---
title: "Data sampling methods"
subject: "Data Mining"
chapter: 2
key: "semester_7/data_mining:data-sampling"
source: https://www.bctengineer.com/q/semester_7/data_mining/data-sampling
site: BCT Engineering
---

# Data sampling methods

Data Mining · Chapter 2: Data Pre-processing

- Asked in: year not given
- Video: [Types of Sampling | Introduction to Data Mining | Part 12](https://www.youtube.com/watch?v=4EVqbRnPLq8) — Data Science Dojo
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_7/data_mining/data-sampling
- Whole chapter: https://www.bctengineer.com/md/semester_7/data_mining/chapter-2

## Question

What do you mean by Data Sampling? Explain various ways of sampling the data.

## Short answer

Data sampling selects a representative subset to speed up mining cheaply. Methods include simple random (with/without replacement), stratified (from each group proportionally), cluster (whole groups), systematic (every k-th tuple), and progressive (adaptive) sampling.

## Answer

**Data sampling** selects a subset of the data objects to represent the whole data set. Mining a sample is faster and cheaper, and it works well if the sample is *representative*, i.e. has approximately the same properties of interest as the full data. Sampling is used both for data selection and for data reduction. Sample size matters: too small a sample misses patterns, while too large a sample reduces the savings.

#### Sampling methods
Let N be the number of tuples and n the sample size.
1. **Simple random sampling without replacement (SRSWOR)**: draw n distinct tuples; each tuple has equal probability 1/N of being drawn and, once drawn, cannot be drawn again.
2. **Simple random sampling with replacement (SRSWR)**: a drawn tuple is put back, so it can be selected more than once.
3. **Stratified sampling**: divide the data into disjoint groups (strata), such as age groups or classes, and draw a random sample from each stratum, usually in proportion to its size. It guarantees that small groups are represented and is used on skewed data.
4. **Cluster sampling**: tuples are grouped into clusters (e.g. cities, disk pages), and a random set of whole clusters is chosen.
5. **Systematic sampling**: select every k-th tuple after a random start (k = N/n).
6. **Progressive (adaptive) sampling**: start with a small sample and increase its size until the model accuracy stops improving.

| Method | Best used when |
| --- | --- |
| Simple random | Population is homogeneous |
| Stratified | Groups differ, data skewed |
| Cluster | Population is naturally grouped, cheap to access whole groups |
| Systematic | Data in a list without periodic pattern |

## Check yourself

The answer's key points as multiple-choice questions, with the correct option and why.

1. When does mining on a sample work well?
   A. When it is the smallest possible
   B. When it has only outliers
   C. When it is the first row
   D. When the sample is representative of the full data
   Answer: D. A representative sample has about the same property of interest. (Key point: Representative sample)

2. How do SRSWOR and SRSWR differ?
   A. In SRSWOR a drawn tuple cannot be drawn again; in SRSWR it can
   B. SRSWOR allows repeats; SRSWR does not
   C. Neither allows repeats
   D. Both pick every k-th tuple
   Answer: A. With replacement, a tuple is returned and may be selected again. (Key point: SRSWOR vs SRSWR)

3. When is stratified sampling best?
   A. Only when data is homogeneous
   B. When groups differ or data are skewed, to guarantee small groups are represented
   C. When picking whole clusters
   D. When selecting every k-th tuple
   Answer: B. Random samples are drawn from each stratum, usually proportionally. (Key point: Stratified sampling)

4. In systematic sampling with N tuples and sample n, what is k?
   A. k = n/N
   B. k = N x n
   C. k = N/n
   D. k = N - n
   Answer: C. Select every k-th tuple after a random start. (Key point: Systematic sampling)

## Related in Data Mining

- [Data warehouse schemas](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Adw-schema-types) — Data Mining, chapter 1
- [Define data warehousing and creation process](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Adw-creation-process) — Data Mining, chapter 1
- [Dimension of data; curse of dimensionality](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Adimension-curse) — Data Mining, chapter 2
- [Skewed data and skewness](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Askewed-data) — Data Mining, chapter 2

## Related in other subjects

- [Sampling distribution of mean; standard error](https://www.bctengineer.com/question.md?key=semester_5%2Fprobability_and_statistics%3Asampling-distribution-mean-se-finite) — Probability and Statistics, chapter 4, asked 2×
- [Sampling distribution of proportion; parameter vs statistic](https://www.bctengineer.com/question.md?key=semester_5%2Fprobability_and_statistics%3Aparameter-vs-statistic-sampling-proportion) — Probability and Statistics, chapter 4
- [Sub-sampling; sampling sinc(5πt) at 10 Hz](https://www.bctengineer.com/question.md?key=semester_6%2Fcommunication_systems%3Asubsampling-sinc5pit-10hz) — Communication Systems, chapter 4
- [Aperture effect correction; bandpass sampling 80–115 kHz](https://www.bctengineer.com/question.md?key=semester_6%2Fcommunication_systems%3Aaperture-correction-bandpass-80-115) — Communication Systems, chapter 4