesis focuses on the enhanced instance segmentation method for multiple types of food and the improved food volume estimation method for better food calorie estimation.
- PublicationEnhancing Svd-Based Image Watermarking Strategies Based On Digital Chaos(2022-08)Wafa’hamdan Suleiman AlshouraA digital image is a universal medium that carries sensitive information and has proliferated in recent years. The watermarking scheme is a technique used for protecting digital images and other content such as audio, video, and text. Image watermarking schemes have the ability to embed the owner’s information into a host image in an imperceptible manner, and can be extracted later in the detection phase. Recently, the hybrid singular value decomposition (SVD)-based watermarking schemes in the frequency domain have received considerable attention. The interest is as a result of SVD having stability and robust properties which makes it resistant to different well-known attacks. However, existing hybrid SVD schemes do not meet some critical watermarking requirements such as successful trade-offs between robustness and imperceptibility, large capacity, and high security. Hence, they produce ineffective results which are not robust and are prone to a variety of attacks. This study aims to bridge the gap by developing enhanced hybrid SVD-based image watermarking schemes to fulfil the aforementioned watermarking requirements. In the proposed schemes, random numbers and new embedding strategies are leveraged upon to address these issues as well as making the proposed schemes flexible and easy to implement. This study proposes three new schemes that can be implemented on different image formats (gray and color image). The design elements and the novel constructions incorporated in the proposed schemes makes sure that they surpass the existing schemes.
- PublicationEthical Framework On Breast Self-Examination System(2022-09)Khana, RajesBreast Cancer is the leading cause of mortality for females in Indonesia. There is a need for women in Indonesia to have a breast self-examination system (BSE) where they can seek information and consult their cases with any physician; However, violation of patient confidentiality, lack of patient trust in physician conduct, and lack of patient control over their data pose a major problem in the use of BSE system. For this reason, there is a need to design a BSE system that can track self-exam data and allow communication between patient and physician. In developing a BSE system, ethical and trustworthiness values should exist as a part of the system’s ability. A survey was conducted to identify the ethical and trustworthiness values that can be applied to the BSE system. Therefore, this research aims to identify ethical values, determine the suitable trustworthiness factors, and propose an ethical framework that suits the development of the BSE system. The explanatory mixed-method carried quantitative survey with 772 respondents. A qualitative focus group discussion is conducted with 32 participants leading to the importance of an ethical framework being used in the healthcare system. The findings indicate that ethical values positively influence the BSE system with a pvalue < 0.001, trustworthiness positively influences the BSE system with a p-value < 0.001, and an ethical framework is essential to implement in the BSE system due to the coefficient of determination (R2) = 0.509.
- PublicationExtension Of Information Systems Success Model For The Gamification E-Learning System(2022-10)Al-Muwassefi, Hala Najwan SabehThe existing literature depicts that graduates’ skills gap between industry expectations and academic preparation has become a trending phenomenon worldwide. Universities have been called to play a vital role to instil in these students the most sought skills in surviving this 21st Century. Hence, it signals the need to propose a new innovative system to enhance students’ skills and examine the net benefits of this system and the predictors of net benefits among university students. Accordingly, this research proposes the Login Career System, a gamification e-learning system, to improve students’ skills by integrating gamification features and certified e-learning courses. Besides, based on the integration of the updated DeLone & McLean Information Systems Success Model and the philosophy of expectancy theory, this study builds a theoretical model to govern the investigation of information quality, system quality, service quality, and collaboration quality as predictors of net benefits, use, and user satisfaction. Additionally, the model is built to investigate the use and user satisfaction as predictors of net benefits, whereas the use of the system is also examined as a predictor of user satisfaction. The research model was extended to include perceived future skills, perceived future personal characteristics, and perceived future labour market knowledge as the three outcome factors generated from the net benefits.
- PublicationFactors Influencing The Adoption Of Internet Of Things For Home Security In Dhaka Households(2023-05)Mahmud, ArifGlobally, burglary crimes in some countries are effectively minimized by using advanced security technology supported by the Internet of Things (IoT). However, these crimes have caused rising concerns among residents in Dhaka. Despite the implementation of appropriate rules and technology development, IoT adoption in Bangladesh remains relatively low. Therefore, it is imperative to determine the factors that influence the intention to adopt IoT for home security. Accordingly, the Protection Motivation Theory (PMT) and Attitude-Social Influence-Self-Efficacy model (ASE) are integrated along with two moderating variables, personal innovativeness, and perceived trust to address the research objectives. This research implements a quantitative approach to analyze the primary data of 348 participants. Based on the findings, perceived severity, perceived vulnerability, response efficacy, response cost, and attitude are factors that impact the intention to adopt IoT with a variance of 34.9%. Moreover, the R2 and Q2 are improved respectively, due to the inclusion of personal innovativeness and perceived trust as moderators. Personal innovativeness positively moderates the relationships between response efficacy–intention, and self-efficacy–intention, and negatively moderates the attitude–intention relationship.
- PublicationLandmark Image Discovery Using Network Clustering(2022-03)Mohammed Al-Zou’Bi, Ala’A AhmedSignificant amounts of Internet photo collections are stored online and continue to grow rapidly. This wealth and availability of visual information enable the development of several computer vision applications. Therefore, there is a need for efficient techniques for structuring and organizing this large number of images. In particular, landmark images form a large portion of such collections. Mining of landmark images relies on clustering to group large-scale image collections by the object they depict. The grouping process is a very challenging task due to the variations in the object’s appearance, which can be caused by illumination conditions, differences in scale and imaging viewpoint.
- PublicationModel Of Cloud Accounting Adoption Among Indonesian Micro, Small And Medium Enterprises (Msmes)(2022-08)Ferdinand Murni, HamunduCloud accounting (CA) is an accounting information system that fits the characteristics of micro small and medium enterprises (MSMEs) for management accounting. MSMEs generally low uptake of management accounting and thus lead to the failure of the business. Meanwhile, MSMEs are considered the major contributor to the economic growth of most nations including Indonesia. This study has examined the effect of the Technololgy-Organization-Environment Framework (TOE) and Technology Acceptance Model (TAM) on the intention to adopt CA. The independent variables consist of cloud computing characteristics, organizational readiness, and environmental context such as industrial & market, mimetic pressures, as well as government intervention. In addition, the moderator variable consisted of perceived ease of use (PEU) and perceived usefulness (PU), whereas trust in the internet (TI and trust in system reliability (TS) as moderator variables. A quantitative method with non-probability purposeful sampling has been employed in this study. The population is MSMEs in Indonesia as GoFood merchants, with data collected of 345 responses. The result found PEU and PU have a mediating role between cloud computing characteristics, organizational readiness, industrial & market, mimetic pressures, and intention to adopt CA. In addition, TS has significantly moderated the relation between PEU and intention to adopt CA, while PU has been rejected. In short, the ownersmanagers of MSMEs are not concerned with the advanced internet, but more concerned with PEU, and TS. This study has highlighted the results and attempted to justify them with logic supported by the relevant literature.
- PublicationModification Of Regression Models To Solve Heterogeneity Problem Using Seaweed Drying Data(2023-09)Joshua, Ibidoja OlayemiDuring the seaweed’s drying process, a lot of drying parameters are involved. One of the problems in regression analysis is the impact of heterogeneity parameters. The seaweed data was collected using sensor-smart farming technology attached to the v-Groove Hybrid Solar Drier. The proposed method used the variance inflation factor to identify the heterogeneity parameters. To determine the 15, 25, 35, and 45 highranking important parameters for the seaweed, models such as ridge, random forest, support vector machine, bagging, boosting, LASSO, and elastic net are used before heterogeneity, after heterogeneity, and for the modified model. To reduce the outliers, robust regressions such as M Huber, M Hampel, M Bi Square, MM, and S estimators are used. Before the heterogeneity parameters were excluded from the model, the hybrid model of the ridge with the M Hampel estimator showed that better significant results were obtained with 2.14% outliers. After the heterogeneity parameters were excluded from the model, the support vector machine with the MM estimator showed that better significant results were obtained with 2.09% outliers. For the modified model, LASSO with M Bi square estimator showed that better significant results were obtained with 1.31% outliers. For future studies, the impact of heterogeneity using a hybrid model with imbalanced data or missing values can be investigated. Ensemble machine learning algorithms such as stacking, XGBoost, and AdaBoost can be used.
- PublicationMulti-fish Detection And Tracking Using Track-mask Region Convolutional Neural Network(2023-09)Alshdaifat, Nawaf Farhan FankurDeep learning has become more common in recent years due to its excellent results in many areas. This thesis primarily focuses on multi-fish detection and tracking methods in underwater videos. The existing multi-fish detection methods for underwater videos have a low detection rate and consumes time in the training and testing process due to the underwater conditions and the overfitting during training. Many multi-fish detection and tracking methods for underwater videos (based on deep learning) where low accuracy for multi-fish tracking and occlusion instances during multi-fish tracking leads to inability to distinguish edges, and inability to handle each detected object over time. Therefore, this research aims to improve and enhance methods for multi-fish detection and tracking in underwater videos based on the latest deep learning algorithms. The proposed improved multi-fish detection method involves three main steps: 1) Improving ResNet-101 backbone for better fish detection, 2) Enhancing the Region Proposal Network (RPN) method based on Faster R-CNN for multi-fish detection and 3) An improved multi-fish detection method in terms of accuracy and with a lower training and testing times by utilising the aforementioned methods. The proposed multi-fish tracking method (Track-Mask R-CNN) also exhibits similar enhanced characteristics compared to the state-of-art methods (using fish dataset). An accuracy of 86.7% and 78.9% have been achieved for the proposed multi-fish detection and tracking respectively.
- PublicationPartial Verification Bias Correction In Diagnostic Accuracy Studies Using Propensity Score-Based Methods(2023-06)Wan Nor Arifin Bin Wan MansorThis research objective was to design PS-based methods with weighted regression and resampling approaches to improve and extend PVB correction under MAR and MNAR assumptions. Three MAR PS-based methods of PVB correction were proposed: 1) Inverse probability weighted logistic regression (IPWLogReg), 2) Scaled inverse probability weighted resampling (SIPW-Resamp) and 3) Scaled inverse probability weighted balanced resampling (SIPW-BalResamp).