Website publishers, search engines, social media service providers, and other online entities need to comply with regulation that requires removal of content that is deemed to be insulting to an individual. This disclosure describes techniques to match and remove text that is similar to insult strings that mandated for removal. One-way hashes of insult strings are used to enable removal of insulting content without exposing the name of the subject of the insult string. Name canonicalization is performed and a semantic fingerprint is obtained using a one-way function.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.