Gulnaz Gimaletdinova
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Gulnaz Gimaletdinova has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Ruslan Isaev
Paragon International University, Cambodia
6 shared publications - Zhusupbek Saipidinov
Ala-too International University, Kyrgyzstan
2 shared publications - Aldaiar Ramis Uulu
Ala-too International University, Kyrgyzstan
1 shared publication - Almazbek Ilash
Ala Too International University, Kyrgyzstan
1 shared publication - Ashin Kurbanov
Ala-too International University, Kyrgyzstan
1 shared publication - Bekturali Orozov
Ala-too International University, Kyrgyzstan
1 shared publication
Latest Publications
- Open Access
- 25 December 2024
- Published by MDPI AG in Preprints.org
Eye Disease Classification Using Deep Learning Approaches: A Case Study on Retinal Images
Glaucoma, cataracts, and diabetic retinopathy are among the world's major causes of blindness. Early detection and classification are the most important ways to prevent further loss of vision. With more and more medical imaging data and development in deep learning technologies, this is an excellent opportunity to automate this diagnostic process that significantly reduces the workload of healthcare professionals while at the same time increases the accuracy of diagnosis. The aim of this paper is to study the performance of a deep learning approach for the classification of eye diseases from retinal images. It is a CNN, specifically EfficientNet-B3, very effective in image classification tasks. The dataset used in this paper consists of retinal images that were divided into categories such as glaucoma, cataract, diabetic retinopathy, and normal ones. The research was performed by exploring the model performance in respect of accuracy, precision, recall, and F1-score in the classification of the data. The outcome showed that the model was able to achieve an overall accuracy of 92.5%, within which all metrics showed very outstanding performances in all categories. This discussion presents the effectiveness, shortcomings, and future possibilities the model may have in clinics for real-time diagnosis.
- 25 December 2024
- Published by MDPI AG in Preprints.org
Open AccessEye Disease Classification Using Deep Learning Approaches: A Case Study on Retinal Images
Glaucoma, cataracts, and diabetic retinopathy are among the world's major causes of blindness. Early detection and classification are the most important ways to prevent further loss of vision. With more and more medical imaging data and development in deep learning technologies, this is an excellent opportunity to automate this diagnostic process that significantly reduces the workload of healthcare professionals while at the same time increases the accuracy of diagnosis. The aim of this paper is to study the performance of a deep learning approach for the classification of eye diseases from retinal images. It is a CNN, specifically EfficientNet-B3, very effective in image classification tasks. The dataset used in this paper consists of retinal images that were divided into categories such as glaucoma, cataract, diabetic retinopathy, and normal ones. The research was performed by exploring the model performance in respect of accuracy, precision, recall, and F1-score in the classification of the data. The outcome showed that the model was able to achieve an overall accuracy of 92.5%, within which all metrics showed very outstanding performances in all categories. This discussion presents the effectiveness, shortcomings, and future possibilities the model may have in clinics for real-time diagnosis.
- Open Access
- 23 December 2024
- Published by MDPI AG in Preprints.org
Traffic Violation Analysis in Kyrgyzstan (2019-2023): Utilizing Data-Driven Strategies for Enhanced Traffic Management and Violation Mitigation
The explosive growth in the use of motor vehicles has caused many issues in ensuring road and traffic safety. The aim of this paper is to establish the trends and factors that explain traffic violations across the Republic of Kyrgyzstan from 2019 to 2023 to provide information that can be utilized by traffic management authorities. To this end we analyze the frequency and types of violations as well as the distribution of violations by date and time, geographical regions, and types of vehicles from a data set that has sourced more than 2,354,392 recorded traffic violations since Safe City Project commenced in Kyrgyzstan. The analysis defines the places and times where there is traffic, most violations of traffic laws and their major causes. The results inform smart and evidence-based approaches aimed at the enhancement of traffic control, improving compliance levels and overall tackling of the problem of road safety. This work also emphasizes the importance of data analytics applications in addressing the problems of managing traffic in big cities efficiently.
- 23 December 2024
- Published by MDPI AG in Preprints.org
Open AccessTraffic Violation Analysis in Kyrgyzstan (2019-2023): Utilizing Data-Driven Strategies for Enhanced Traffic Management and Violation Mitigation
The explosive growth in the use of motor vehicles has caused many issues in ensuring road and traffic safety. The aim of this paper is to establish the trends and factors that explain traffic violations across the Republic of Kyrgyzstan from 2019 to 2023 to provide information that can be utilized by traffic management authorities. To this end we analyze the frequency and types of violations as well as the distribution of violations by date and time, geographical regions, and types of vehicles from a data set that has sourced more than 2,354,392 recorded traffic violations since Safe City Project commenced in Kyrgyzstan. The analysis defines the places and times where there is traffic, most violations of traffic laws and their major causes. The results inform smart and evidence-based approaches aimed at the enhancement of traffic control, improving compliance levels and overall tackling of the problem of road safety. This work also emphasizes the importance of data analytics applications in addressing the problems of managing traffic in big cities efficiently.
