The Profanity Filter and Image Moderator prevents our valued customers from using offensive language and “not-allowed” images to customize products.
1. How it works
The tool works with both texts and/or images. Zwelli.com and our partners make a check on the words/images that our customers add to the product at the time of one of the following actions:
- Add to cart
- Generate PDF
If a not-allowed word or image was used, then the customer receives a visual warning indicating the not-allowed word/image and won’t be able to continue with any of the actions mentioned above until he/she has removed or replaced the word or image not allowed. That way, you won’t be able to buy, share or save any product including words or images that are not allowed to use to customize your items. See the video from our partner below.
The Profanity Filter works based on pre-populated lists of not-allowed words. Actually, is not just a list, it’s a system of detecting profanity. Our service is designed to capture many variations of every kind of profanity you can imagine (and even some that you don’t want to).
The tool is not only capable of recognizing profanities as text, but also of recognizing profanity that is embedded within images.
The Image Moderator is based on Artificial Intelligence to detect violating images in real-time. We block images by category:
- WAD (weapons, alcohol, and medical drugs)
- Hate and offensive (offensive symbols, gestures and flags)
- Celebrities (unless in Public Domain or Creative Commons)
- Images that include texts
- Gender, minors, and faces
The tool works with the following languages:
The chosen language is valid for both the Profanity Filter and the “Text” category for the Image Moderator.