Series temporales para el índice Diferencial Normalizado de Vegetación mediante una Red Neuronal Artificial de corto y largo plazo, y el algoritmo Prophet

Marco Javier Castelo Cabay, Edgar Francisco Merino Villa, Mayra Elizabeth Peñafiel Tixi, Bélgica Marcela Basantes Erazo

Resumen


La presente investigación se realizó una evaluación de modelos para el pronóstico de series temporales del Índice Normalizado de Vegetación (NDVI) por medio de una Red Neuronal Recurrente (RNR) de corto y largo plazo, y el algoritmo Prophet de Facebook.


Los datos se obtuvieron del sensor espacial Moderate Resolution Imaging Spectroradiometer (MODIS) que emite información con una periodicidad de 16 días, se obtuvieron valores desde enero de 2013 hasta diciembre del 2021 por medio de la plataforma Google Earth Engine (GEE). Utilizando el lenguaje de programación Python en un entorno Jupyter se construyó la red neuronal Long-Short Term Memory (LSTM), y el algoritmo Prophet, tomando como datos de entrenamiento 172 valores y 36 para prueba en ambos casos. Como métrica de evaluación se consideró Root Mean Square Error RMSE (RMSE) y Mean Square Error (MSE), obteniéndose valores de 0.509, 0.259 para LSTM y 0.5311, 02820 para Prophet, demostrando que la red LSTM tiene mejor rendimiento para la predicción de NDVI.


Palabras clave


NDVI; series temporales; pronóstico; LSTM; Prophet.

Texto completo:

PDF HTML XML

Referencias


Al-Shateri, H. A. (2020). Land use land cover change detection in the mining areas of V.D. Yalevsky coal mine-russia. Mining Informational and Analytical Bulletin, (6-1), 212–223. https://doi.org/10.25018/0236-1493-2020-61-0-212-223

Bashir, T., Haoyong, C., Tahir, M. F., & Liqiang, Z. (2022). Short term electricity load forecasting using hybrid prophet-LSTM model optimized by BPNN. Energy Reports, 8, 1678–1686. https://doi.org/10.1016/j.egyr.2021.12.067

Cotton, P. (2021, February 3). Is Facebook's "Prophet" the time-series messiah, or just a very naughty boy? Microprediction. Retrieved June 23, 2022, from https://www.microprediction.com/blog/prophet

Dubrovin, K. N., Stepanov, A. S., & Aseeva, T. A. (2022). Application of lai and NDVI to model soybean yield in the regions of the Russian Far East. IOP Conference Series: Earth and Environmental Science, 949(1), 012030. https://doi.org/10.1088/1755-1315/949/1/012030

Gonzalez, J. M. C. (2020, April 24). Forecasting de series temporales con redes neuronales. LinkedIn. Retrieved June 22, 2022, from https://www.linkedin.com/pulse/forecasting-de-series-temporales-con-redes-neuronales-casas-gonzalez/

Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment, 202, 18–27. https://doi.org/10.1016/j.rse.2017.06.031

Hayashi, H. (2017, October 18). Is prophet really better than Arima for forecasting time series data? Medium. Retrieved June 23, 2022, from https://blog.exploratory.io/is-prophet-better-than-arima-for-forecasting-time-series-fa9ae08a5851

Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735

Malki, A., Atlam, E.-S., & Gad, I. (2022). Machine learning approach of detecting anomalies and forecasting time-series of IOT devices. Alexandria Engineering Journal, 61(11), 8973–8986. https://doi.org/10.1016/j.aej.2022.02.038

Mañas, A. (2019). Notas sobre pronostico del Flujo de Trafico en la Ciudad de Madrid. bookdown (Vol. I). GRIN VERLAG. Retrieved June 23, 2022, from https://bookdown.org/amanas/traficomadrid/resumen.html.

