7264 research outputs found
Sort by
BAMBOOK: PERSONALIZED BOOK RECOMMENDATION AND ENGAGEMENT PLATFORM
No full textIn this report we are going to present a developed book recommendation mobile application system “BamBook”. Purpose of BamBook is to address the gap in digital platforms offering comprehensive book recommendations. Our mobile application is powered by Go, Python and Swift programming languages, while utilising Google’s YouTube Retrieval model for recommendation feature, trained on GoodReads dataset. In the process of development, we encountered challenges mostly in choosing, integrating the model, handling extensive datasets and optimising for the iOS platform. Our system architecture was improved and an iterative method was used to overcome these problems. Although our product still needs more improvements in both back and front parts, evaluation from user’s shows that BamBook meets our main objective which is to provide for users an application with easily understandable interface and relevant recommendation
ADAPTIVE NANOPORE SEQUENCING FOR INHERITED CARDIAC CONDITIONS
No full textInherited Cardiac Conditions pose a significant global health burden, described as heart structural or functional anomalies often inherited in an autosomal dominant pattern. It is also highly contributing to the mortality rates. Early diagnosis remains challenging, as most of the individuals may have asymptomatic initial stages subsequently leading to sudden cardiac death, therefore diagnostics is of high significance, especially in younger patients. Current genetic testing methods, such as Sanger sequencing and targeted panels, offer limited insight into structural genetic variations, necessitating the exploration of more comprehensive approaches. This study aims to design and study the adaptive sampling sequencing techniques, a method for targeting specific genomic regions without DNA modifications, for future use in genetic diagnosis for individuals with inherited cardiac conditions and compare the efficiency with currently used next-generation sequencing targeted panels. Through multiple sequencing runs and subsequent analysis, our findings revealed challenges in achieving consistent target region enrichment. The short sequence length observed in our study may contribute to the failure in target enrichment, highlighting the need for longer reads to improve coverage uniformity. Variant analysis using the Epi2me labs platform and annotations revealed the absence of pathogenic variants detected by Oxford Nanopore Technologies (ONT), in comparison with Illumina. This highlights the imperative for enhancing the variant calling algorithm’s sensitivity and specificity. Adaptive sampling sequencing exhibits promise for diagnosis implementation. However, additional optimization is required to improve target region enrichment, particularly through addressing the limitations posed by sequence length
FINITE-TIME OUTPUT-FEEDBACK PASSIFICATION OF UNCERTAIN FRACTIONAL-ORDER NEURAL NETWORKS HAVING TIME-VARYING DELAYS
No full textThe purpose of the study was to investigate the finite-time stability and output feedback finite-time passification of fractional order uncertain neural networks with time-varying delay. We derived conditions for finite-time boundedness and finite-time passivity using linear matrix inequality, Lyapunov functional, and Schur complement Lemma we reached finite-time stability of the system. Certainly derivation and integration for fractional order calculus, that we used for neural network system, differs from conventional integer order calculus, by its hereditary characteristics. The findings have substantial implications for the design and control of complex neural network systems, paving the way for improved robustness and reliability in real-world applications
NAVIGATING THE COMPLEXITIES OF 60 GHZ 5G WIRELESS COMMUNICATION SYSTEMS AT 60 GHZ FREQUENCY BAND FOR SECURE V2V COMMUNICATION
