Volume 15, Issue 10, 10 2024 Edition - IJSER Journal Publication


Publication for Volume 15, Issue 10, 10 2024 Edition - IJSER Journal Publication


IJSER Research Group https://www.ijser.org/forum/index.php Register for IJSER Research Forum      IJSER Xplore IJSER Xplore Research Paper Database

Pages   [1]
 



Analyzing Speech Impairments: A Machine Learning Approach to Dysarthria Detection []


Dysarthria includes dysfunction in the nerves and muscles controlling speech, leading to unclear spoken words. While many studies have been carried out to examine speech impairment, the variation of this problem among people with a similar dysarthria diagnosis has necessitated the need for more research in this area. The particular type and severity of the impairment are essential to monitor the progress of dysarthria and make effective therapeutic interventions. This project describes a Convolutional Neural Network (CNN) model for dysarthria detection, where several acoustic features are extracted in the form of zero crossing rates, Mel Frequency Cepstral Coefficients (MFCCs), spectral centroids, and spectral roll-off. Using the TORGO database of speech signals, training the model, and testing it for its efficiency has shown much promise in the early diagnosis of dysarthric speech. The numerical results indicate that the model design provides an efficiency of nearly 95%, which is higher than previous model architectures. This model aims to identify the condition early and help improve the management of dysarthria through timely and accurate diagnosis.


Applying SWARA Technique to Assess Risk Factors in PPP Waste Water Treatment Plant Projects in Egypt []


The aims of this study are: 1) to illustrate and cluster the risk factors in accordance with the public private partnership (PPP)waste water treatment plant (WWTP) projects in Egypt, 2) to assess the risk factors’ criticality degrees according to Step-Wise Weight Assessment Ratio Analysis (SWARA) method. A questionnaire survey was conducted on 20 experts in PPP projects to assess the severity of 57 risk factors gathered from literature. SWARA technique was applied to arrange these risk factors. Price change, contract termination, political corruption, political interference and technical risk are the most important risk factors. The originality of this research stems from the new technique SWARA in assessing risk factors in PPPWWTP projects. The major contribution of this research is a message that geared toward the governments and their policy-makers, especially those of the developing markets. The message is that economic and administrative unconventional actions should take as soon as possible in their systems to face these risk factors


Multifaceted Impact Analysis of Economic and Environmental Indicators on Sustainable Development in China []


This detailed analysis looks into the complex relationship between economic growth, environmental impacts, and technological advances in China's quest for sustainable development, as outlined by the United Nations in 2024. The study employs sophisticated statistical techniques such as Lasso regression analysis, OLS (Ordinary Least Squares), Granger causality tests, and Johansen cointegration to identify long-term and consistent relationships between crucial variables influencing China's environmental and economic conditions. This research provides insights into the intricate connections among technology, foreign direct investment, urban expansion, and environmental sustainability, particularly regarding their influence on CO2 emissions (World Bank, 2022; Lin, Lam, Shi, Chen, & Chen, 2023). The results question existing assumptions, providing a more profound comprehension of China's approaches to sustainable progress (United Nations, 2023). These observations benefit academics and decision-makers, aiding them in achieving harmony between environmental preservation and economic advancement


Application of CNN Architecture for Automated Human Activity Recognition In Deep Learning Network []


Convolutional neural networks (CNNs) are unique and produces great results in the image analysis and classification used for Human Activity Recognition (HAR). CNNs which is a subset of deep learning architectures known for their efficacy in processing and analyzing image data. This report explores the investigation of CNN models for automated HAR, leveraging its ability to automatically extract features and patterns from raw input data. This report uses convolutional neural network algorithms in deep learning to automatically extract features of activities related to different human activities and sporting life. The sample images are collected from Kaggle and randomly collected from Google. The network structure of the constructed CNN model consists of an input layer, two convolutional layers with Dropouts and two pooling layers. After comparing the average accuracy of each set of experiments and the test set of the best model obtained from it, the best model is then selected.




Pages   [1]