ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Implementation of Preprocessing Techniques in Datamining?

Journal: International Journal of Computer Science and Mobile Computing - IJCSMC (Vol.3, No. 5)

Publication Date:

Authors : ; ;

Page : 464-471

Keywords : Discretization; Correlation; Normalization; Euclidean distance; Cosine similarity;

Source : Downloadexternal Find it from : Google Scholarexternal


carefully screened can produce misleading results. Thus, the raw data needs to pre-process before doing data mining. And often-times, this step can take considerable amount of processing time. Usually, data from experiments are not suitable for doing data mining tasks. Because of the raw data may contain out-of-range-values, impossible data combination or missing value etc. Analyzing data without being Data pre-processing includes cleaning, normalization, transformation, feature selection and extraction etc. The product of data pre-processing is the final training data set. In our research, we do discretization, calculating similarity or distance between objects, normalization, and find a correlation between objects or attributes in a data set to gain better analyze before main pre-processing steps.

Last modified: 2014-05-21 21:17:55