Use the Spam Detection API
Detect unwanted links and fraud within user messages
Membrace is a Hybrid AI platform with pre-trained AI models to identify and moderate any type of spam content
What type of spam does Membrace block?
While our Spam Moderation tool always takes context into account, here are some common examples of spam content below
Suspicious messages-links with no accompanying text
Calls to follow the link for instant success
Promises of earnings for merely clicking on a link
Money transfer requests
Links to adult content resources
Fraudulent schemes
Casino and gambling links
Сlickbait
What doesn't qualify as spam?
Link serves as the response to the question posed in the initial message and is not spam
Link to an authoritative and well-known service
Invite links that are not related to violations
Solution for digital platforms
Membrace suggests Anti-Spam Solution for:
Messages
Reviews
Comments
Profiles
Contact
Emails
Posts, Publications
Request a free demo
See firsthand how our solution seamlessly integrates with and protects your service
Case Studies
We've worked with the world's leading AI teams for years, and delivered more high-quality data than anybody else.
Use the Spam Detection API
Request
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
import requests
token = 'abacaba' # set real token here
api_url = 'https://api.membrace.ai/v1/item/process'
headers = {'Authorization': token}
# Text request
text = 'Hello world'
resp = requests.post(
api_url,
headers=headers,
json={
'request_id': '1',
'project_id': 'text',
'project_version': '1',
'content': {
'text': text
}
}
)
resp.raise_for_status()
print(resp.json())
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
{
"request_id": "1",
"item_id": "safdhgjkl",
"project_id": <project_id>,
"project_version": "1",
"status": "SUCCESS",
"result": {
"alcohol": false,
"clean_text": true,
"clickbait": false,
"drugs": false,
"erotica": false,
"explicit": false,
"insult": false,
"insult_light": false,
"insult_strong": false,
"obscene_light": false,
"obscene_strong": false,
"personal_data": false,
"swearing": false
}
}
Request
1
2
3
4
5
6
7
8
9
10
11
12
curl -X POST \
-H 'Authorization: <token>' \
-H 'Content-Type: application/JSON' \
-d '{
"request_id": "1",
"project_id": <project_id>,
"project_version": "1",
"content": {
"text": "Example message"
}
}' \
https://api.membrace.ai/v1/item/process
Response
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
{
"request_id": "1",
"item_id": "safdhgjkl",
"project_id": <project_id>,
"project_version": "1",
"status": "SUCCESS",
"result": {
"alcohol": false,
"clean_text": true,
"clickbait": false,
"drugs": false,
"erotica": false,
"explicit": false,
"insult": false,
"insult_light": false,
"insult_strong": false,
"obscene_light": false,
"obscene_strong": false,
"personal_data": false,
"swearing": false
}
}
Access cutting-edge Hybrid AI through our API
Membrace AI
Ready-to-use pretrained models
200ms response time
To boost content validation and generation
Human-in-the-loop
Processes to evaluate and retrain models
Guidelines
Easy to fine-tune
40+
Languages, 100+ Countries
>15M cases per day
Scalable to process any data volume