Text moderation and filtering API
Filtering and moderation of toxic texts
ZPK Moderator is an API that allows you to detect harmful or dangerous texts using artificial intelligence in your application or in user communities.
Avoid or control the publication of texts that incite hatred, self-harm, or content inappropriate for minors.
"hate": {
"score": "0.000185327", "danger": false
},
"self_harm": {
"score": "0.000000909", "danger": false
},
"sexual": {
"score": "0.348992828", "danger": false
},
"sexual_minors": {
"score": "0.000000048", "danger": false
},
"violence": {
"score": "0.767623782", "danger": true
}
I like this video game, but the enemies keep killing me. I'm going to take my AK-47 and kill them all LOL
I like this school, but the classmates are disgusting, I'm going to get a**47 and **** them all.
Context-sensitive AI moderation
Our moderation API is not just word filtering, we use artificial intelligence to detect the context of messages, minimizing false positives, and detecting positives that would normally be missed.
The moderation API will detect context in messages, and you can optionally specify context when requesting analysis, for example to contextualize a comment within a particular news story.
Hey guys, do you want me to set up a Counter-Strike game? Let's shoot some rounds?
Transparent prices
No monthly fees or minimum usage.
ZPK Systems uses a transparent pricing model, there are no fees, no usage minimums, and no volume discounts because we offer the best price from the first request to our APIs.
Some providers force their users to sign up for 'packs' of thousands of requests that a project that starts will probably not use, at ZPK Systems payments are 100% pay-per-use.
(up to 500 characters)
You will pay for the exact usage of the API, not a penny more.
Mónica
| Sexual: | |
| Hate: | |
| Violence: |
Darth Maul
| Sexual: | |
| Hate: | |
| Violence: |
Satirus Vicius
| Sexual: | |
| Hate: | |
| Violence: |
Track harmful users or users who maintain inappropriate behavior.
With this moderation API, you can map the origin of messages to an identifier, this will allow your application to extract data about the attitude of a particular messages source.
Whether it is a user who maintains an overly aggressive attitude, or auto-imported data from a social network source who is inciting hatred.
If you set a source on messages, you can extract attitude statistics from that specific source.
Easy integration in a multitude of languages
Even though we have an SDK specifically for PHP, the moderation API works through HTTPs/JSON requests.
Allowing easy and quick integration into a multitude of programming languages, any language that can make HTTPs requests can use our text moderation API.
composer require zpksystems/phpzpk
Moderation SDK available for PHP
If your project is developed in PHP, we offer an SDK that will allow you to filter content and detect harmful comments or text in your application.
Our SDK allows you to access the moderation API and any of the ZPK APIs at no extra cost, and without maintenance fees. Access this and 6 more APIs, whenever your application requires it.
Ready to get started?
Try our moderation API without costs.
Sign up as a ZPK Systems user and we'll add credit to your account for you to run integration tests.