Access the API
You can access tour deployed runtime API at the runtime URL. The API is a RESTfull API that accepts POST requests. The API is secured with an API key that you will need to include in the header of your request.
curl -X POST https://api.ecosystem.ai/invocations \
-H "Content-Type: application/json" \
-d '{
"campaign": "spending_personality",
"subcampaign": "openai",
"channel": "chatgpt",
"customer": "1234",
"numberoffers": "1",
"userid": "openai",
"params": "{}"
}'Here is another example payload with a in_param set that can later be used in the model.
{
"campaign": "spending_personality",
"subcampaign": "openai",
"channel": "chatgpt",
"customer": "1234",
"numberoffers": "1",
"userid": "openai",
"params": "{}",
"in_param": {
"airtime_balance": 200,
"airtime_advance_limit": 30,
"api_payment_method": "p"
}
}Modules have different endpoints and payloads. The following is an example of the Spend Personality API in OpenAPI 3.0 standard.
This is the standard return for the Spend Personality API:
openapi: 3.0.0
info:
title: Spend Personality API
description: API that obtains spending personality insights and pushes responses once customer values are returned.
version: "1.0.0"
servers:
- url: https://bankruntime2.ecosystem.ai
paths:
/invocations:
post:
summary: Obtain Spend personality
operationId: getSpendPersonality
requestBody:
required: true
content:
application/json:
schema:
type: object
properties:
campaign:
type: string
description: Campaign identifier
default: spend_personality
subcampaign:
type: string
description: Subcampaign identifier
default: spend_personality
channel:
type: string
description: Customer interaction channel
default: copilot
customer:
type: string
description: Customer identifier
default: 590
numberoffers:
type: string
description: Number of offers to display
default: "1"
userid:
type: string
description: User ID
default: copilot
params:
type: string
description: Additional parameters in JSON format
default: "{}"
responses:
"200":
description: Success response
content:
application/json:
schema:
type: object
properties:
cache:
type: integer
description: Indicates if the response is cached (0 for no, 1 for yes)
example: 0
datetime:
type: string
format: date-time
description: Timestamp of the response
example: "2024-12-20T13:04:43.000774Z"
explore:
type: integer
description: Exploration score (default 0)
example: 0
final_result:
type: array
description: Array of final results for spend personality analysis
items:
type: object
properties:
result:
type: object
properties:
offer:
type: string
description: The recommended offer
example: "Industrious"
score:
type: number
description: Personality trait score associated with the offer
example: 0.4742771310552009
final_score:
type: number
description: Final calculated score
example: 0.4742771310552009
cost:
type: integer
description: Cost associated with the recommendation
example: 1
price:
type: integer
description: Price of the offer
example: 1
offer_value:
type: integer
description: Value of the offer
example: 1
uuid:
type: string
description: Unique ID for the recommendation
example: "fdfbe8f4-7f99-4315-9614-ca68e6c50b4b"
arm_reward:
type: number
description: Reward score associated with the offer
example: 0.4742771310552009
modified_offer_score:
type: number
description: Adjusted score for the offer
example: 0.5204656402553511
offer_name:
type: string
description: Name of the offer
example: "Industrious"
result_full:
type: object
properties:
cost:
type: number
description: Cost metric
example: 1
industrious:
type: number
description: Industrious personality score
example: 0.4742771310552009
offer_name:
type: string
description: Name of the offer
example: "Industrious"
personality:
type: string
description: Detected personality type
example: "Industrious"
score:
type: number
description: Overall personality score
example: 0.4742771310552009
trait:
type: string
description: Prominent personality trait detected
example: "Extrovert"
rank:
type: integer
description: Rank or priority of this result
example: 1
id:
type: string
description: Unique identifier for the analysis type
example: spend_personality
uuid:
type: string
description: UUID for the overall response
example: "fdfbe8f4-7f99-4315-9614-ca68e6c50b4b"
in_params:
type: object
description: Input parameters for the request
example: {}
/response:
post:
summary: Respond to Spend personality when customer lookup happens
operationId: customerResponse
requestBody:
required: true
content:
application/json:
schema:
type: object
properties:
uuid:
type: string
description: Unique identifier for the request
example: "fdfbe8f4-7f99-4315-9614-ca68e6c50b4b"
offers_accepted:
type: array
description: List of offers accepted by the customer
items:
type: string
example: ["offer1"]
channel:
type: string
description: Customer channel type
default: chatgpt
responses:
"200":
description: Response confirmation
content:
application/json:
schema:
type: object
properties:
final_result:
type: array
description: The final response result from the API
items:
type: object
components:
schemas:
SpendPersonalityRequest:
type: object
properties:
campaign:
type: string
description: Campaign identifier
default: spend_personality
subcampaign:
type: string
description: Subcampaign identifier
default: spend_personality
channel:
type: string
description: Customer interaction channel
default: copilot
customer:
type: string
description: Customer identifier
default: 590
numberoffers:
type: string
description: Number of offers to display
default: "1"
userid:
type: string
description: User ID
default: copilot
params:
type: string
description: Additional parameters in JSON format
default: "{}"
SpendPersonalityResponse:
type: object
properties:
cache:
type: integer
description: Indicates if the response is cached (0 for no, 1 for yes)
example: 0
datetime:
type: string
description: Timestamp of the response
example: "2024-12-20T13:04:43.000774Z"If you do not want the API to return an array or document in document JSON objects then use the flatten option. This is especially useful for UI and other technologies that have difficulty with processing nested JSON objects. The example uses the in_param to force a flattened response: {"in_param": {"flatten_json": true}}
openapi: 3.0.0
info:
title: Spend Personality API
description: API that obtains spending personality insights and pushes responses once customer values are returned.
