TY  - JOUR
TI  - A Survey on Different Techniques for Handling Missing Values in Dataset
AU  - Sukanya Gupta
AU  -  Dr. Manoj Kumar Gupta
JO  - International Journal of Scientific Research in Computer Science, Engineering and Information Technology
PB  - Technoscience Academy
DA  - 2018/04/25
PY  - 2018
DO  - https://doi.org/10.32628/IJSRCSEIT
UR  - https://ijsrcseit.com/CSEIT411849
VL  - 4
IS  - 1
SP  - 295	
EP  - 301
AB  - Abundant of information is being collected and stored every day. That data can be used to extract interesting patterns. The data that we collect is incomplete normally. Now, using that data to extract any information may give misleading results. So, before using that we need to pre process the data to eradicate the abnormalities. In case of small percentage of missing values, those instances can be ignored but in case of large amounts, ignoring them won’t give desired results. Large amount of missing spaces in a dataset is a big problem faced by researchers as it can lead to many problems in quantitative research. So, before performing any data mining techniques to extract some valuable information out of a dataset some pre processing of data can be done to avoid such fallacies and thereby improving the quality of data. To handle such missing values many techniques have been proposed since 1980.The simplest technique is to ignore the records containing missing values other technique include imputation, which involves replacing those missing spaces with some estimates by doing certain computations. This would increase the quality of data and would improvise prediction results. This paper gives a review on different types of techniques available for handle missing data like k nearest neighbor (KNN), multiple imputation, case deletion, most common method (MC) etc.