Name: A Performance Comparison Study on Forest Fire Prediction (en Inglés)
SKU: 59015093
Price: 168312.00 COP
Availability: InStock
Author: Balamurugan R.

portada A Performance Comparison Study on Forest Fire Prediction (en Inglés)

A Performance Comparison Study on Forest Fire Prediction (en Inglés)

Formato

Libro Físico

Autor

Balamurugan R.

Editorial

LAP Lambert Academic Publishing

Idioma

Inglés

N° páginas

Encuadernación

Tapa Blanda

Dimensiones

22.9x15.2x0.3 cm

Peso

0.09 kg.

ISBN13

9786206145363

A Performance Comparison Study on Forest Fire Prediction (en Inglés)

Balamurugan R. (Autor) · LAP Lambert Academic Publishing · Tapa Blanda

Libro Nuevo Importado

Envío: 20 a 24 días háb.

$ 336.623$ 168.312

-50%

Costos de importación incluídos en el precio ✅

Libro Nuevo

Quedan 10 unidades

$ 168.312

¡Envío Gratis! Llega entre el 17 Sep y el 24 Sep a Bogota, Cundinamarca. Seleccionar ubicación

Agregar a lista de deseos

Reseña del libro "A Performance Comparison Study on Forest Fire Prediction (en Inglés)"

Forest fire is the uncontrolled and non-prescribed combustion of natural vegetation posing a great threat towards the environment. Being a natural phenomenon, it is impossible for mankind to prevent it. There is a need for early forecasting, quick detection and prompt action for controlling such phenomena for protecting the ecosystem. In this book, a predictor model which is able to predict forest fires are derived using soft computing and machine learning techniques. The dataset consists of 517 records of time series data for Montesinho natural park, in Portugal. To find some of the critical patterns and to segment the fire regions (using clustering), PCA and clustering methods through K-means are used respectively on the dataset. Five soft computing techniques namely MPNN, PNN, KNN, RBF and SVM are applied concurrently. The Python libraries such as Scikit-learn, Pandas, Matplotlib and Seaborn are utilized for the execution of the algorithms. Finally, each soft computing technique is assessed under the evaluation parameters such as MSE, RMSE, MAE, RAE & IG and the appropriate model having the best values is identified.