Science Google Developing Troll-Detection Software

http://www.theverge.com/2017/2/23/1...ctive-software-ai-machine-learning-developers

Google’s Jigsaw unit, as part of a larger effort to battle online trolling, said earlier today that it was releasing a new tool called Perspective, software that uses machine learning to detect harassment and abuse online.

While Jigsaw and its software are not new — it’s been around in some form since 2010, though with a different name, and Jigsaw software has even been used to spot potential ISIS recruits — this new Perspective API makes it available to developers who want to use and build on top of the software. The software works by determining the “toxicity” of online comments, a scale that has been established by mining millions of comments from the web and then presenting them to panels of 10 people (humans!) at a clip to get their feedback.

Try using Perspective to determine the toxicity of your average post

http://www.perspectiveapi.com/

@Null should embed this to grade our posts based on how toxic they are

Spoiler

 

[Xarpho]

This is my "high score", in only three words. (I tried "kill yourself", got 94%)

 

[Sailor_Jupiter]

@Dynastia and @AnOminous both scored 34%. So did I. @Mariposa Electrique is 30% and rated less toxic... With which I agree.

@Cosmos scored 2% toxic, which is less toxic than "fluffy kittens" (which scored a whopping 9%.) Lol. Lowest toxicity score thus far. She barely edged out Thomas Jefferson, who shared a 3% score with his epic rival/frenemy John Adams. (Lol I love it that they tied. Still competing! XD )

Ed: Apparently Batman and Superman are 2%, as well. Wow. So hero. Much wholesome. Wow.

 

[Ruin]

@Dynastia and @AnOminous both scored 34%. So did I.

@Cosmos scored 2% toxic, which is less toxic than "fluffy kittens" (which scored a whopping 9%.) Lol. Lowest toxicity score thus far. She barely edged out Thomas Jefferson, who shared a 3% score with his epic rival/frenemy John Adams. (Lol I love it that they tied. Still competing! XD )

"Ruin" is apparently 53% toxic.

 

[Sailor_Jupiter]

"Ruin" is apparently 53% toxic.

Well, you've got us all beat! What do you plan to do with your new title?

 

[Ruin]

Well, you've got us all beat! What do you plan to do with your new title?

Shit self and do nothing duh.

 

[nonnewtonian]

 

[Sperglord Dante]

 

[Cosmos]

@Cosmos scored 2% toxic, which is less toxic than "fluffy kittens" (which scored a whopping 9%.) Lol. Lowest toxicity score thus far. She barely edged out Thomas Jefferson, who shared a 3% score with his epic rival/frenemy John Adams.

Learning that I'm only 2% toxic made my day, thank you

 

[AA 102]

lmao

 

[Alberto Balsalm]

 

[IwegalBadnik]

I would pretend to be surprised, but I'm not.

EDIT: LMFAO

Starting to see a trend develop here...

Spoiler

 

[GS 281]

Anything kiwi farms related is 34% toxic, even kiwifarms

 

[Batman VS Tony Danza]

This thing is terrible.

Spoiler: Rollercoaster of Toxicity

 

[Coconut Gun]

Black people are 21% more toxic than white people.

 

[Xarpho]

More on weird trends, I ran some retail names into it, Walmart, Target, Amazon.com, and Sears got less than 10%, Macy's got 11%, but a motley assortment hung out at 34% (JCPenney, Wegmans, Kmart, Dillard's, Albertsons).

 

[Varg Did Nothing Wrong]

This is soooo problematic. When trans people say things like "die cis scum" it's obviously just them letting off steam from a life persecuted by the cisnormative patriarchy

 

[WireSponge]

Spoiler: Many Images

Spaces are toxic.

U TOXIC BOI

Random letters are apparently quite toxic, but never more than 34 percent. Exhibit B:

Spoiler: Racism

Spoiler: Islamophobia: Round 2

Plot twist it actually was kind if islamophobic this time

 

[Hatoful Dandy]

Nice to see they agree with me

Reported as non toxic

XD

 

[Un Platano]

If you live in California this looks like a good opportunity to sue over labelling

 

[Lazuli]

Turns out the company that produced this is called Jigsaw, formerly.... Google Ideas.