A Hybrid Genetic Algorithm and Artificial Neural Networks (GANN) for Feature Reduction in Forest Fire Detection
Publish place: The first international conference of modern research engineers in electricity and computer
Publish Year: 1395
نوع سند: مقاله کنفرانسی
زبان: English
View: 715
This Paper With 20 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
این Paper در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
CBCONF01_0181
تاریخ نمایه سازی: 16 شهریور 1395
Abstract:
Early fire detection is very important in fighting fires. To this end, researchers have carried out work on feature reduction in automatic fire detection systems. One such method used in the fire detection field is Self-Organizing Maps (SOM). SOM has some limitations such as exponential complexity in large search spaces. This paper proposed GANN method based on the combination of Genetic Algorithms (GAs) and Artificial Neural Networks (ANNs) for forest fire detection. Applying GA will result in reduced number of features which could be used as inputs to ANN. The fitness function of the GA chromosomes is defined as the output value returned by ANN. Also we compare our results with SOM, Principal Component Analysis (PCA) and Independent Component Analysis (ICA). To demonstrate the efficiency of these methods, we have used the Canadian forest fire management information systems dataset. In this paper the performance of algorithms are compared based on mean square error (MSE) rate. Our simulation results demonstrated that the MSE in the GANN is less than other methods and the proposed method can reduce features without any reduction in accuracy.
Keywords:
Authors
Elham Mahdipour
Faculty member of Khavaran Higher Education Institute, Mashhad, Iran
Chitra Dadkhah
Assistant professor, K.N.Toosi University of Technology, Tehran, Iran
Mohammad Teshnehlab
professor, K.N.Toosi University of Technology, Tehran, Iran
مراجع و منابع این Paper:
لیست زیر مراجع و منابع استفاده شده در این Paper را نمایش می دهد. این مراجع به صورت کاملا ماشینی و بر اساس هوش مصنوعی استخراج شده اند و لذا ممکن است دارای اشکالاتی باشند که به مرور زمان دقت استخراج این محتوا افزایش می یابد. مراجعی که مقالات مربوط به آنها در سیویلیکا نمایه شده و پیدا شده اند، به خود Paper لینک شده اند :