التفاصيل البيبلوغرافية
| العنوان: |
Analysis for Predicting Respiratory Diseases from Air Quality Attributes Using Recurrent Neural Networks and Other Deep Learning Techniques. |
| المؤلفون: |
Deo, Arpit, Khan, Safdar Sardar, Doohan, Nitika Vats, Jain, Aviral, Nighoskar, Mitali, Dandawate, Aditi |
| المصدر: |
Ingénierie des Systèmes d'Information; Apr2024, Vol. 29 Issue 2, p731-739, 9p |
| مصطلحات موضوعية: |
RECURRENT neural networks, DEEP learning, AIR quality, FEDERATED learning, RESPIRATORY diseases, STANDARD deviations |
| مستخلص: |
The primary objective of this investigation is to establish a clear correlation between air quality and the prevalence of respiratory conditions, this is done by employing deep learning (DL) methodologies. RNN was compared against alternative methods such as GNN, CNN, feed-forward technique (DL), k-nearest neighbours, linear regression, decision tree, and neural network. Performance evaluations were conducted employing an Octuple crossvalidation approach, with the root mean square error (RMSE) employed to perform comparative analysis. RMSE serves as the primary metric for evaluating regression models, additional criteria for model comparison include computational efficiency, scalability, assessing model performance with larger datasets; and generalization to new data. While RMSE's simplicity and sensitivity to large errors are strengths, limitations include sensitivity to outliers and overlooking distributional differences. The data was collected on two major features, the first being the Air quality dataset which contained various gases present in the atmosphere that affect the air quality, and the second being the hospital patient dataset which indicates the number of people suffering from respiratory diseases due to the air quality. The study acknowledges the superior performance of the Recurrent Neural Network (RNN) model but suggests the need for new learning methods handling small datasets and explores efficient, scalable approaches like distributed and federated learning. [ABSTRACT FROM AUTHOR] |
|
Copyright of Ingénierie des Systèmes d'Information is the property of International Information & Engineering Technology Association (IIETA) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| قاعدة البيانات: |
Complementary Index |