Welcome to SDU Institutional Repository
Communities
Select a community to browse its collections.
Recent Submissions
Study of the transformation of Kazakh language speech into text data
(Faculty of Engineering and Natural Science, 2024) Kursabayeva A.
The transformation of speech into text data is a key component in the development of modern language technologies and artificial intelligence. Despite significant advances in this field, support for languages with unique grammatical and phonetic characteristics, such as Kazakh, remains a challenge. The purpose of this study is to analyze the existing method of converting speech in the Kazakh language into text and evaluate their effectiveness. The research methodology includes the analysis of the VOSK model for speech transformation in the Kazakh language. An experimental study is being conducted based on the KazakhTTS dataset using machine learning and natural language processing methods. The results of the experiment, presented as an indicator of the error rate in the word (WER), showed that VOSK big and VOSK small have almost the same indicators (51% and 53% respectively). It was also noted that there are limitations in recognizing word endings and that some errors occur during speech recognition. The discussion of the results highlights the potential of the model and points to the need for further improvement and training in working with more diverse data. In conclusion, the key conclusions are outlined, as well as potential directions for further research in the field of Kazakh speech recognition.
Development of transcript-Driven IT Specialization Recommendation System using ML
(Faculty of Engineering and Natural Science, 2024) Myrzabayeva A.
Recommendation systems in education are pivotal for guiding students through their academic and career paths. However, traditional systems often fail to address the unique challenges and rapid changes within the Information Technology (IT) sector. This study proposes a machine learning-driven approach to enhance the precision and personalization of IT career guidance.This research develops a sophisticated machine learning model using a variety of algorithms, including Random Forest, Logistic Regression and Decision Trees, to analyze and process detailed student transcripts. The study aims to predict and align students’ IT specializations with both their capabilities and market demands. A robust validation framework, including cross-validation and algorithm comparison, ensures the accuracy and reliability of the recommendation system. The model demonstrates a high degree of predictive accuracy, outperforming traditional recommendation systems. It effectively identifies individual strengths and market opportunities, providing tailored recommendations that improve educational outcomes and job market readiness. Integrating machine learning with educational recommendation systems offers a promising avenue for addressing the specialization needs within the IT sector. By leveraging detailed transcript data and advanced predictive analytics, the proposed system aligns educational paths with professional demands, enhancing student employability and meeting industry needs.
Research models and methods for science strategic planning
(Faculty of Engineering and Natural Science, 2024) Abdrakhym A.
The purpose of this thesis is research models and methods for strategic science planning, to develop a website to showcase potential partners from 520 international universities around the world, which further facilitates collaboration. I hope to contribute to the development of strategic planning in the field of science in Kazakhstan by establishing links between scientists and promoting international cooperation. Using methods of analysis and data collection from various sources, the study established links between Kazakhstan’s scientists and the international scientific community. Using network analysis and statistical methods, the most positive areas of research in Kazakhstan and areas of international cooperation have been identified.
Denoising face recognition system
(Faculty of Engineering and Natural Science, 2024) Abdrakhim D.
This thesis investigates the influence of sophisticated denoising techniques on the efficacy of face recognition systems, particularly in environments characterized by substantial image noise. Considering the dependency of face recognition algorithms on the quality of input images, this research conducts a comprehensive evaluation of various denoising strategies, ranging from conventional filters like Median, Bilate-ral, and Gaussian, to advanced deep learning approaches, exemplified by the Deep Convolutional Neural Network (DnCNN). The Extended Yale B dataset, augmented with synthetically introduced noise, provides the basis for this empirical study. Employing quantitative metrics such as Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM), coupled with qualitative evaluations, this dissertation quantifies the enhancements in image quality and recognition precision afforded by each denoising method. The findings affirm that the integration of advanced denoising algorithms markedly improves the accuracy of face recognition systems, highlighting the efficacy of adaptive deep learning solutions in addressing the complexities introduced by noisy visual environments.
AI & Feedback Analytics in Higher Education: Strategies to Enhance Your Teaching and Learning Experience
(Annual Explorance 2024 conference, 2025-02) Prof. Andrew Turner; Samit Lotlikar; Dr. Shifan Khanday; Prof. Ahmed Ghonim; Prof. Abhilasha Singh; Dr. Amit Sareen; Dr. Hanadi Kadbey; Dr. Reem Al Gurg; Suzan Abu Shakra; Gulnaz Toguzbayeva; Samer Jaffar
Artificial intelligence is no longer a future concept; it’s here to stay, and it’s revolutionizing how organizations innovate and thrive. For higher education institutions, AI offers a winning formula to prioritize growth and enhance student engagement. By harnessing the power of AI-powered feedback analytics, institutions move beyond surface-level insights, uncovering actionable opportunities that drive meaningful change.
This eBook dives into how AI transforms the way higher education institutions collect, analyze, and act on student feedback. With these tools, institutions can uncover patterns, detect early warning signs, and implement strategies that align with their mission and vision—making AI an indispensable ally in shaping the future of higher education.