This document discusses methods for handling noisy data in data mining. It describes how noise can be introduced through faulty data collection, entry, transmission or technology limitations. It then outlines three approaches to handling noisy data: binning, clustering and regression. Binning involves partitioning data into bins of equal width or depth and then smoothing the data by taking the bin mean, median or boundaries. Clustering can detect outliers by organizing similar values into clusters. Regression fits data to a linear or multidimensional surface to smooth out noise. Examples are provided to illustrate binning for data smoothing.

1. G H Patel College of Engineering & Technology
Submitted By: -
Vivek Gandhi
(140110107017)
Gujarat Technological University
Submitted To: -
Ms. Manpreet Bagga

2. Data Mining
Handling Noisy Data

3. Noisy Data: -
•Noise: - Random error or variance in a measured variable
or we can say meaningless data.

4. Incorrect attribute values may due to: -
•Faulty data collection
•Data entry problems
•Data transmission problems
•Technology limitation
•Inconsistency in naming convention

5. How to Handle Noisy Data?
•Binning method
•Clustering
•Regression

6. Binning Method: -
• First sort data and partition
• Then one can smooth by bin mean, median and boundaries.
•Equal-width (distance) partitioning:
• It divides the range into N intervals of equal size: uniform grid
• If A and B are the lowest and highest values of the attribute, the
width of intervals will be: W = (B-A)/N.
• The most straightforward
• Skewed data is not handled well.

7. Continue…….
•Equal-depth (frequency) partitioning:
• It divides the range into N intervals, each containing approximately same
number of samples
• Good data scaling
• Managing categorical attributes can be tricky.

8. Binning Methods for Data Smoothing
* Sorted data for price (in dollars): 4, 8, 9, 15, 21, 21, 24, 25, 26, 28, 29, 34
* Partition into bins:
- Bin 1: 4, 8, 9, 15
- Bin 2: 21, 21, 24, 25
- Bin 3: 26, 28, 29, 34
* Smoothing by bin means:
- Bin 1: 9, 9, 9, 9
- Bin 2: 23, 23, 23, 23
- Bin 3: 29, 29, 29, 29
* Smoothing by bin boundaries:
- Bin 1: 4, 4, 4, 15
- Bin 2: 21, 21, 25, 25
- Bin 3: 26, 26, 26, 34

9. Cluster Analysis: -
• Outliers may be detected by clustering, where similar values
are organized into group of cluster.

11. Example:-
• Data:- 4, 8, 15 ,21 ,21 ,24 ,25 ,28 ,34
1. Partition into equidepth
Bin1:- 4 ,8 ,15
Bin2:- 21 ,21 ,24
Bin3:- 25 ,28 ,34
2. Smooth by bin
Bin1:- 9 ,9 ,9
Bin2:- 22 ,22 ,22
Bin3:- 29 ,29 ,39

12. Example:-
3. Smooth by Boundaries
Bin1:- 4 ,4 ,15
Bin2:- 21 ,21 ,24
Bin3:- 25 ,25 ,34

13. Regression: -
• Here data can be smoothed by fitting the data to a function.
• Linear regression involves finding the “best” line to fit two attributes,
so that one attribute can be used to predict the other.
• Multiple linear regression is an extension of linear regression, where
more than two attributes are involved and the data are fit to a
multidimensional surface.

14. Continue…….