An entity is a collection of similar words, such as cities, dates, or colors. The system uses entities to understand semantics. It extracts them and outputs them in a machine-readable format. For example, a user might say, "I like green." A machine does not know that green is a color unless you tell it. To teach the machine, you can create a dictionary of common colors. Then, the machine can recognize these words as colors.
Create a standard entity
In daily life and business, different words can mean the same thing. For example, "Beijing," "Imperial Capital," and "Capital of China" all refer to the same city. You can configure these words to belong to a single entity.
You can create an entity for train seat classes and define different seat levels within it. You can then use this entity directly in an intent.
Create an entity from an intent
You can also use a shortcut to create an entity from an intent.
System entities
Dialog Studio provides many pre-built system entities to help you get started quickly.
Interface name |
Meaning |
Description |
Example |
sys.number |
System.Number |
Parses all numbers in a sentence, including Chinese and Arabic numerals. |
one hundred twenty-five, 100, zero point three |
sys.phone-number |
System.Phone Number |
Parses mobile phone numbers. |
1*78******* |
sys.date |
System.Date |
Parses all date expressions. |
tomorrow, next Monday, March 20th, June next year, National Day, 2017-10-5 |
sys.time |
System.Time |
Parses times of the day. |
nine in the morning, 12:30 PM, a quarter past eight in the evening |
sys.date-time |
System.Date and Time |
Parses dates and times together. |
tomorrow at 12:30 PM, next Wednesday at a quarter past eight in the morning |
sys.geo-city |
System.City |
Parses all cities in China. |
Beijing, Hangzhou, Shanghai |
sys.geo-province |
System.Province |
Parses all provinces in China. |
Zhejiang, Hebei |
sys.geo-country |
System.Country |
Parses all countries in the world. |
China, Singapore, Mauritius |
sys.age |
System.Age |
Parses age numbers. |
15 years old, 30, fifty years old |
sys.address |
System.Address |
Parses complex, multi-level addresses. |
No. 882, Julu Road, Jing'an District, Shanghai, China |
sys.email |
System.Email |
Parses standard mailboxes. |
de***@163.com, 12***@qq.com |
sys.duration |
System.Duration |
Parses a time period or duration. |
one hour, 15 minutes, one hour and five minutes |
sys.timeset |
System.Recurring Time |
Parses a recurring time point entity. |
every Monday, every workday |
sys.ordinal |
System.Ordinal |
Parses ordinal numbers. |
first, second |
sys.express |
System.Express Delivery |
Parses express delivery services. |
SF Express, ZTO Express |
sys.receipt_object |
System.Invoice Object |
Parses invoice objects in the E-commerce realm, such as individuals. |
individual, enterprise |
sys.receipt_type |
System.Invoice Type |
Parses invoice types in the E-commerce realm, such as electronic invoicing. |
electronic invoicing, paper invoice |
sys.receipt_form |
System.Invoice Form |
Parses invoice forms in the E-commerce realm, such as value-added tax. |
value-added tax |
Create a regular expression entity
A regular expression is a sequence of characters that specifies a search pattern. This pattern consists of ordinary characters, such as 'a' through 'z', and special characters, called metacharacters. You can use regular expressions to match patterns in strings.
Regular expression rules