- Open Access
- 19 December 2024
- Published by MDPI AG in Preprints.org
Stock Market Prediction and Forecasting Using Historical Data
Stock market prediction is a critical area of financial research that aims to forecast future price movements of stocks using historical data, economic indicators, and advanced machine learning algorithms. This thesis focuses on utilizing the dataset titled "Stock Market Prediction & Forecasting" from Kaggle to develop predictive models capable of generating accurate market forecasts. The study investigates various methodologies, including time series analysis, feature engineering, and the application of machine learning techniques such as Random Forest, XGBoost, and Long Short-Term Memory (LSTM) networks. Results demonstrate the feasibility of achieving meaningful insights into stock price trends, enabling investors and analysts to make informed decisions.
- 19 December 2024
- Published by MDPI AG in Preprints.org
Open AccessStock Market Prediction and Forecasting Using Historical Data
Stock market prediction is a critical area of financial research that aims to forecast future price movements of stocks using historical data, economic indicators, and advanced machine learning algorithms. This thesis focuses on utilizing the dataset titled "Stock Market Prediction & Forecasting" from Kaggle to develop predictive models capable of generating accurate market forecasts. The study investigates various methodologies, including time series analysis, feature engineering, and the application of machine learning techniques such as Random Forest, XGBoost, and Long Short-Term Memory (LSTM) networks. Results demonstrate the feasibility of achieving meaningful insights into stock price trends, enabling investors and analysts to make informed decisions.
- Open Access
- 18 December 2024
- Published by MDPI AG in Preprints.org
Improvement of Speed-Strength Qualities in Young Football Players Aged 12-13
The development of speed-strength qualities, or "explosive power," is very important for young football players aged 12–13 who want to be successful in the game. This study looks at how effective different training methods are for improving these abilities. We used key training approaches, such as plyometric exercises, resistance training, and dynamic drills with weights, to help build explosive power, The research focused on exercises with various types of weights—above, equal to, and below what players usually lift—so they can improve their skills and mimic real game situations. We tested the players abilities in vertical and horizontal jumps, multi-jump series, and strength during resisted movements. The results showed that the players made significant progress in explosive power This study highlights the importance of using a variety of training methods to help young athletes develop the speed and strength they need to perform better on the field.
- 18 December 2024
- Published by MDPI AG in Preprints.org
Open AccessImprovement of Speed-Strength Qualities in Young Football Players Aged 12-13
The development of speed-strength qualities, or "explosive power," is very important for young football players aged 12–13 who want to be successful in the game. This study looks at how effective different training methods are for improving these abilities. We used key training approaches, such as plyometric exercises, resistance training, and dynamic drills with weights, to help build explosive power, The research focused on exercises with various types of weights—above, equal to, and below what players usually lift—so they can improve their skills and mimic real game situations. We tested the players abilities in vertical and horizontal jumps, multi-jump series, and strength during resisted movements. The results showed that the players made significant progress in explosive power This study highlights the importance of using a variety of training methods to help young athletes develop the speed and strength they need to perform better on the field.
- Open Access
- 16 December 2024
- Published by MDPI AG in Preprints.org
Diagnosing Heart Diseases With One-Dimensional Evolutionary Neural Networks Based On Heart Sounds
This study aims to investigate the effectiveness of deep learning methods in classifying heart sounds (phonocardiograms - PCG). Early diagnosis of heart diseases is critical for improving patients' quality of life and reducing mortality rates. The project uses a comprehensive heart sounds database provided by PhysioNet/CinC Challenge 2016.The data is collected from various sources, enhancing the model's generalizability. During the feature extraction process, MFCC, Chroma, and Mel-Spectrogram features were extracted from the sound signals. A high-pass filter was applied for noise reduction.In the modeling phase, a 1D Convolutional Neural Network (CNN) was used. The model was evaluated using accuracy, classification report, and ROC-AUC scores. The results show that the proposed CNN model provides high accuracy in classifying heart sounds as normal or abnormal. This study represents a significant step towards developing an automatic system for early diagnosis of heart diseases.
- 16 December 2024
- Published by MDPI AG in Preprints.org
Open AccessDiagnosing Heart Diseases With One-Dimensional Evolutionary Neural Networks Based On Heart Sounds
This study aims to investigate the effectiveness of deep learning methods in classifying heart sounds (phonocardiograms - PCG). Early diagnosis of heart diseases is critical for improving patients' quality of life and reducing mortality rates. The project uses a comprehensive heart sounds database provided by PhysioNet/CinC Challenge 2016.The data is collected from various sources, enhancing the model's generalizability. During the feature extraction process, MFCC, Chroma, and Mel-Spectrogram features were extracted from the sound signals. A high-pass filter was applied for noise reduction.In the modeling phase, a 1D Convolutional Neural Network (CNN) was used. The model was evaluated using accuracy, classification report, and ROC-AUC scores. The results show that the proposed CNN model provides high accuracy in classifying heart sounds as normal or abnormal. This study represents a significant step towards developing an automatic system for early diagnosis of heart diseases.
