Show IH: Shut Your Mouth API - An API to End Cyber Bullying

Looks good. Tips:

1. You might need to change your name. "Shut Your Mouth" is an irrational order (usually out of anger), and sometimes considered a form of bullying. I get where you're coming from, but ... you can be more professional.
2. On the homepage, top right corner the code example, remove those red buttons (that look like STOP signs) that collapse the code. It's confusing and a code example shouldn't have code collapsing on it, it doesn't add any benefit.
3. Look if you can add a "negativity/profanity score" to your API. Your current return values are interesting, but if you'd have a score as well it'd make it so much easier for me to implement into my app.
4. On top of #3: If it contains any kind of extreme bullying (e.g. racial slur, "kill yourself"), return a separate flag that tells me. Extreme cases must be handled with higher urgency and I don't want to maintain my own word list.
5. When looking at your pricing, I thought you had a free tier w/ 100k requests, but only afterward looked at the title (NGO). Make it much clearer that this is for NGOs only, it shouldn't look like a regular pricing tier.
6. Your number "Around 4400 teenagers commit suicide yearly" is wrong. In the US alone it's already higher, and you should show the worldwide number because this is a global problem (and you'd be telling EU customers they're not as important as the US if you don't include them).
7. Find and include real examples of how negativity/bullying detection has helped other companies. I'm familiar with gaming so I can recommend League of Legends (and recently Tom Clancy's Rainbow Six Siege), but you should include non-gaming examples. Use real numbers & link to credible news sources.

It seems like you're just testing the string for some words out of an array. This tool needs NLP if you want it to be useful in any way. So right now, it's too simplistic. Here are some errors I found:

• If I put in 'hello', it returns as offensive because it thinks I'm using 'hell' as a word.
• If someone says 'I suck at math' it will also return as offensive because 'suck' is found.
• 'Hitler is awesome' will return 'neutral'.
• Your own brand name is offensive yet if I enter 'Shut your mouth', it comes back as neutral.
• Entering 'motherf*cker' does not come back as offensive.

I applaud startups in this space, but the tool needs a lot more work imo.

Also, please change the name of your API as it's quite ironic how that name is offensive when used as a normal sentence. Nobody will take you seriously.

Wow this is awesome .this is what im looking for .Great Work..Thank you..

This comment was deleted 7 years ago

This comment was deleted 7 years ago