JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD
JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD MCA 2020 DATA WAREHOUSING AND DATAMINING Previous Question Paper (October/ November - 2020 - R17)
Answer any five questions
All questions carry equal marks
Main
Define data mining. Explain the steps in applying data mining to any database or data warehouse and the significance of data mining primitives.
Answer
Data mining is the process of discovering meaningful patterns, correlations, trends, and anomalies within large datasets using techniques from statistics, machine learning, and database systems to extract actionable knowledge.
The process of applying data mining follows these systematic steps:
- Data Cleaning: Removing noise, handling missing values, and resolving inconsistencies in the raw data.
- Data Integration: Combining data from multiple sources, such as multiple databases or data warehouses, into a coherent data store.
- Data Selection: Retrieving relevant data from the database that is necessary for the specific analysis task.
- Data Transformation: Converting data into forms appropriate for mining by performing operations like normalization, aggregation, or generalization.
- Data Mining: Applying intelligent methods and algorithms to extract patterns or models from the prepared data.
- Pattern Evaluation: Identifying truly interesting patterns representing knowledge based on specific measures like support, confidence, or interestingness thresholds.
- Knowledge Presentation: Visualizing and presenting the mined knowledge to the user using techniques like charts, tables, or reports.
Data mining primitives are essential components that allow users to interactively specify the mining task. Their significance lies in providing a flexible framework that defines:
- Task-relevant data: The specific portion of the database to be mined.
- Kind of knowledge to be mined: Specifying the function (e.g., classification, clustering, or association).
- Background knowledge: Using concept hierarchies to guide the mining process.
- Interestingness measures: Criteria to filter and rank the discovered patterns.
- Visualization techniques: How the results should be displayed.
These primitives enable a user-centered approach, ensuring the mining process is guided by domain expertise and specific business objectives rather than blind automated searching.
What is the use of principal component analysis in data preprocessing?
Answer
Principal Component Analysis (PCA) is a dimensionality reduction technique used in data preprocessing to transform a large set of variables into a smaller set of uncorrelated variables called principal components, while retaining as much variance as possible. Its primary uses include:
- Noise Reduction: By discarding components with low variance, PCA helps filter out noise and irrelevant features from the dataset.
- Mitigation of Multicollinearity: PCA eliminates correlations between input variables, which is essential for algorithms sensitive to redundant features.
- Computational Efficiency: Reducing the number of dimensions significantly lowers the computational cost and memory requirements for training machine learning models.
- Visualization: It allows for the projection of high-dimensional data into 2D or 3D space, facilitating the identification of patterns, clusters, or outliers.
- Overfitting Prevention: By simplifying the model input, PCA helps prevent overfitting, leading to better generalization on unseen data.
How to handle missing values in data mining process?
Answer
Sign in to view the complete answer paper.
With the help of a neat sketch describe the three-tier architecture of data warehouse.
Answer
Sign in to view the complete answer paper.
Differentiate between dependent and independent data marts.
Answer
Sign in to view the complete answer paper.
Why are data models insufficient for data warehouse? Explain snowflake schema and fact constellation schema for data warehouse.
Answer
Sign in to view the complete answer paper.
Find strong association rules from the following transactional data using Apriori algorithm. Let support=30%, confidence=40%
| TID | List of Items |
|---|---|
| T1 | Soap, shampoo, body lotion, toothpaste |
| T2 | Mug, bucket, soap, scrub |
| T3 | Dettol liquid, bucket, shower caps, soap |
| T4 | Tooth brush, foot scrapper, shower caps |
| T5 | Harpic, Dettol liquid, shower gel, shower caps |
| T6 | Soap, body wash, toothpaste |
Answer
Sign in to view the complete answer paper.
What is the role of information gain in decision tree induction?
Answer
Sign in to view the complete answer paper.
Why is k-nearest neighbor classifier known as lazy learner?
Answer
Sign in to view the complete answer paper.
What is an outlier? What are the applications of outlier analysis? Demonstrate deviation based approach for outlier detection.
Answer
Sign in to view the complete answer paper.
What is the objective of Page Ranking algorithm? Explain it as web structure mining algorithm.
Answer
Sign in to view the complete answer paper.