There are various Database attributes types in the Database management system. But before going through the types we must understand the meaning and purpose of attributes.
An attribute provides the characteristics of the entity. In other words, an attribute describes the property of an entity. Each entity consists of one or more attributes.
What Are Attributes? Different Database Attributes Types with Examples
Database Management System (DBMS) consists of ER model. The full form of ER model is the Entity-Relationship model. We use ER model to describe the data elements and their relation with the specified system.
The ER model consists of entities and attributes. An entity can be an object, person, or place. In the ER model, we represent Entity as rectangles. For example, In an organization, we can take employees, departments, executives, as an entity.
Attributes give us additional information about the entity. It describes the property of an entity. In the ER model, we represent the Attributes as an Eclipse. For example, If Employee is an entity, employee id, contact number, name, date of joining, etc can be the attributes of an employee.
There are 5 different types of attributes in DBMS.
Simple Attributes are independent attributes that cannot be classified further. In other words, it is also known as atomic attributes.
For example, a Student is an entity that consists of attributes Roll No, Age, Class. Here, we cannot divide the Roll no attribute into sub-attributes. Therefore, if we cannot divide the attribute further then it is a Simple Attribute.
When it is possible to divide the attributes into different components then that attribute is called a Composite Attribute. We divide Composite Attribute into sub-parts that form simple attributes.
For example, If Name is an attribute for Student entity. We can divide the Name attribute into first name, middle name, last name attributes. These sub-attributes that are classified from the composite Attribute works as Simple Attributes.
Single Valued Attributes
Attributes stores values that are used to describe the entity. The attributes which are able to store only one value are known as Single Valued Attributes. These attributes cannot store more than one value.
For example, The attributes of an Employee entity are Employee id, DOB, Gender. An employee has only one employee id which is unique and it also has a single date of birth. So these attributes can store only one value in it. Therefore, it is known as Single Valued Attributes.
The attributes which are able to store more than one value are known as Multi-Valued Attributes.
For example, let’s assume Email id and Contact No are the attributes of the Employee entity. An employee can provide more than one email id and contact no. Therefore multiple values can be stored in Multi-Valued Attributes.
The name itself describes the attribute. Derived attributes are those attributes that are derived from the value of another attribute.
For example, We can calculate the age from the date of birth value. Therefore Age attribute can be derived from the DOB attribute.
Every entity has a special attribute that holds a unique value to identify the entity in the entity set. The value of key attributes must be unique and cannot be used again.
For example, Employee id is the key attribute for employee entity, Roll No is the key attribute for Student entity, and Pincode is the key attribute for the place attribute.
Besides DBMS there are different database attributes types in Data Mining. Data Mining is a computational process of analyzing the data. It gathers more information about the data. It is also referred to as knowledge extraction of the data.
In Data Mining there are Data objects which act as an entity and these entities have various types of data attributes. A group of attributes of an entity forms a data object. It has a different concept as compared to DBMS. when a data object is ready to be used in a database, the data object is called data tuples.
There are 3 main phases in Data Mining. Data Pre-processing, Data Extraction, Data Evaluation. Data Mining uses data objects and attributes in the first phase of Data Pre-processing. The Database divides the attributes into two main categories.
There are 3 types of attributes that describe the quality of the entity.
Nominal attributes consist of names. It also describes the category or state of the attribute. It does not follow any order or sequence.
For example, Assume the Attributes is Colours, the values of this attribute can be Black, Brown, White.
Binary Attributes (B)
Binary attributes consist of only two values. For example, Pass and Fail, Agree and Disagree, etc.
There are two different types of Binary attributes. Symmetric Binary attributes occur when both the values are important. For instance, Gender has two values Male and Female both the values are equally important. Asymmetric Binary attributes occur when both the values are not important. For instance, In the Result attribute, Pass and Fail are not equally important.
Ordinal Attributes (O)
The values in the Ordinal Attributes must follow a meaningful sequence. Ordinal Attributes are attributes when the order of values is sequential and describes what is important.
For example, the attribute Grade has values A, B, C, D, E, F.
There are 3 different types of data that describe the quantity of the entity.
A numeric data consists of integers. It is a subcategory of Quantitative Attributes because it can be measurable. There are two different types of Numeric Data.
The interval-scaled attribute consists of numeric values. The difference between the two values is meaningful. We can add or remove the data at an interval but we cannot multiply the data.
The ratio-scaled attribute consists of values that are multiples or ratios of another value. The values must be in a sequence. We can also calculate the mean, median, and differences of the values.
Discrete data consist of both Numerical and Nominal values. The main feature of this data is that it has a finite set of values.
For example, Zipcode attributes consist of a finite set of numerical values.
Continuous data consist of an infinite set of values. The values of this data are in float type. For example, if height is an attribute the values will be 5.2, 6.4, 7.2, ….etc.
Attributes are describing the entity. It consists of values. We have also seen various types of attributes. In DBMS there are different types of attributes that store the values for the Entity. In the ER model, attributes are also important to describe the entity-relationship.
We have also seen the use of Attributes in data mining. The Pre-processing phase of Data Mining consists of Data objects and Attributes. Without attributes, we cannot define the Entity and it will be impossible to choose the entity in an entity set with unique attributes.
You may also like to read: Pros and Cons of Data Mining Explained