- DATA SET INFORMATION
Data comes originally from INMET - Instituto Nacional de Meteorologia (inmet.gov.br)
These datasets contain monthly data of air humidity, cloudiness, rainfall, maximum temperature, minimum temperature, and mean temperature obtained from four brazilian weather stations: Manaus, Sao Paulo, Natal, and Porto Alegre. Data were collected from January of 1990 to December of 2015.
Number of samples in each dataset: 312. Number of attributes: 6.
The datasets were used in:
Soares, E.; Costa Jr., P.; Costa, B.; Leite, D. "Ensemble of Evolving Data Clouds and Fuzzy Models for Weather Time Series Prediction." Applied Soft Computing - Elsevier, xx-x, 2018.
to develop evolving prediction models.
The intention of the datasets is to help others to obtain the same data and replicate our experiments.
- ATTRIBUTE INFORMATION
Name / Data Type / Measurement Unit
1 - Humidity / continuous / Percentage (%)
2 - Cloudiness / continuous / Tenth (T)
3 - Rainfall / continuous / Millimeters per square meter (mm/m^2)
4 - Maximum temperature / continuous / Degrees Celsius (C)
5 - Minimum temperature / continuous / Degrees Celsius (C)
6 - Mean temperature / continuous / Degrees Celsius (C)
The mean temperature is the value to predict.
- RELEVANT PAPERS
Soares, E.; Costa Jr., P.; Costa, B.; Leite, D. "Ensemble of evolving data clouds and fuzzy models for weather time series prediction." Applied Soft Computing - Elsevier, xx (x), 2018.
Soares, E.; Mota, V.; Poucas, R.; Leite, D. "Cloud-based evolving intelligent method for weather time series prediction." 2017 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE) - Naples IT, 6p. 2017.
Leite, D.; Ballini, R.; Costa, P.; Gomide, F. "Evolving fuzzy granular modeling from nonstationary fuzzy data streams." Evolving Systems 3 (2), 65-79, 2012.