Introduction to Data Models

Introduction to Data Models

• Data models are simplified representations of complex real-world data structures.
• They are typically graphical in nature.
• They facilitate interaction among designers, programmers, and end-users.
• Data models organize data for diverse user needs.

Data Model Building Blocks

• Entity: Anything about which data is collected and stored (e.g., customer, product).
• Attribute: A characteristic of an entity (e.g., customer name, product price).
• Relationship: Describes the association among entities (e.g., a customer purchases a product).
• Constraint: A restriction on data or user-defined structures, limiting attributes' behaviors (e.g., NOT NULL, CHECK, PRIMARY KEY).
• Common entities include person, vehicle, building, plant, animal, etc.
• Common attributes include name, family, scientific name, etc.
• Relationships include 1:1, 1:M, M:N (one-to-one, one-to-many, many-to-many).
• Specific examples include relationships in business rules about employees and departments, indicating that one department can have many employees, but an employee can only be assigned to one.

Business Rules

• Business rules are brief, precise descriptions of policies, procedures, or principles within an organization.
• These rules apply to organizations storing and utilizing data to generate information.
• Descriptions are clear and precise.
• Rules should be documented, kept up-to-date, easy to understand, and widely distributed.
• They describe characteristics of data as viewed by the company and correspond to database table (ERD).
• These rules are derived from corporate sources such as managers, policy-makers, and documentation like standards.
• Examples include requirements that a department have at least one employee, and an employee must be attached to only one department.

Discovering Business Rules

• Sources of business rules include company managers, policy-makers, department managers, documented procedures, standards, and operational manuals.
• Direct interviews with end-users are also essential.
• Nouns in business language typically translate into entities.
• Verbs usually translate into relationships.
• Fact-finding techniques like interviews and questionnaires gather information about requirements, preferences and the system.
• The process aims to capture crucial facts for building databases, focusing on both current and future systems.
• Examples of business rules include varying prices for customers with different statuses (bronze, silver, gold) based on the items they want to purchase.

Data Model Components

• Standardize company's view of data.
• Act as a communication tool between users and designers, fostering understanding of data's nature, role, and scope along with business processes involved.
• Data should be correctly organized based on participation rules and constraints.
• Ensuring an accurate data model is crucial.

Fact Finding Techniques

• Techniques such as observations of documents, questionnaires, interviews, and research, and observation of organizational operations to acquire specifics on the requirements are crucial for gathering facts about the environment.

Evolution Of Data Models

• Hierarchical Model: Data organized in a tree-like structure.
  • Data represented by records related to each other via pointers.
  • Topmost record is called the root, lower records are children.
  • The top-down hierarchical relationship describes dependencies like parent/child between records.
  • Developed in 1960s to manage complex data like in manufacturing projects.
• Network Model: Set-based approach, composed of at least two record types, or sets (owner, member).
  • A collection of records represented in 1:M relationships.
  • Each parent can have many child records, while a child can have more than one parent record.
  • Developed in 1970s to manage complex data relations more efficiently, providing a set-oriented approach.
• Relational Model: Data organized into tables with rows and columns.
  • Related by sharing common attributes.
  • Represents relations amongst entities and attributes.
  • Stores a collection of related entities and resembles a file.
  • A powerful example of a relational table includes a table of biscuits with attributes like biscuit number, name, flavor, price, and an image, allowing data to be linked efficiently for real world uses.
  • Developed by Codd in 1970, utilizing the mathematical concept of relations.
• Object-Oriented Model: Data and relationships in a single object structure (data model - OODM).
  • An abstraction that models real world entities.
  • Attributes describe the object's properties.
  • A shared structure (attributes) and related behaviors are grouped in classes; the objects with similar attributes and behaviors are ordered in a class hierarchy.
  • The concept of inheritance allows an object within the class hierarchy to gain access to inherited attributes and methods of the higher classes in the hierarchy.

Degrees of Data Abstraction

• Classifies data models starting with a broad abstraction, then progressively narrowing the focus to increasing degrees of detail.
• Database system purpose is to provide a user-friendly abstract view of data, concealing storage details.
• ANSI/SPARC Framework defines three degrees: External, Conceptual, and Internal.
  • External: End-user view.
  • Conceptual: Entire database, with logical structures and relationships (using ER Model).
  • Internal: Specific DBMS representation (hardware/software).

Summary

• Data models serve as simplified portrayals of intricate real-world data environments.
• Data modeling components include entities (objects), attributes (properties), relationships (associations), and constraints (rules).
• Models progress through hierarchical, network, relational, and object-oriented stages.
• Data modeling requirements stem from different perspectives (global versus local), and degrees of abstraction.