No full textRoad safety concerns have increased with the rising surge in vehicular traffic and can be tackled by breakthrough solutions. One of such approaches which has drawn much attention is Vehicular-to-Vehicular communication through Massive MIMO at 60 GHz mmWave technology in the 5G spectrum. Massive MIMO utilizes multiple antennas to improve spectral efficiency, throughput, coverage, energy efficiency, and reduce latency. Installation of Massive MIMO for mmWave technology entails greater complexities more so in channel estimation. The problem is solved in this paper as this work designs a sparsity adaptive algorithm that manages the tradeoff between accuracy and computational complexity. The algorithm works for real-time V2V communication taking massive MIMO into consideration and especially for 60 GHz environments. The authors conducted research from which this paper is designed to compare existing channel estimation techniques’ effectiveness in different environments. It offers prospects for improvement in V2V communication hence better road network in the traffic scenario
"MOMENTUM MAYHEM" 3D PUZZLE GAME PROTOTYPE
No full text"Momentum Mayhem" is a 3D puzzle game prototype that aims to develop physics concepts in the game environment. The project comprises four stages: ideation, pre-production, production, and post-production. The ideation phase involved analyzing popular physics-based games such as "Fall Guys" and "We Were Here" to inspire unique gameplay concepts. In the pre-production phase, the team outlined the game's requirements and designed its architecture. The production phase focused on creating and integrating essential game components, including character controllers, environmental objects, UI/UX, level maps, and multiplayer functionality. "Momentum Mayhem" leverages Unity for its game engine and Photon PUN 2 for multiplayer networking. The game features physics-based puzzles requiring cooperative gameplay, aiming to enhance problem-solving skills and teamwork. This project highlights the growing game development scene in Kazakhstan by showcasing creative gameplay and technical skills. The document further elaborates on the background, related work, project approach, execution, and evaluation of the system
BREAST CANCER DETECTION USING BAYESIAN NETWORK
No full textBreast cancer remains a major global health problem requiring effective diagnostic methods for early detection. Mammograms, although essential, may miss a significant percentage of cancer indicators. This thesis explores the potential of using machine learning techniques, specifically Bayesian Networks (BN) combined with Convolutional Neural Networks (CNN), to improve breast cancer diagnosis. The literature review highlights the urgency of improving diagnostic methods and identifies thermography as a promising, cost-effective, and non-invasive and alternative detection method. Systematically, many data sets have been collected and used to develop and train robust BN models. The so-called Model A, combining thermal images and medical records, achieved an accuracy of 84.07%, while Model B, incorporating CNN predictions on top of the datasets, achieved an accuracy of 90.9341%. These results demonstrate the potential of machine learning to transform breast cancer diagnosis, increasing accuracy and reducing the risk of misdiagnosis. Future research aims to increase dataset sizes and improve model performance, ultimately improving healthcare outcomes in breast cancer detection, in order to achieve WHO’s ultimate goal of breast self-examination (BSE)
PERFORMANCE COMPARISON OF ENSEMBLE ALGORITHMS WITH NEURAL NETWORKS BASED METHODS FOR SMALL-SIGNAL MODELING OF GaN HEMTs
No full textThis thesis investigates and compares the ensemble modeling methods and neural networks approaches for small-signal modeling of Gallium Nitride High Electron Mobility Transistors (GaN HEMTs). Specifically, ensemble methods are represented by Random Forests and eXtreme Gradient Boosting (XGBoost) algorithms, while neural networks techniques consist of Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Feed-forward Artificial Neural Networks (FFANN). To carry out the research to a higher standard, this work utilizes two distinct GaN HEMT devices. The first, a GaN HEMT grown on Diamond, is characterized by a smaller dataset and fewer modeling parameters. Conversely, the second device, a GaN HEMT on Silicon, possesses a larger dataset and a greater number of training parameters.
