Abstract
<p class="AbstractL-MAG">—<em> One of data mining techniques is Classification, used to predict relationships between data on a dataset. The prediction performed by classifying data into several different classes considering certain factor. Classification is a performance of Supervised Learning application where the training data already has a label when entered as input data. Classification is an approach of empirical techniques that can be utilized for short-term weather prediction. The most widely used algorithms in Classification Techniques are Classification Tree, Naïve Bayes and K-Nearest Neighbors. In this study, the author used these three algorithms to predict rain with validation parameters of Brier Score, Confusion Matrix and ROC curves. The input data is synoptic data of Kemayoran Meteorological Station, Jakarta (96745) for 10 years (2006 - 2015) consists of 3528 datasets and 8 attributes. Based on a series of data processing, selection and model testing shows that the Naïve Bayes Algorithm has the best accuracy rate of 77.1% with the category of fair classification so it is quite potential to be used in the operational. The dominant weather attributes in rain formation are moisture (RHavg), minimum temperature (Tmin), maximum temperature (Tmax), average temperature (Tavg) and wind direction (ddd).</em></p>
Highlights
Weather prediction is a challenge in meteorology that has been a major subject of meteorological research
The research in [2] using Artificial Neural Network (ANN) and Decision Tree (DT) algorithms to analyze meteorological data for 10 years (2000 - 2009) period at the Ibadan Meteorology Flight Station, Nigeria, the results show that this technique is good for weather prediction, the C5 Decision Tree is used to generate decision trees and rules to classify weather parameters
Based on research on short-term weather prediction using the classification algorithm on rain prediction based on probabilistic supervised learning with Brier Score, Confusion Matrix and Receiver Operating Characteristic test parameters, the following conclusions can be drawn: 1. Based on the results of the three test parameters namely Brier Score, Confusion Matrix and ROC curve, the three algorithms can be applied to weather data with a fairly good category, namely fair classification
Summary
Weather prediction is a challenge in meteorology that has been a major subject of meteorological research. Approach in weather prediction can be done by empirical or dynamic method. Short-term weather predictions have been using dynamic methods which are an analytical approach based on the principles of fluid dynamics, while empirical methods performed with statistical and mathematical approaches are more widely used for long-term weather predictions. Both approaches have their own flaws and advantages. The use of empirical methods in BMKG for short-term weather prediction are not much done yet. The researchers are interested to examine more about how the use of empirical methods, especially data mining techniques for short-term weather prediction. The weather parameters discussed in this study are dew point, humidity, wind speed, pressure, mean sea level, wind speed and rainfall using kNN, Naïve Bayes, Multiple Regression and ID3 algorithm with good result in predicting weather [1]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
