[database] differences among structured data, unstructured data and semi-structured data

Linzhonglu

Structured data can obtain corresponding information through inherent key values , And the data format is fixed , Such as RDBMS data

Semi structured data can obtain corresponding information through flexible key value adjustment , And the format of the data is not fixed , Such as json, The information stored under the same key value may be numeric , It may be textual , It could also be a dictionary or a list

<person>
    <name>A</name>
    <age>13</age>
    <gender>female</gender>
</person>

<person>
    <name>B</name>
    <gender>male</gender>
</person>

Unstructured data cannot obtain corresponding information through key value .

In fact, we can't say that documents are unstructured data in general , It depends on the level of information you want to get , For example, I just want to count the company's financial statements , And take the financial report as a whole as my information unit , So the financial report here is also structured data ; And if I want to get the specific information in the financial report , Such as total income , So here, the financial report is unstructured data .

link ：https://www.zhihu.com/question/50986354/answer/276099917
 

Ai Play badminton and call me :

structured ：SQL

Semi structured ：json

Unstructured ： picture

SQL: Structured query language . Its data is structured . Get the corresponding information through the intrinsic key value

json: Through flexible key value adjustment , Get the information

NOSQL database : It does not require structured data design . So its fault tolerance is very strong , Nor is there too strict a design , the

It is easy to expand and modify in the future .

NOSQL The concept of relation does not exist in the database , If you want to achieve a relationship , for instance 1 Yes 1, One to many , Many to many , You need to program it , Instead of using the database itself .

https://www.zhihu.com/question/50986354/answer/1849823026

Jidao Technology （ Beijing ） Co., LTD.

Structured data
Fixed data model Schema, A data combination of a specific data type , For example, database tables
• Stored in RDBMS perhaps Spreadsheet in
• advantage : Association query and modification are simple
• shortcoming : The table structure is fixed , Adding an attribute change table structure is difficult . The introduction of table Association will bring more application troubles .
• application :CRM,ERP, Flight and train reservation system

Unstructured data
• There is no fixed data structure and type , There is no fixed data model schema
• Stored in file storage system or object storage system
• advantage : Follow one's inclinations , All kinds of
• shortcoming : There is no standard , Difficult to standardize management , retrieval , Inquire about
• application :BLOB, Audio and video files , Office documents , report form , journal

Semi-structured data

• There is a format but no fixed data model Schema, Have self describing attribute information to express data content .

• Store in key value pairs ,JSON/XML Similar documentation , Stored in a file or document shaped database , Or stored in the graph database as a graph , In the data warehouse .

• advantage : Good scalability , Any data can be added 、 Delete 、 Change description information , It can be retrieved and analyzed through specific algorithms and methods .

• shortcoming : Specific usage for specific application scenarios , Not good at storing BLOB


link ：https://www.zhihu.com/question/50986354/answer/2214580874
 

Brother Piao :

summary ：

• structured 、 Semi structured 、 Unstructured is actually classified according to data format .
• Structured data belongs to unstructured data , Is a special case of unstructured data .
• Both structured and semi-structured data are data with basic fixed structure patterns
• There is only domain overlap between semi-structured and unstructured data and the current popular big data , In essence, there is no necessary connection between the two .
• The industry regards big data as semi-structured / Unstructured data , It is because big data technology initially played a role in the field of semi-structured data , Its essence is to confuse data processing technology with data format , It's not true .

link ：https://www.zhihu.com/question/50986354/answer/683873405