Preprocessing Farmer Query Data Using Classic Method and Building Classifier Model

Being important preliminary step, preprocessing is critical phase in text mining and other related fields. Real world data contains errors of varying magnitude with multiple interrelated issues. Data preprocessing used to transform it into a form, which is readable, acceptable by tools, data that is free from ambiguities, duplicity. In this research work, we are dealing with farmer query data set, which is kind of text data, structured in tabular form. In case of text data, before any meaningful information retrieval, preprocessing techniques are applied on the target data set to reduce the size of the data set which will increase its effectiveness. The objective of our work is to analyze the issues of preprocessing operation such as tokenization, formatting, stop word removal for our text data. After preprocessing operations , further we have used logistic classifier to binary classify and model the farmer dataset. Logistic classifier gives good accuracy results and thus proves machine learning role in farmer query classification area.