Data Warehouse and Data Mining - Important Questions with Answers

Short Answer Questions

Q: What is a Data Warehouse?

A: A Data Warehouse is a centralized repository used to store data from multiple sources. It supports analytical

reporting, structured queries, and decision-making.

Q: Define OLAP and OLTP.

A: OLAP (Online Analytical Processing) is used for complex analysis and reporting. OLTP (Online Transaction

Processing) supports daily transactions like insert, update, delete.

Q: What is data cleaning?

A: It is the process of fixing or removing incorrect, corrupted, or incomplete data within a dataset to improve data quality.

Q: What is ETL in data warehousing?

A: ETL stands for Extract, Transform, Load. It extracts data from source systems, transforms it into a suitable format,

and loads it into the data warehouse.

Q: Define metadata.

A: Metadata is data that describes other data. In data warehousing, it includes information about data source, structure,

transformations, and access methods.

Q: What is dimensional modeling?

A: It is a design concept used in data warehouses to structure data into fact and dimension tables for easier retrieval and

analysis.

Q: Name types of OLAP systems.

A: The three main types are: MOLAP (Multidimensional OLAP), ROLAP (Relational OLAP), and HOLAP (Hybrid OLAP).

Q: What is a snowflake schema?

A: It is a type of schema where dimension tables are normalized into multiple related tables, resembling a snowflake

structure.

Q: What is a star schema?

A: It consists of a central fact table linked to dimension tables. It is simple and optimized for querying large data.

Q: What is data cube?

A: A data cube is a multi-dimensional array of values used in OLAP to represent data along some measure of interest.

Q: Define clustering.
 Data Warehouse and Data Mining - Important Questions with Answers

A: Clustering is a data mining technique used to group similar data points into clusters based on characteristics.

Q: What is data mining?

A: Data mining is the process of extracting useful information and patterns from large datasets using statistical and

computational methods.

Q: What is the difference between supervised and unsupervised learning?

A: Supervised learning uses labeled data to train models, while unsupervised learning works with unlabeled data to

identify patterns.

Q: Define association rule.

A: Association rule shows how items are related to each other in large datasets. Example: {Milk} => {Bread}.

Q: What is a decision tree?

A: It is a tree-like model used for classification and prediction. It splits data into branches based on conditions.

Long Answer Questions

Q: Explain the architecture of a data warehouse with a neat diagram.

A: The architecture includes:

1. Data Sources (Operational DBs, Flat files)

2. ETL Process

3. Staging Area

4. Data Storage (Warehouse)

5. Metadata Repository

6. Data Marts

7. Query Tools

This structure supports data consolidation and analysis.

Q: Compare OLAP and OLTP with examples.

A: OLTP handles routine transactions like banking or online purchases; it's optimized for write operations.

OLAP supports complex analytical queries like sales forecasting and is optimized for reading large volumes of data.

Q: Describe the steps in the ETL process.

A: 1. Extract: Get data from multiple sources.

2. Transform: Cleanse and convert data formats.

3. Load: Store the transformed data in a warehouse.

Q: Explain star schema and snowflake schema with diagrams.

A: Star Schema: Central fact table connected to denormalized dimension tables.

Snowflake Schema: Fact table connected to normalized dimension tables with multiple levels.

Star is faster; Snowflake saves storage.

Q: Discuss different types of OLAP (ROLAP, MOLAP, HOLAP).

A: ROLAP: Uses relational DBs, handles large data.

MOLAP: Uses multidimensional cubes, faster querying.

HOLAP: Hybrid approach using both MOLAP and ROLAP features.

Q: Write a note on data preprocessing techniques.

A: Includes:

- Data Cleaning

- Data Integration

- Data Transformation

- Data Reduction

These steps ensure high data quality before analysis.

Q: What are fact and dimension tables? Explain with examples.

A: Fact Table: Contains numeric data for analysis (e.g., sales amount).

Dimension Table: Contains descriptive data (e.g., product, region). They help slice data from the fact table.

Q: Describe the concept and advantages of data marts.

A: Data marts are smaller, subject-specific subsets of a data warehouse. They are faster, easier to maintain, and

provide focused analytics (e.g., marketing data mart).

Q: Explain the role of metadata in data warehousing.

A: Metadata describes how, when, and by whom data is collected and formatted. It improves understanding, data

quality, and usage in a warehouse.

Q: What are the characteristics of a data warehouse?

A: 1. Subject-Oriented

2. Integrated

3. Time-Variant

4. Non-Volatile
 Data Warehouse and Data Mining - Important Questions with Answers

These features make data warehouses effective for analytical queries.

Q: What is data mining? Explain its process with a diagram.

A: Data mining is the process of extracting patterns from large datasets. The process includes data collection,

preprocessing, model building, evaluation, and deployment.

Q: Explain classification and prediction techniques with examples.

A: Classification assigns data to categories (e.g., spam detection). Prediction estimates future values (e.g., stock prices).

Techniques include Decision Trees, SVM, Regression.

Q: What is clustering? Explain k-means clustering algorithm.

A: Clustering groups similar data. K-means assigns data points to k clusters based on distance from centroids. It repeats

until clusters stabilize.

Q: Describe decision tree induction with example.

A: A decision tree splits data based on attribute values. Example: If income > 50k, then 'Approved', else 'Rejected'. It

continues until classification is done.

Q: Explain association rule mining and Apriori algorithm.

A: Apriori identifies frequent itemsets using minimum support. Then, rules are generated with confidence values.

Example: {diapers} => {beer} with 70% confidence.

Q: Write a note on web mining, text mining, and spatial mining.

A: Web Mining: Extracts patterns from web data.

Text Mining: Derives insights from text sources.

Spatial Mining: Analyzes spatial/geographical data.

Q: Discuss challenges and issues in data mining.

A: Includes data quality, data integration, scalability, privacy, and algorithm complexity. Addressing these ensures

accurate and ethical mining results.

Q: Compare classification and clustering with examples.

A: Classification: Supervised, e.g., Email = spam/ham.

Clustering: Unsupervised, e.g., grouping customers by purchasing behavior.

Q: Explain any two applications of data mining.

A: 1. Market Basket Analysis: Finding associations between products.

2. Fraud Detection: Identifying unusual transaction patterns.

Q: What are the stages in the data mining process?

A: 1. Business Understanding

2. Data Understanding

3. Data Preparation

4. Modeling

5. Evaluation

6. Deployment