JAWAHARLAL NEHRU TECHNOLOGICAL UNIVERSITY HYDERABAD MCA 2017 DATA WAREHOUSING AND DATA MINING Previous Question Paper (August - 2017 - R13)

Data mining functionalities are used to specify the kind of patterns to be found in data mining tasks. The primary functionalities include:

• Characterization and Discrimination: Characterization summarizes the general features of a target class of data, while discrimination compares the target class with one or more contrasting classes.
• Mining Frequent Patterns, Associations, and Correlations: This involves discovering patterns that occur frequently in data, such as association rules (e.g., 'if a customer buys bread, they also buy butter') and correlations between variables.
• Classification and Regression: Classification builds models to predict categorical class labels (e.g., 'yes' or 'no'), whereas regression models continuous-valued functions to predict numerical values.
• Clustering: This is an unsupervised learning process that groups data objects into clusters based on similarity, such that objects within a cluster are more similar to each other than to those in other clusters.
• Outlier Analysis: This identifies data objects that do not comply with the general behavior or model of the data, which is crucial for fraud detection and error identification.

A data mart is a subset of a data warehouse that is focused on a specific functional area, department, or business line, such as sales, marketing, or finance. Unlike a data warehouse, which provides an enterprise-wide view of data, a data mart is designed to meet the specific analytical needs of a particular group of users, offering faster access to relevant data and improved performance for departmental queries.

Data marts can be categorized as dependent (extracted from an existing enterprise data warehouse) or independent (built directly from operational sources without a central warehouse).

For example, a large retail company may have a central data warehouse containing all transactional data. The Sales Department might create a dedicated data mart that contains only historical sales records, regional performance metrics, and product inventory levels for the last three years. This allows the sales team to run complex analytical reports on their specific KPIs without querying the massive, multi-terabyte enterprise data warehouse, resulting in faster response times and simplified data management.