Apply a complex model

<aside> 🚧 Coming soon: the options described in this page will be available in the coming weeks (updated December 12, 2024)

</aside>

Why apply a complex model?

The challenges of a keyword classifier is, in summary, that the classification does not account for context. We can overcome this challenge to a certain extent by applying a classification that uses a large language model (or LLM). LLMs are trained on large amounts of data to learn how language works. They can then use this knowledge to perform a variety of natural language processing (NLP) tasks, including classifying text.

Put simply for our use case: LLMs are much better at classifying complicated things such as broad topics (”politics” or “gender”) or tone (“intimidation” or “negative sentiment” or “toxicity”).

Complex models available to apply in Phoenix

When you click “create” in the Classify tab on the Phoenix platform, you will see options to create an author classifier and a keyword text classifier (these two options are described in the previous sections). You will also see options to apply complex models to classify text (in posts or comments), with the name of the model and a short description. The models listed below are currently available on Phoenix.

<aside> ➡️

We aim to add more classifier models as we grow. Do you have access to a model that you think would help Phoenix? Email [email protected] to let us know and we’ll try to integrate it!

</aside>

Perspective API — coming soon

Perspective API is a free classifier that uses machine learning to identify toxic language. You can use Perspective API to classify based on the text of a post and/or comment. This will be applied to all posts and/or comments that have been gathered, and data gathered in the future.

You can find the different classes of toxicity identified by the model here.
This model works only for some languages, listed here.

Intimidatory language in the Sahel — coming in Q1 2025

In 2024, Build Up and datavaluepeople collaborated with Search for Common Ground and local peacebuilders across the Sahel to train a model that identifies Facebook and TikTok comments aimed at intimidating people. The model classifies a comment by its level of intimidation: low, medium, high or not intimidating.

You can find a description of the methodology used to train this model here.
This model works best in French language, but can also classify for other local languages (coming soon).

Can’t see a model that works for you? It is possible for peacebuilders to train their own models, here is how we think about this process. We work with a community of developers interested in building models relevant to conflict analysis; you can join our Discord here to find collaborators.

</aside>