Time Series Database Preprocessing for Data Mining Using Python

Abstract-Database Preprocessing for Data Mining Using Python lays the groundwork for data mining. Before the discovery of useful information/knowledge, the target data set must be properly prepared. Using Data Mining in census data has very high learning value and vast marketplace space. While census data are being collected and accumulated at dramatically high rates, concept hierarchy, one of the Data Mining techniques, can be used to reduce the data by collecting and replacing low level concepts with higher-level concepts. Data pre-processing lays the groundwork for data mining yet most researchers unfortunately, ignore it. This project walks through the process of analysing the characteristics of the daily minimum temperature dataset.

The unified data model is used as a standard representation for the incoming data so that it can be mined. It not only provides flexibility for data pre-processing but also reduce complexity and difficulty of preparation for mining customer survey data. Time series analysis is the preparatory step that is needed to develop the future values. Through the series analysis we can understand different structures of the inherent nature of the series so that we can create an accurate forecast. The data was decomposed into the base level, trend, seasonal and residue components. Finally, the data stationary status was checked using Dickey Fuller’s stationary test and the test shows that the data is stationary.