Data Models and Relationships

Data Models

• Data models are conceptual tools used to describe data, data relationships, data semantics, and consistency constraints.
• Data models describe the structure or design of a database.
• Data models demonstrate how data is related, stored, retrieved, and modified in database management systems.
• Information can be depicted using symbols and terminology for organizational members to understand and comprehend.

Data Model Building Blocks

• An entity is anything about which data is collected and stored, which represents a particular type of object.
• An entity set is a set of entities of the same type that share the same properties.
• An attribute is a characteristic of an entity.
• Constraints are restrictions placed on data and ensure data integrity.

Relationships

• A relationship is an association among entities.
• Different types of relationships exist.
• A One to One (1:1) relationship describes that one entity is associated with only one other entity.
• A One to Many (1:M) relationship describes that one entity is associated with many other entities.
• A Many to Many (M:N) relationship describes that many entities are associated with many other entities.

Hierarchical Model

• Hierarchical models classify data into a tree-like structure with a single parent or root for each record.
• Sibling records are arranged in a certain sequence.
• The physical order is how the database is stored.
• Hierarchical models are useful for explaining real-world interactions.
• Hierarchical models depict a set of 1:M relationships between a parent and its children segments.

Advantages of the Hierarchical Model

• Simplicity and speed in traversing through a tree-like structure are benefits of the hierarchical model.
• Any change in the parent node is automatically reflected in the child node, so data integrity is maintained.

Disadvantages of the Hierarchical Model

• Complex relationships are not supported
• More than one parent node of the child node is unsupported, therefore complex relationship where a child node needs multiple parent nodes can't be represented.
• If a parent node is deleted, the child node is also automatically deleted.

Network Model

• The network model is an expansion of the hierarchical model.
• The network model replaces a hierarchical tree with a graph and allows a record to have several parents.
• The network model was the most prevalent model before the relational model.

Advantages of Network Model

• Data can be obtained more quickly in a network model compared to a hierarchical approach.
• Data may be more linked, with multiple paths to a certain node, aiding in versatile information access.
• Data integrity is maintained through parent-child connections, with parent record changes mirrored in the child record.

Disadvantages of Network Model

• The system may be increasingly complicated as additional relationships are added.
• User must have a full understanding of the system to operate it
• Any modification, such as an update, deletion, or insertion, is quite difficult.

Entity-Relationship Model

• The Entity-Relationship (ER) Model is a high-level data model diagram.
• It expresses a real-world problem in visual form so it is easier for stakeholders to follow.
• The ER diagram aids developers in understanding the system.
• The ER diagram is made up of three parts.

Advantages of Entity-Relationship Model

• Conceptually simple, making ER Models easy to build.
• Knowing the relationship between attributes and entities helps building the ER Diagram.
• The model is an Effective Communication Tool used widely by database designers to communicate.
• ER models map well to the relational model.
• ER models can also be converted to any other model like the network model or the hierarchical model, making conversion to any model easy.

Disadvantages of Entity-Relationship Model

• There is no industry standard for developing an ER model.
• One developer using notations not understood by other developers.
• Some information might be lost or hidden in the ER model.
• High level view implies loss of information might be hidden.

Relational Model

• The most common model in the Relational Model.
• Data is kept in the form of a two-dimensional table.
• All data is saved in the form of rows and columns.
• Tables are the foundation of a relational paradigm.
• Tables are referred to as relations

Advantages of Relational Data Model

• This model is more simple compared to the network and hierarchical model.
• Can be easily scaled as to add additional rows and columns.
• Structural independence is achieved when database structural changes do not change the way to access the data, and do not affect the Database Management System's capability to access the data.

Disadvantages of Relational Data Model

• The model requires more powerful hardware computers and data storage devices, introducing hardware Overheads.
• The model is very easy to design and use.
• Ease of design can lead to the development of a poor database which would slow down if the database grows.

Object-Oriented (OO) Model

• OO model paradigm contains a collection of objects/reusable software parts, with related characteristics and operations.
• Object-oriented databases are classified in a variety of types.
• Multimedia Database - contains various forms of media impossible to store in a relational database.
• Hypertext Database - Any object can link to any other object, effective for organizing a large amount of heterogeneous data, but is imperfect for numerical analysis.
• Object-oriented database models are also known as hybrid database models, because they integrate, but are not limited to tables.
• OODs are most often used with object-oriented programming languages like Java, Kotlin, C#, Node JS (React), and Swift.

Advantages of Object Oriented Data Model

• Complex data sets can be saved and retrieved quickly and easily.
• Object IDs are assigned automatically.
• It works effectively with object-oriented programming languages.
• Semantic content is added.
• Model supports for complex objects and visual representation includes semantic content.

Disadvantages of Object Oriented Data Model

• Object databases are not widely adopted.
• High Complexity can cause performance problems.
• High system overheads cause slower transactions.

Object-Relational (OR) Model

• Object-Relational (OR) Models are hybrids of the relational and object-oriented models.
• The paradigm was created to bridge the gap between the object-oriented and relational models.
• Can have many sophisticated features with the ability to create complicated data types based on our needs utilizing current data types.
• This paradigm is difficult to manage.
• Thorough understanding of this paradigm is essential.

OO vs OR Data Model

• Object-Oriented database
  • A database that represents information in the form of objects as inspired by Object-oriented programming.
  • Depends on OOP.
  • Relationships are represented by references via the object identifier (OID).
  • Handles larger and more complex data than RDBMS.
  • Less efficient.
  • Data management language is typically incorporated into a programming language like #C++.
• Object-Relational database
  • A database that depends on the Relational data model and Object-oriented data model.
  • Hybrid of relational and object-oriented model.
  • Connections between two relations are represented by foreign key attributes in one relation that reference the primary key of another relation
  • Handles comparatively simpler data.
  • Improvement and comparatively more efficient.
  • Includes data manipulation languages like SQL, QUEL, and QBE based on relational calculus.

Semi-structured Model

• The semi-structured model emerged from the relational model.
• There is not a difference between data and schema in the semi-structured model.
• Semi-structured models are often found in a Web-based data source.
• Some entities have missing or have additional attributes that allows for greater data storage flexibility, possessing qualities of flexibility.
• Email, XML, Markup languages, binary and executable files, TCP/IP packets, zipped files, data integrated from different sources, and web pages are a few examples of semi-structured data sources.