Hacker News new | past | comments | ask | show | jobs | submit login
Sherloq: A tool to stop hate speech (sherloq.io)
19 points by marcosnils on March 17, 2017 | hide | past | favorite | 8 comments

An awesome tool you have created... I typed in a few things that I [normally would never] said about "groups of people" and it came up with almost 100% hatefulness.

"I hate people." 0%

"I hate you." 92%

"I hate the world." 0%

"I hate my job." 0%

"I dislike John Smith." 5%

"I hate John Smith." 0%

"I hate you, John Smith." 92%

"I hate my life." 1%

"I hate my wife." 0%

"You suck." 92%

"Everyone sucks." 16%

"John Smith sucks." 70%

"If I hate everyone, than does that mean I hate myself?" 0%

"I don't like olives." 0%

"I don't like Antarctica." 0%

I see the system is also analyzing swear words and adds a few % points.

It does take into account words like "Jews" (99%) and "Muslims" (95%) and "Catholics" (12%) [why do certain groups get a bigger percentage rating than others?].

But it also fails on other groups of people like "Christians", "Baptists", "Atheists", "Methodists", "Lutherans", "Indians", "Canadians", etc.

So.. there is always room for improvement!

The UI is gorgeous, by the way, and mobile friendly!

I'm sure the system can (and should) learn from user input as people enter more phrases and words into the system? Can you add the capability of adding custom phrases or words to the admin? I didn't see this feature, only the ability to add "criteria" but it doesn't let me actually type anything in.

It would also be nice to see something that allows for "levels" of toleration as well. Something like "Relaxed" "Moderate" or "Strict".

This is a tool I've bookmarked and will be looking into using as I develop more web apps. But I'd like to see a way to add my own phrases and words that I would personally deem offensive.

Thank you for sharing!

Hello mattbgates, I'm one of the engineers working on the project. Thanks a lot for taking the time to play around with the tool and writing this feedback. We're going to add a feature for giving feedback to moderation (whether the moderated text is actually hate speech or not) to learn from users input.

We're actively working on improving the model so hopefully next time you try it you'll see the improvements!

This writing experiment/demo is pretty fun.

  All immigrants should be deported.
  All boy bands should be executed.
Both got me a 0%.

  All Germans are evil rats.

All my (self)hate wasn't sufficient to trigger this algorithm.

Hello fwn, interesting examples you have there. We're still working on further improvements to the model at this stage, aiming to detect even more sophisticated examples. Thanks for your feedback and hope you try it again and see improvements soon.

Given this example from the page `diversity is white genocide forced on us by the libtard elite` I decided to test for liberal bias.

"Fuck liberals." 97%

"Fuck conservatives." 0%


I copied the text from the following and pasted it into the engine via your web site:


I then deleted sections randomly (mostly from the tail) to see what effect this would have on the rating. The rankings seem to shift radically with what would seem like insignificant edits to a body of text. The following text ranks as 92% "hatefulness". Then, removing only the text "HomeMy Books Browse Community Sign" (which I would think are not 'hateful' in any way) the ranking of "hatefulness" drops to 39%.

I realize the technical challenges that must be involved so I'm not really surprised at the behavior. But I will say that I question the merits of attempting to detect "hate speech" in the first place. Assume that this tool were to eventually reach a state where it could detect "hate" with absolute precision. Then what? What purposes shall it be used for? I can not imagine a single purpose that I personally would endorse. The notion of subjectively ranking the free speech of any other individual for any purposes whatsoever is repugnant to me.

[* BEGIN SAMPLE ] HomeMy Books Browse Community Sign

Get quotes daily Join Goodreads Quotes About Freedom Of Speech Quotes tagged as "freedom-of-speech" (showing 1-30 of 250) “I disapprove of what you say, but I will defend to the death your right to say it.” ― S.G. Tallentyre, The Friends of Voltaire tags: disagreement, evelyn-beatrice-hall, freedom-of-speech, freedom-of-thought, misattributed-to-voltaire

Oscar Wilde “I may not agree with you, but I will defend to the death your right to make an ass of yourself.” ― Oscar Wilde tags: freedom-of-speech, paraphrased, self-expression 1124 likes Like

Henry Louis Gates Jr. “Censorship is to art as lynching is to justice.” ― Henry Louis Gates Jr. tags: african-american, american, censorship, freedom-of-speech, freedom-of-thought, freedom-to-read, literary 1067 likes Like

Theodore Roosevelt “To announce that there must be no criticism of the President, or that we are to stand by the President, right or wrong, is not only unpatriotic and servile, but is morally treasonable to the American public.” ― Theodore Roosevelt tags: dissent, freedom-of-speech, truth-telling 644 likes Like

George Washington “If freedom of speech is taken away, then dumb and silent we may be led, like sheep to the slaughter.” ― George Washington

Harry Truman “Once a government is committed to the principle of silencing the voice of opposition, it has only one way to go, and that is down the path of increasingly repressive measures, until it becomes a source of terror to all its citizens and creates a country where everyone lives in fear."

[Special Message to the Congress on the Internal Security of the United States, August 8, 1950]” ― Harry Truman

Christopher Hitchens “My own opinion is enough for me, and I claim the right to have it defended against any consensus, any majority, anywhere, any place, any time. And anyone who disagrees with this can pick a number, get in line, and kiss my ass.” ― Christopher Hitchens

Jess C. Scott “Hypocrites get offended by the truth.” ― Jess C. Scott, Bad Romance: Seven Deadly Sins Anthology [ END SAMPLE *]

there are several use cases for a product like this. Lots of social media companies like twitter, facebook, reddit / etc have huge problems because of hate speech. In addition, as sirfz mentioned, child abuse and harassment can also be applied in this case.

Online gaming is another area where hate speech is big problem and a tool like this would encourage participation and possibly detect toxic players that don't help to grow communities.

Here are some articles related to the problem:





Thanks for the feedback mamaniscalco, please check back again soon to see if our detection has improved (if you're interested).

As for the the purpose of the tool, one application could be protecting children from online hate/harassment for example.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact