Machine learning based smart weather prediction
In one sentence. Compares linear regression, SVM and decision trees with empirical models for station precipitation, humidity, wind and radiation forecasts in India, reporting MSE 0.1397.
Abstract
Weather forecasting refers to the prediction of atmospheric conditions depending on a given time and location. Weather prediction is essential and it plays a significant role in many sectors namely energy and utililities, marine transportation, aviation, agriculture and forestry to a greater extent. Accurate weather forecast mechanism help the farmers for suitable planning of farming operations that will prevent crop losses. In this work, the weather parameters namely precipitation, relative humidity, wind speed and solar radiation were predicted for few Indian locations using the conventional temperature based empirical models and machine learning algorithms such as linear regression, support-vector machine (SVM) and decision tree. Forecasting of weather parameters, on which agriculture depends, will increase the overall yield and it helps farmers and agricultural-based businesses to plan better. From the current results, it is observed that machine learning (ML) based methods had a better prediction results than the physics based conventional models for weather forecasting with mean square error of 0.1397 and correlation coefficient of 0.9259. The objective of this work is to arrive at an optimized end result and a better weather prediction using the Machine learning models with lesser computational effort.
Labels
- Topic
- Local & station forecasting
- Type
- Evaluation or intercomparison of existing models
- Method
- Classical ML (trees, GP, linear, kernels)
- Time scale
- 6 h–3 days
- Domain
- Local / station
- Data
- Station / in-situ
- Status
- Research
- Identifiers
- DOI:10.11591/ijeecs.v28.i1.pp508-515 OpenAlex:W4293324552
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.