- Open Access
- 11 December 2024
- Published by MDPI AG in Preprints.org
Predicting Passenger Car Prices with Machine Learning Models
The car resale industry is broad and precise price forecast is essential for preventing customers from overpaying to resellers. Machine learning is currently the most effective forecasting tool, which is why three machine learning models—Random Forest, CatBoost Regressor, and Support Vector Machine (SVM)—were chosen to analyze car features such as model, mileage, and year, as well as additional attributes such as color, conditioner, and car body type extracted from car advertisement descriptions. The data for the prediction was collected from the website mashina.kg[1] using a web scraper implemented in the Python programming language. The performances of multiple models were then evaluated to determine which one best suited the available data set. As a result, the model CatBoost provides the most effective way for accurate car price predictions in Bishkek, with an accuracy of 86.05%, making it a helpful tool for both customers and financial institutions in making well-informed decisions regarding cars price.
- 11 December 2024
- Published by MDPI AG in Preprints.org
Open AccessPredicting Passenger Car Prices with Machine Learning Models
The car resale industry is broad and precise price forecast is essential for preventing customers from overpaying to resellers. Machine learning is currently the most effective forecasting tool, which is why three machine learning models—Random Forest, CatBoost Regressor, and Support Vector Machine (SVM)—were chosen to analyze car features such as model, mileage, and year, as well as additional attributes such as color, conditioner, and car body type extracted from car advertisement descriptions. The data for the prediction was collected from the website mashina.kg[1] using a web scraper implemented in the Python programming language. The performances of multiple models were then evaluated to determine which one best suited the available data set. As a result, the model CatBoost provides the most effective way for accurate car price predictions in Bishkek, with an accuracy of 86.05%, making it a helpful tool for both customers and financial institutions in making well-informed decisions regarding cars price.
- Open Access
- 11 December 2024
- Published by MDPI AG in Preprints.org
Diabetes Prediction Using Machine Learning Techniques: A Comprehensive Analysis
This article analyzes machine learning methods for predicting diabetes using clinical data. The study focuses on four models: Logistic Regression, Decision Tree, Gradient Boosting, and XGBoost. Patient data, including age, gender, blood glucose levels, and body mass index, were used to evaluate the accuracy of these models. Data preprocessing techniques such as scaling and normalization were applied to improve results. The findings reveal that ensemble methods like Gradient Boosting and XGBoost outperform traditional models in prediction accuracy. These machine learning approaches not only enhance prediction but also identify key risk factors, aiding in early diagnosis and timely prevention of diabetes.
- 11 December 2024
- Published by MDPI AG in Preprints.org
Open AccessDiabetes Prediction Using Machine Learning Techniques: A Comprehensive Analysis
This article analyzes machine learning methods for predicting diabetes using clinical data. The study focuses on four models: Logistic Regression, Decision Tree, Gradient Boosting, and XGBoost. Patient data, including age, gender, blood glucose levels, and body mass index, were used to evaluate the accuracy of these models. Data preprocessing techniques such as scaling and normalization were applied to improve results. The findings reveal that ensemble methods like Gradient Boosting and XGBoost outperform traditional models in prediction accuracy. These machine learning approaches not only enhance prediction but also identify key risk factors, aiding in early diagnosis and timely prevention of diabetes.
- Open Access
- 12 November 2024
- Published by MDPI AG in Preprints.org
Forecasting Energy Consumption Using Machine Learning: A Case Study of Kyrgyzstan Using Socioeconomic Data
As global energy demands soar, understanding and accurately forecasting energy consumption becomes crucial for both economic stability and sustainable development. This paper explores the application of machine learning techniques to predict energy consumption, using the "World Energy Consumption" dataset from Kaggle. Focusing on Kyrgyzstan as a case study, we investigate how factors such as energy per capita, population, and GDP influence energy demand. Employing a regression model, we evaluate the effectiveness of socioeconomic indicators in predicting energy needs. Results demonstrate the potential for these methods to contribute to energy management strategies, though limitations point toward the need for more complex models and broader datasets.
- 12 November 2024
- Published by MDPI AG in Preprints.org
Open AccessForecasting Energy Consumption Using Machine Learning: A Case Study of Kyrgyzstan Using Socioeconomic Data
As global energy demands soar, understanding and accurately forecasting energy consumption becomes crucial for both economic stability and sustainable development. This paper explores the application of machine learning techniques to predict energy consumption, using the "World Energy Consumption" dataset from Kaggle. Focusing on Kyrgyzstan as a case study, we investigate how factors such as energy per capita, population, and GDP influence energy demand. Employing a regression model, we evaluate the effectiveness of socioeconomic indicators in predicting energy needs. Results demonstrate the potential for these methods to contribute to energy management strategies, though limitations point toward the need for more complex models and broader datasets.