---
title: "Discussion on \"Detect Toxic Comments in Your App Using Toxicity Space\" | Hashnode"
description: "Discussion on \"Detect Toxic Comments in Your App Using Toxicity Space\". How Toxicity Space can be used to enhance the toxic commentary detection in your application"
url: "https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Detect Toxic Comments in Your App Using Toxicity Space](https://note.hadna.space/detect-toxic-comments-in-your-app-using-toxicity-space?utm_source=hashnode&utm_medium=feed)

How Toxicity Space can be used to enhance the toxic commentary detection in your application

Jun 24, 2022 · 11 min read · note.hadna.space  
Updated Oct 22, 2023  
Read the article

-   [#linode](https://hashnode.com/tag/linode)
-   [#linodehackathon](https://hashnode.com/tag/linodehackathon)
-   [#tensorflow](https://hashnode.com/tag/tensorflow)
-   [#artificial-intelligence](https://hashnode.com/tag/artificial-intelligence)
-   [#text-processing](https://hashnode.com/tag/text-processing)

## Written by

[Diky Hadna 👾](https://hashnode.com/@dkhd)

@dkhd · ~/.zshrc

  
[View profile](https://hashnode.com/@dkhd)

Read my profile at https://hadna.space

## Responses 3

Markdown supported · ⌘↵ to post

[Eleftheria Batsou](https://hashnode.com/@eleftheriabatsou)

[Jul 4, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62c2da0cccecfb25bbc787fd)

Hi [Diky Hadna 👾](https://hashnode.com/@dkhd), could you please update the GitHub repo with an MIT license so this project is eligible for the hackathon?

Thank you!

[Diky Hadna 👾](https://hashnode.com/@dkhd)

[Jul 4, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62c2da0cccecfb25bbc787fd#reply-62c30861c0401b3dcbcfdd91)

Hi [Eleftheria Batsou](https://hashnode.com/@eleftheriabatsou),

It's done! The Github repo now has an MIT license.

[Omar Moustafa](https://hashnode.com/@Omar-The-Dev)

[Jun 25, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62b67cebb86f939ac81b481c)

Pretty cool and nice idea! The idea is cool but it would be more better if there is an **API** to automatically integrate with your apps :)

Yeah found the API, but how like I can integrate with an Android, etc. application?

[Diky Hadna 👾](https://hashnode.com/@dkhd)

[Jun 25, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62b67cebb86f939ac81b481c#reply-62b7192ab86f939ac81b5b6a)

[Omar Moustafa](https://hashnode.com/@Omar-The-Dev) I already provided an example of how to utilize this API in the application (using Javascript and `curl`).

Have you tried it?

[Omar Moustafa](https://hashnode.com/@Omar-The-Dev)

[Jun 25, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62b67cebb86f939ac81b481c#reply-62b7267e92447c9aed5b7f13)

[Diky Hadna 👾](https://hashnode.com/@dkhd) Yes, but most people now use **Kotlin for Android :) Don't worry! I will figure it out!**

[Diky Hadna 👾](https://hashnode.com/@dkhd)

[Jun 26, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62b67cebb86f939ac81b481c#reply-62b84966b86f939ac81b748a)

[Omar Moustafa](https://hashnode.com/@Omar-The-Dev) thank's for the suggestion. I've added it to the article so other Devs can provide API call examples in various programming languages.

If you want, you can also write down how to call the API using Kotlin, and I'll add it to the article as well.

[Miles Wallio](https://hashnode.com/@kmwallio)

[Jun 24, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62b5f14a92447c9aed5b5b72)

This is pretty cool. Did you create the dataset (or is it available somewhere)?

The labelling takes it to the next level for auto-moderation. You could create a community based on "We're obscene, but we're not toxic or insulting". A lot of the datasets online are binary on a single level.

[Diky Hadna 👾](https://hashnode.com/@dkhd)

[Jun 25, 2022](https://hashnode.com/posts/detect-toxic-comments-in-your-app-using-toxicity-space/62b5a367b86f939ac81b2eb5/comment/62b5f14a92447c9aed5b5b72#reply-62b668a192447c9aed5b6617)

Hi [Miles Wallio](https://hashnode.com/@kmwallio)

The model was trained on the civil comment dataset which can be accessed here [figshare.com/articles/dataset/data\_json/7376747](https://figshare.com/articles/dataset/data_json/7376747) and it contains about 2M comments labeled for toxicity, specific for this purpose.

Yes, things like this are pretty cool actually 😁 there are a lot of possibilities if we want to use.

In another cases, Chatbots are built on top of Natural Language Processing as well. Take another step to implement the speech recognition and text-to-speech, then we'll have Siri or Alexa or Cortana 😁
