--- title: "Data mining and business knowledge" subject: "Information System" chapter: 4 key: "semester_8/information_system:data-mining-business-knowledge" source: https://www.bctengineer.com/q/semester_8/information_system/data-mining-business-knowledge site: BCT Engineering --- # Data mining and business knowledge Information System · Chapter 4: Decision support and Intelligent systems - Asked in: 2069 Bhadra (old course), 2+2+2 marks - Video: [What is Data Mining?](https://www.youtube.com/watch?v=7rs0i-9nOjo) — IBM Technology - Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14 - Page: https://www.bctengineer.com/q/semester_8/information_system/data-mining-business-knowledge - Whole chapter: https://www.bctengineer.com/md/semester_8/information_system/chapter-4 ## Question What do you mean by Data Mining? How is it related to Data Warehouse? Discuss the process of generating business knowledge using data mining techniques. ## Short answer Data mining automatically discovers hidden patterns in large data. The process: define business goal, select and preprocess data from warehouse, transform, apply mining techniques (classification, clustering), evaluate patterns, and deploy knowledge for decisions. ## Answer #### Data mining **Data mining** is the process of automatically discovering hidden, valid and useful patterns, trends and relationships from large volumes of data using statistical, machine learning and database methods. #### Relation with a data warehouse A **data warehouse** stores integrated, cleaned, historical data from many sources. Data mining works best on this data because it is consistent, large and covers time. The warehouse is the data source; mining is the analysis layer on top of it (along with OLAP). Findings may be stored back to the warehouse. #### Process of generating business knowledge (KDD steps) ```text Business goal -> Data selection -> Cleaning -> Transformation -> Mining -> Evaluation -> Deployment ``` 1. **Define the business problem** - e.g. reduce customer churn. 2. **Select data** from the warehouse relevant to the problem. 3. **Pre-process and clean** - remove noise, handle missing values, remove duplicates. 4. **Transform** - integrate, aggregate, create attributes, reduce dimensions. 5. **Apply mining techniques** - classification, clustering, association rules, regression, anomaly detection. 6. **Evaluate patterns** - check accuracy, validity and usefulness with domain experts. 7. **Present and deploy knowledge** - reports and visualisation; use in decision making, such as targeted offers. 8. **Monitor and refine** the model as data changes. Example: mining sales data finds that customers who buy a laptop often buy a bag within a week, so the shop bundles them. ## Check yourself The answer's key points as multiple-choice questions, with the correct option and why. 1. What does data mining do? A. Stores daily transactions B. Only prints reports C. Automatically discovers hidden, valid and useful patterns in large volumes of data D. Deletes old data Answer: C. Mining discovers hidden patterns using statistics, machine learning and database methods. (Key point: Definition) 2. What is the first step in generating business knowledge? A. Deploy the model B. Clean data C. Apply mining D. Define the business problem, e.g. reduce churn Answer: D. The process begins with the business goal. (Key point: First step) 3. Which sequence is correct? A. Select data, pre-process/clean, transform, mine, evaluate, deploy B. Mine, select data, clean, deploy, evaluate C. Deploy, mine, clean D. Evaluate, select, mine Answer: A. KDD follows selection, cleaning, transformation, mining, evaluation, deployment. (Key point: Order of steps) 4. Why are patterns evaluated before deployment? A. To make the data larger B. To check accuracy, validity and usefulness with domain experts C. To remove the warehouse D. To skip decisions Answer: B. Not every discovered pattern is valid or useful. (Key point: Evaluation) 5. Why is a warehouse ideal for mining? A. It is small and raw B. It holds only current inconsistent data C. Its data is consistent, large and covers time D. It cannot be queried Answer: C. Cleaned integrated historical data suits mining. (Key point: Relation) ## Related in Information System - [Data warehouse and data mining](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Adata-warehouse-data-mining) — Information System, chapter 4, asked 2× - [Short note: data mining and warehousing](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Adata-mining-and-warehousing-note) — Information System, chapter 4 - [Planning for business/IT strategies](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Aplanning-business-it-strategies) — Information System, chapter 5 - [Cloud technology for small business](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Acloud-for-small-business) — Information System, chapter 8 - [Knowledge management and KM triangle](https://www.bctengineer.com/question.md?key=semester_8%2Finformation_system%3Aknowledge-management-hierarchy) — Information System, chapter 4 ## Related in other subjects - [National data warehouse process architecture](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Anepal-national-data-warehouse) — Data Mining, chapter 1 - [Typical process flow in warehouse](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Adw-process-flow) — Data Mining, chapter 1 - [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