JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD B.Tech IV Year DATA MINING Previous Question Paper (July/August - 2022 - R18)

Data mining task primitives define the scope and requirements of a data mining process. They allow users to interact with the system to specify the data mining task. The key primitives include:

1. Task-relevant data: Specifies the portion of the database to be investigated, including attributes or dimensions of interest.
2. Kind of knowledge to be mined: Defines the specific function, such as characterization, discrimination, association, classification, clustering, or trend analysis.
3. Background knowledge: Utilizes concept hierarchies or user beliefs to guide the discovery process and express discovered patterns at multiple abstraction levels.
4. Pattern interestingness measures: Criteria used to evaluate the discovered patterns, such as support and confidence for association rules, or simplicity and certainty for classification.
5. Visualization and presentation of discovered patterns: Specifies the format in which the results should be displayed, such as decision trees, tables, graphs, or rules.

Data preprocessing is a critical phase in data mining that transforms raw, often noisy or incomplete data into a clean, structured format suitable for analysis. The primary steps are:

1. Data Cleaning: Addresses missing values (by ignoring, filling manually, or using mean/median imputation), smooths noisy data (using binning, regression, or clustering), and identifies/removes outliers.
2. Data Integration: Combines data from multiple sources (databases, data cubes, or files) into a coherent data store. This step involves resolving schema integration conflicts and identifying entity duplicates.
3. Data Transformation: Converts data into forms appropriate for mining. This includes normalization (scaling values to a specific range like 0 to 1), discretization (reducing the number of values by grouping them into intervals), and attribute construction (creating new features from existing ones).
4. Data Reduction: Reduces the volume of data while maintaining analytical integrity. Techniques include dimensionality reduction (e.g., Principal Component Analysis), numerosity reduction (e.g., histograms, sampling, or data cube aggregation), and data compression.