---
title: "KDD applied to WWW mining"
subject: "Data Mining"
chapter: 7
key: "semester_7/data_mining:kdd-in-www-mining"
source: https://www.bctengineer.com/q/semester_7/data_mining/kdd-in-www-mining
site: BCT Engineering
---

# KDD applied to WWW mining

Data Mining · Chapter 7: Advanced Applications

- Asked in: year not given, 2+8 marks
- Video: [Data Mining & Business Intelligence | Tutorial #1 | The KDD Process](https://www.youtube.com/watch?v=U-hUnxuaw7w) — RANJI RAJ
- Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14
- Page: https://www.bctengineer.com/q/semester_7/data_mining/kdd-in-www-mining
- Whole chapter: https://www.bctengineer.com/md/semester_7/data_mining/chapter-7

## Question

What is KDD? How do you apply KDD in World Wide Web mining?

## Short answer

KDD is the overall process of discovering useful knowledge, with steps cleaning, integration, selection, transformation, mining, evaluation and presentation. In web mining, selection means crawling pages and collecting logs, cleaning removes tags and noise, mining applies content, structure and usage algorithms, and evaluation filters trivial patterns.

## Answer

**KDD (Knowledge Discovery in Databases)** is the overall process of discovering useful, valid, previously unknown knowledge from large data. Data mining is the central step of this process.

#### KDD steps
1. **Data cleaning**: remove noise and inconsistent data.
2. **Data integration**: combine multiple sources.
3. **Data selection**: choose data relevant to the task.
4. **Data transformation**: normalise or aggregate into forms suitable for mining.
5. **Data mining**: apply algorithms to extract patterns.
6. **Pattern evaluation**: identify the interesting patterns by measures.
7. **Knowledge presentation**: visualise and report the knowledge.

#### Applying KDD to World Wide Web mining
| KDD step | In web mining |
| --- | --- |
| Selection | Resource finding: crawl and retrieve web pages, collect server logs |
| Cleaning | Remove HTML tags, stop words, image requests, robot accesses and failed requests |
| Integration | Combine page content, link structure, logs, user registration data |
| Transformation | Convert to features: term vectors (TF-IDF), link graph, sessions and user paths |
| Data mining | Content mining (classification, clustering), structure mining (PageRank, HITS), usage mining (association rules, sequence patterns) |
| Evaluation | Measure interestingness, filter out trivial patterns (e.g. every user visits the home page) |
| Presentation | Visualise navigation paths, recommendations, reports |

*Example*: for an online shop, logs are cleaned and divided into sessions, association rules find that users who view "laptop" pages also view "laptop bag" pages, and the shop recommends bags on laptop pages. The process is iterative: results may lead to returning to earlier steps.

## Check yourself

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

1. What is the role of data mining in KDD?
   A. It is the whole KDD process
   B. It is the central step in which algorithms extract patterns
   C. It is the final visualisation step
   D. It is the data cleaning step
   Answer: B. Data mining is one central step of the larger KDD process. (Key point: KDD)

2. Which order of KDD steps is correct?
   A. Cleaning, integration, selection, transformation, mining, evaluation, presentation
   B. Mining, cleaning, selection, evaluation
   C. Selection, presentation, mining, cleaning
   D. Integration, mining, transformation, cleaning
   Answer: A. The process runs from cleaning up to knowledge presentation. (Key point: KDD steps)

3. In web mining, what does the cleaning step involve?
   A. Crawling pages
   B. Removing HTML tags, stop words, image requests, robot accesses and failed requests
   C. Computing PageRank
   D. Visualising navigation paths
   Answer: B. Cleaning removes noise from content and logs. (Key point: Web mapping)

4. Which pattern would evaluation filter out as trivial?
   A. Laptop viewers also view laptop bags
   B. Every user visits the home page
   C. Users who log in view their orders
   D. Sessions longer than 30 minutes
   Answer: B. Evaluation filters uninteresting patterns such as 'everyone visits the home page'. (Key point: Evaluation)

## Related in Data Mining

- [Web log mining and WWW mining architecture](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aweb-log-mining-architecture) — Data Mining, chapter 7
- [Web mining, importance and techniques](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aweb-mining-importance-techniques) — Data Mining, chapter 7
- [Short note: WWW mining](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Awww-mining-note) — Data Mining, chapter 7
- [Data mining as a step in KDD](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Adm-as-kdd-step) — Data Mining, chapter 1
- [Web structure/content/usage mining; privacy](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aweb-structure-content-usage-privacy) — Data Mining, chapter 7
- [Web mining structure and challenges](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Aweb-mining-structure-challenges) — Data Mining, chapter 7

## Related in other subjects

- [Web mining with structure mining](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Aweb-mining-structure-mining) — Information System, chapter 7
- [Web mining importance and types](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Aweb-mining-importance-types) — Information System, chapter 7, asked 3×
- [Text mining vs web usage mining](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Atext-mining-vs-web-usage-mining) — Information System, chapter 7
- [Internet vs WWW; client-server paradigm](https://www.bctengineer.com/question.md?key=semester_7%2Fweb_technology%3Ainternet-vs-www-client-server) — Web Technology, chapter 1