A Study on Handling Missing Values and Noisy Data using Weka Tool |
Author(s): |
| R. Vinodhini , Rajah Serfoji Govt. College (Autonomous), Thanjavur-5; A. Rajalakshmi, Rajah Serfoji Govt. College (Autonomous), Thanjavur-5; K. Fathima Bibi, Rajah Serfoji Govt. College (Autonomous), Thanjavur-5 |
Keywords: |
| Data Preprocessing, Data Cleaning, Filters, WEKA Tool, Classification, Missing Values, Noisy Data |
Abstract |
|
Many people treat data mining as a synonym for another popularly used term, Knowledge Discovery from Data, or KDD. Data preparation and preprocessing is the key to solve the problem. Data pre-processing is the important step in data mining process. Data pre-processing includes cleaning, normalization, transformation, feature extraction and selection, pattern evaluation, knowledge presentation. In this paper, we choose data cleaning as experimental study. The main objective of data cleaning is to reduce the time and complexity of mining process and increase the quality of datum in data warehouse. In this paper, we have used several methods and filters for removing missing values and noisy data, namely Replace Missing Values filter, Remove Misclassified Filter, Interquartile Range filter, Binning, Clustering, etc. The performances of the filters are measured by J48 Classification Algorithm. |
Other Details |
|
Paper ID: IJSRDV4I50673 Published in: Volume : 4, Issue : 5 Publication Date: 01/08/2016 Page(s): 1127-1131 |
Article Preview |
|
|
|
|