Middya, A. I., & Roy, S. (2022). Pollutant Specific Optimal Deep Learning and Statistical Model Building for Air Quality Forecasting. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4010743

Ning, Y., Kazemi, H., & Tahmasebi, P. (2022). A comparative machine learning study for time series oil production forecasting: Arima, LSTM, and prophet. Computers & Geosciences, 164, 105126. https://doi.org/10.1016/j.cageo.2022.105126

Olah, C. (2015, August 27). Understanding LSTM networks. Understanding LSTM Networks -- colah's blog. Retrieved June 22, 2022, from https://colah.github.io/posts/2015-08-Understanding-LSTMs/

Omar, M. S., & Kawamukai, H. (2022). Evaluation of stochastic and artificial neural network models for multi-step lead forecasting of Ndvi. IOP Conference Series: Earth and Environmental Science, 1008(1), 012014. https://doi.org/10.1088/1755-1315/1008/1/012014

Pascual, A., Tupinambá-Simões, F., Guerra-Hernández, J., & Bravo, F. (2022). High-resolution planet satellite imagery and multi-temporal surveys to predict risk of tree mortality in tropical eucalypt forestry. Journal of Environmental Management, 310, 114804. https://doi.org/10.1016/j.jenvman.2022.114804

Rahman, M., Ahmed, M. T., Nur, S., & Touhidul Islam, A. Z. (2022). The prediction of Coronavirus Disease 2019 outbreak on Bangladesh Perspective Using Machine Learning: A Comparative Study. International Journal of Electrical and Computer Engineering (IJECE), 12(4), 4276. https://doi.org/10.11591/ijece.v12i4.pp4276-4287

Rodríguez, A. A. (2019). Análisis de las series temporales a la luz de Deep Learning. Anuario Jurídico y Económico Escurialense, II, 257–276.

Shohan, M. J., Faruque, M. O., & Foo, S. Y. (2022). Forecasting of Electric Load using a hybrid LSTM-neural prophet model. Energies, 15(6), 2158. https://doi.org/10.3390/en15062158

Siami-Namini, S., Tavakoli, N., & Siami Namin, A. (2018). A comparison of Arima and LSTM in forecasting time series. 2018 17th IEEE International Conference on Machine Learning and Applications (ICMLA). https://doi.org/10.1109/icmla.2018.00227

Sirikonda, S., Kumar, S. N., Sravanthi, T., Srinivas, J., Manchikatla, S. T., & Kumaraswamy, E. (2022). Forecast the death and recovery rate of COVID 2019 using Arima and Prophet Models. INTERNATIONAL CONFERENCE ON RESEARCH IN SCIENCES, ENGINEERING & TECHNOLOGY. https://doi.org/10.1063/5.0081771

Touhami, I., Moutahir, H., Assoul, D., Bergaoui, K., Aouinti, H., Bellot, J., & Andreu, J. M. (2022). Multi-year monitoring land surface phenology in relation to climatic variables using Modis-ndvi time-series in Mediterranean Forest, Northeast Tunisia. Acta Oecologica, 114, 103804. https://doi.org/10.1016/j.actao.2021.103804

Wibawa, A. P., Utama, A. B., Elmunsyah, H., Pujianto, U., Dwiyanto, F. A., & Hernandez, L. (2022). Time-series analysis with smoothed convolutional neural network. Journal of Big Data, 9(1). https://doi.org/10.1186/s40537-022-00599-y

Xiao, W., Deng, X., He, T., & Chen, W. (2020). Mapping annual land disturbance and reclamation in a surface coal mining region using Google Earth engine and the LANDTRENDR algorithm: A case study of the shengli coalfield in Inner Mongolia, China. Remote Sensing, 12(10), 1612. https://doi.org/10.3390/rs12101612

Zhu, W., & Lei, H. (2018). Urban Vegetation Coverage Monitoring Technology based on NDVI. Proceedings of the 2018 7th International Conference on Energy, Environment and Sustainable Development (ICEESD 2018). https://doi.org/10.2991/iceesd-18.2018.291




DOI: https://doi.org/10.23857/pc.v7i8.4427

Enlaces de Referencia

  • Por el momento, no existen enlaces de referencia
';

Bookmark and Share




Polo del Conocimiento              

Revista Científico-Académica Multidisciplinaria

ISSN: 2550-682X

Casa Editora del Polo                                                 

Manta - Ecuador       

Dirección: Ciudadela El Palmar, II Etapa,  Manta - Manabí - Ecuador.

Código Postal: 130801

Teléfonos: 056051775/0991871420

Email: polodelconocimientorevista@gmail.com / director@polodelconocimiento.com

URL: https://www.polodelconocimiento.com/