Furthermore, the model performance is meticulously evaluated using Mean Square Error
(MSE), Mean Absolute Error (MAE), and the coefficient of determination (R2). Findings suggest that ensemble models exhibit enhanced stability and greater robustness against overfitting. While the neural networks-based models demonstrate superior accuracy and a more streamlined development process. This research provides critical guidance for researchers and engineers in selecting the most suitable modeling approach for certain GaN HEMT devices. The choice hinges on a careful balance between prioritizing accuracy, mitigating overfitting, and managing the complexities inherent in model development
HIGH VOLTAGE INSULATOR REAL-TIME CONDITION CLASSIFICATION
No full textInsulators have a dual purpose of mechanically
supporting and electrically isolating live phase
conductors from the support tower in power
systems. Due to experiencing harsh weather
conditions, insulators may become contami
nated or damaged. As a result, electrical and
mechanical properties of insulator may deteri
orate. Thus, it is significant to automize the
process of condition classification of High Volt
age insulator to prevent accidents and there
fore ensuring secure service of transmission
lines. This paper examines the effect of pol
lution at insulator’s surfaces to the Dielectric
Dissipation Factor and therefore research was
conducted in 2 different disciplines: 1) Calculation of dielectric dissipation factor using
Ansys Maxwell software; 2) High voltage in
sulators contamination type classification us
ing Convolutional Neural Networks. Ansys
Maxwell software is used to simulate the Di
electric Dissipation Factor of insulator under
different types of contamination. In this part,
three contamination levels (light, medium and
heavy) will be considered which include soil,
cement, iron, calcium and aluminum as pollu
tants. The results show trend in which DDF
increases with an increase of contamination
level. In image classification part, Convolu
tional Neural Networks will be used to cre
ate a classifier. Insulators will be classified be
tween 4 types and these are: clean, contami
nated with soil, contaminated with water and
contaminated with snow. Validation set assess
ment results in 91% of accuracy
PREDICTING MICROPROCESSOR POWER CONSUMPTION BASED ON HARDWARE PERFORMANCE COUNTERS
No full textThis thesis presents a foundation for microprocessor power consumption estimation and prediction framework development for ARM-based devices with limited power resources using hardware performance counters. The study introduces a LSTM RNN model to estimate power consumption based on CPU HPC data without evaluation of other hardware events.. This method has a potential advantage for battery-powered embedded systems, where traditional power measurement tools have small efficiency. The research builds upon previous work in the field, highlighting the importance of energy-efficient designs in the growing IoT market. The proposed framework aims to enhance the battery life of portable devices, by helping developers to optimise the software and enabling devices with real-time power management. The model was trained on the dataset collected in idle, video recording, video streaming and audio recording scenarios and evaluated on RMSE performance. Results of the paper suggest that prediction performance of the RNN LSTM model are lacking, but the use of adaptive algorithms in a regression like evaluation have potential to be effective
FACTORS INFLUENCING ACADEMIC PERFORMANCE OF 10TH - 11TH GRADE GIRLS AT CO-EDUCATION PUBLIC SCHOOLS IN SOUTHERN KAZAKHSTAN
No full textGlobally, prioritizing girls' education has emerged as a critical focus for nations aiming to boost economic growth and human development. Girls' education has the potential to transform lives, yet substantial obstacles and problems still exist within Kazakhstan and other parts of Central Asia, limiting its full impact.
Therefore, the purpose of this study was to examine factors influencing the academic performance of recent 10th and 11th-grade female graduates from public co-educational high schools in the southern part of Kazakhstan by exploring the factors and challenges that girls face as they pursue their academic aspirations. The research questions guiding this study were: how do cultural, societal norms and expectations, socioeconomic and environmental factors influence the academic outcomes of recent female graduates, and what are the policy, practice and research implications of the findings.
This qualitative research study utilizes Vygotsky's Sociocultural Theory to gain a broader understanding of inclusivity and gender equity issues within educational settings. Data was collected through in-depth, semi-structured interviews with ten recent female graduates from co-educational public schools in southern Kazakhstan. All participants were 18 years old or older. Interviews were analyzed using thematic analysis approach. The key findings of the study sheds light on the complex interplay of cultural, socioeconomic, and environmental factors influencing the academic performance of female students in Kazakhstan. The study revealed both challenges and motivators for girls' educational success in southern Kazakhstan. High societal expectations for girls to outperform boys academically, coupled with financial concerns, create significant challenges. However, these pressures can also drive motivation and ambition. Supportive school environments, teachers who employ non-discriminatory gender-neutral teaching practices, and the prospect of state scholarships for free higher education all play pivotal roles in shaping positive academic outcomes.
The study findings could influence existing laws, regulations, guidelines, and resource allocation by providing a basis for developing gender-sensitive educational policies and practices. At the national level, Kazakhstan could lead in Central Asia by implementing policies that serve as models for gender inclusivity in education. Policies should mitigate societal expectations' pressures on girls and harness societal norms' motivation to promote educational aspirations. This approach aligns with the Sustainable Development Goals (SDGs), particularly Goal 4 (Quality Education) and Goal 5 (Gender Equality), emphasizing the importance of inclusive and equitable quality education and promoting lifelong learning opportunities for all. Internationally, these findings could contribute to global discussions on gender equality in education, influencing policy-making and decision-making processes to consider the unique cultural, societal, and economic contexts of different regions