Weather Forecasting from Historical Weather Data using Data Mining

Background:

There is a Smart Grid Center that is conducting huge research to design an Intelligent Energy Use system. This system detects the weather conditions outside and would take an action accordingly (these actions would be like turning on the air conditioner, turning off the heater, etc). A huge number of historical weather information is collected from various weather sensor devices that are located across the RVR building. The information is gathered and found to be climates for a period of time (June 2012 to Present). The purpose of this project is to extract the patterns for day-to-day weather prediction from historical weather data using data mining.

Problem Statement:

Weather forecasting is a prediction of what the weather will be like in the future, it had been invented many years ago. The purpose of this project is to extract the patterns for day-to-day weather prediction from historical weather data using data mining. In this project, a prototype of the system will be developed which includes the main components of the system such as training, analysis, and prediction.

A web framework will be developed using Java Net Beans which is used to present the predicted result in a meaningful and understandable manner. The data would be presented in a graphical form which would be much easier for a user to analyze the data. The graphical view will be developed using SSRS (SQL Server Reporting Service) tool, The SSRS tool is a Reporting Service tool developed by Microsoft which is used to prepare and deliver a variety of interactive and printed reports.

Technologies:

Database: MSSQL

Frontend: Java J2EE, SSRS

Connectivity: JDBC

Operating system: Windows/Linux

Rough Timeline:

Task Id

Task Name

Time Required

1

Setup and Research

5 weeks

2

Developing the prototype using WEKA

4 weeks

3

Building a user application using Net Beans

3 weeks

4

Developing a graph view using SSRS

2 weeks

5

Integrating graph view and user application

3 weeks

6

Testing

2 weeks

7

Documentation

4 weeks