version: "1.0.0"
servers:
- url: https://bankruntime2.ecosystem.ai
paths:
/invocations:
post:
summary: Obtain Spend personality
operationId: getSpendPersonality
requestBody:
required: true
content:
application/json:
schema:
type: object
properties:
campaign:
type: string
description: Campaign identifier
default: spend_personality
subcampaign:
type: string
description: Subcampaign identifier
default: spend_personality
channel:
type: string
description: Customer interaction channel
default: copilot
customer:
type: string
description: Customer identifier
default: 590
numberoffers:
type: string
description: Number of offers to display
default: "1"
userid:
type: string
description: User ID
default: copilot
params:
type: string
description: Additional parameters in JSON format
default: "{}"
responses:
"200":
description: Success response
content:
application/json:
schema:
type: object
properties:
final_result_0_result_full_industrious:
type: number
description: Industrious personality score
example: 0.4742771310552009
final_result_0_result_full_transaction_count:
type: integer
description: Total transaction count
example: 2663
final_result_0_result_score:
type: number
description: Personality trait score associated with the offer
example: 0.4742771310552009
final_result_0_result_offer_name:
type: string
description: Name of the offer
example: "Industrious"
final_result_0_result_full_uuid:
type: string
description: Unique ID associated with the personality result
example: "e98aeb34-786d-453f-8798-a03145a34e2f"
uuid:
type: string
description: Primary UUID of the response
example: "e98aeb34-786d-453f-8798-a03145a34e2f"
final_result_0_result_full_offer:
type: string
description: The recommended offer
example: "Industrious"
final_result_0_result_cost:
type: integer
description: Cost associated with the recommendation
example: 1
datetime:
type: string
format: date-time
description: Timestamp of the response
example: "2024-12-20T18:08:52.000565Z"
final_result_0_result_full_modified_offer_score:
type: number
description: Adjusted score for the offer
example: 0.5204656402553511
final_result_0_result_full_personality:
type: string
description: Detected personality type
example: "Industrious"
final_result_0_result_full_extrovert:
type: number
description: Extrovert score for the customer
example: 0.5204656402553511
final_result_0_result_full_offer_details_1:
type: integer
description: Offer detail, key `1`
example: 1
final_result_0_result_full_offer_details_0:
type: integer
description: Offer detail, key `0`
example: 0
final_result_0_result_full_trait:
type: string
description: Prominent personality trait detected
example: "Extrovert"
final_result_0_result_full_offer_id:
type: string
description: Identifier for the offer
example: "Industrious"
final_result_0_result_full_price:
type: integer
description: Price of the offer
example: 1
id:
type: string
description: Unique identifier for the analysis type
example: spend_personality
final_result_0_result_final_score:
type: number
description: Final calculated score
example: 0.4742771310552009
final_result_0_result_full_offer_value:
type: integer
description: Value of the offer
example: 1
final_result_0_result_full_intentional:
type: number
description: Intentional score for the customer
example: 0.09500563274502441
final_result_0_result_price:
type: integer
description: Price of the offer
example: 1
in_params_flatten_json:
type: boolean
description: Indicates if the input parameters were flattened
example: true
cache:
type: integer
description: Indicates if the response is cached (0 for no, 1 for yes)
example: 0
explore:
type: integer
description: Exploration score (default 0)
example: 0
final_result_0_result_full_offer_name:
type: string
description: Name of the offer
example: "Industrious"
final_result_0_result_uuid:
type: string
description: Unique ID for the recommendation
example: "e98aeb34-786d-453f-8798-a03145a34e2f"
final_result_0_result_modified_offer_score:
type: number
description: Adjusted score for the offer
example: 0.5204656402553511
final_result_0_result_full_enthusiastic:
type: number
description: Enthusiastic personality score
example: 0.06759294029290275
final_result_0_result_full_score:
type: number
description: Overall personality score
example: 0.4742771310552009
final_result_0_rank:
type: integer
description: Rank or priority of this result
example: 1
final_result_0_result_full_introvert:
type: number
description: Introvert score for the customer
example: 0.2377018400300413
final_result_0_result_full_experiential:
type: number
description: Experiential personality score
example: 0.12129177619226436
final_result_0_result_full_cost:
type: integer
description: Cost metric
example: 1
final_result_0_result_offer:
type: string
description: The recommended offer
example: "Industrious"
final_result_0_result_offer_value:
type: integer
description: Value of the offer
example: 1
final_result_0_result_arm_reward:
type: number
description: Reward score associated with the offer
example: 0.4742771310552009URL Encoding is required for the in_param field.
{
"campaign": "justforyou_nb",
"subcampaign": "justforyou_nb",
"channel": "app",
"customer": "1846573d-ac20-45df-a9f0-2fb14da6e7d1",
"userid": "ecosystem",
"numberoffers": "3",
"params": "%22%7B%5C%22airtime_balance%5C%22:%20200,%20%5C%22airtime_advance_limit%5C%22:%2030,%20%5C%22api_payment_method%5C%22:%20%5C%22p%5C%22%7D%22"
}Use the /response API to send the selected uuid as return for the model to convergen in real-time if it’s configured to do so.