Database-Conceptual Structure design

Conceptual structure Design

What is conceptual structure design
The process of abstracting the user requirement analysis into the information structure i.e. the conceptual model is the conceptual structure design
The conceptual structure is the common foundation of various data models, which is more independent of the machine, more abstract, and more stable than the data model.
Conceptual structure design is the key to the whole database design
The characteristics of conceptual structure design
(1) Real and full reflection of the real world
(2) Easy to understand
(3) Easy to change
(4) Easy to switch to relational, network, level and other data models
Tools to describe the conceptual model
E-r model
The conceptual model is independent of the specific DBMS

Methods and steps of conceptual structure design

Four types of methods for designing conceptual structures
From top to bottom
First define the framework of the global conceptual structure, and then refine

From bottom up
First define the conceptual structure of each local application, and then integrate them to get a global conceptual structure

Gradual expansion
First, define the most important core conceptual structure, and then scale out to create other conceptual structures in a snowball fashion until the overall conceptual structure

Hybrid strategy
By combining top-down and bottom-up, a framework of global conceptual structure is designed with top-down strategy, which integrates the local conceptual structures designed by the bottom-up strategy in the skeleton.

Data abstraction

Abstraction is the common characteristic of taking care of the actual people, objects, things, and concepts, ignoring the non-essential details, and describing these characteristics precisely with various concepts.
Conceptual structure is an abstraction of the real world
Three common abstractions
1. Classification (classification)
Define a class of concepts as a type of object in the real world
Abstract the semantics of "is member of" between object values and types
2. Aggregation (Aggregation)
Define a component of a type
Abstract the semantics of "is part of" between the object's inner type and the composition
3. Summary (generalization)
Define a subset of types to contact
Abstract the semantics of "is subset of" between types
Inheritance of

Partial View Design

Steps to design a e-r diagram:
⒈ Select Local Application
⒉ Design sub-e-r diagram

Choose an appropriate level of flow chart in the multi-layer flow chart, as the starting point of the design of e-r graph
Based on the middle-level data flow chart as the design e-r graph
Task
Extract data from the data dictionary for each local application
According to the flow chart, the attributes of entities and entities in each local application are calibrated, and the codes for identifying entities
Determine the relationships between entities and their types (1:1,1:N,M:N)
Two guidelines:
(1) A property can no longer have a property that needs to be described. That is, the attribute must be an irreducible data item and no longer be made up of another property
(2) A property cannot be associated with another entity. Contact only occurs between entities
[Example] design of sub-e-r Chart of sales management subsystem

The main functions of the sales management subsystem are:
Handle orders sent by customers and sales staff
The factory is produced according to the order arrangement.
Handing over the goods and issuing the invoice
Receivables are processed according to invoice stubs and credit conditions upon receipt of customer payment

Types of conflicts

attribute Conflicts
Two types of attribute conflicts

Attribute domain conflict
Type of property value
Range of values
Different collection of values

Property Value Unit Conflict

Naming Conflicts
Two types of naming conflicts

The same name: objects of different meanings have the same names in different local applications

Synonyms (more than one name): objects of the same meaning have different names in different local applications

Structural Conflicts
Three types of structural conflicts
The same object has different abstractions in different applications
The same entity has not exactly the same number of attributes and attributes in different e-r charts
Relationships between entities render different types in different detail views

Redundancy

Redundant data refers to data that can be exported from basic data
A redundant connection refers to a connection that can be exported by another contact

Redundant data and redundant connections are prone to compromise database integrity and make database maintenance more difficult
The initial e-r diagram after eliminating unnecessary redundancy is called the basic e-r diagram.

Ways to eliminate redundancy

Analysis method
Based on data dictionary and Data flow graph
Based on the logical relationship between data items in a data dictionary
Efficiency vs Redundancy information
Needs to be determined according to the overall needs of the user
If some redundant data is artificially retained, the description of the data association in the data dictionary should be used as the integrity constraint
Q4=∑q5
Once the Q5 has been modified, the integrity check should be triggered and the Q4 modified
Normalization theory
The concept of function dependency provides a formal tool for eliminating redundant connections
The integration process addresses the following issues:
Synonyms are synonymous, items and products have the same meaning
The working relationship between staff and warehouse in inventory management has been included in the contact between Labor and personnel management departments and employees, so it can be canceled
The relationship between the leader and the leader can be derived from the leadership relationship between the department and the employee (manager), the subordinate relationship between the department and the employee, so it can also be canceled.

Validate the overall conceptual structure

Once the view is integrated into a holistic database conceptual structure, further validation of the overall conceptual structure is required to ensure that it meets the following conditions:
There must be consistency within the conceptual structure of the whole, there is no contradictory expression
The overall conceptual structure can accurately reflect the original structure of each view, including the relationship between attributes, entities and entities
Overall conceptual structure to meet all requirements identified during the analysis phase

The overall concept structure should also be submitted to the user, solicit the views of users and relevant personnel, review, revise and optimize, and then identify it as the conceptual structure of the database as the basis for further design of the database.
The steps of conceptual structure design
Abstract data and design a partial view
Integrated local view to get a global conceptual structure
Validate the overall conceptual structure
Design a partial view
⒈ Select Local Application
2. Design sub-e-r diagram individually
Calibration of entities, attributes, codes, and relationships between entities in local applications
Describe it with a e-r diagram.
Integrated Detail View
1. Combine the E-r diagram to generate a preliminary e-r diagram
Conflict elimination
Attribute conflicts
Naming conflicts
Structural conflicts
2. Modification and refactoring
Eliminate unnecessary redundancy, design and generate basic E-r diagram
Analysis method
Normalization theory

Database-Conceptual Structure design