Time Series Data Mining and Forecasting Using SQL Server 2008 - Thesis Example

The first chapter provides an introduction to data mining concepts and an overview of how data processing is carried out in Ghana’s agricultural sector. The second chapter describes the thesis objectives, methodology and the formulated hypothesis. The third chapter presents a review literature on agriculture in Ghana and historical trends of data mining. The fourth chapter presents the analytical part of this thesis. The fifth chapter constitutes the results and discussions pertaining to this project.

Chapter six contains the conclusion and finally, chapter seven contains the list of references used in this thesis. CHAPTER ONE1. IntroductionData mining refers to the analysis of data to discover the hidden patterns through the use of hidden methodologies. Though the concept has been in place for quite a long time, it only became more prevalent after the advent of larger data sets inside organizations. Big data sets resulted into explosion in the utilization of extensive data mining techniques because of the increasing variation and size of the nature and content of the stored information.

From the larger data sets, obtaining simple and straightforward statistics from the system under use will never suffice in good decision making hence the need for discovery of the hidden patterns. The present economy is increasingly becoming information based. This state has brought about changes in the operation environment for businesses or organizations. Consequently, it has also changed the manner of collection and analysis of business data. Bearing in mind the widespread embracement of information technology, a greater percentage of data is collected in the online and real time environments that bring forth large volumes of data.

Experts can be in a position of aggregating such time ordered data using a desirable time interval, producing great volumes of time series data that is equally spaced. The data can then be explored followed by analysis through the use of multiple important tools or methodologies that are developed using modernized time series analysis. The storage or management technology associated with the relational databases is sufficient for a number of data mining applications which are below 50 GB.