Meteorological measurements for weather forecasting and climatology have been carried out on a regular basis for centuries. There are several meteorological parameters like temperature, rainfall, humidity, speed of the wind etc. By studying and observing these parameters one can tell about the air pollution of an area or maybe the humidity present in the atmosphere. One can also predict cyclones or any natural calamities related to it. Numerous methodologies and strategies have been adopted for the analysis of these parameters. However, the data acquired can only be evaluated and interpreted after having statistically recorded medium-term and long-term atmospheric conditions. One of the most efficient tools for analysis is the soft computing techniques. These techniques have numerous advantages, as these techniques can be used for prediction studies and also for finding out any trends or patterns. In this paper, several soft computing techniques like linear regression, logistic regression, k-nearest neighbor, random forest regression (RFR) and support vector regression (SVR) are used for modeling of these meteorological parameters and a comparative analysis has been shown. The linear regression technique is giving very poor results for the modeling of most of the parameters. RFR and SVR mostly showing high accuracy rates for most of the meteorological parameters and these two techniques are quite efficient in comparison to other methods for showing the trend.
Previous Article in event
Identification and characterization of a highly chitinase-producing Paenibacillus timonensis LK-DZ15 strainPrevious Article in congress
Next Article in event Next Article in congress
Meteorological Parameter Modeling with Different Soft Computing Techniques
Published: 09 March 2021 by MDPI in MOL2NET'21, Conference on Molecular, Biomed., Comput. & Network Science and Engineering, 7th ed. congress USE.DAT-07: USA-Europe Data Analysis Trends Congress, Cambridge, UK-Bilbao, Basque Country-Miami, USA, 2021
Keywords: Rainfall; Temperature; Random Forest Regression; Support Vector Regression; k-Nearest Neighbor