Support the data validation of hate-speech related Tweets for the pilot Gender AI Social Media Monitoring tool (English & Filipino)

Contract

This is a UNV contract. More about UNV contracts.

A Global Observatory of Public Policies is being set up by UNDP to develop gender responsive policy measures to achieve gender equality. The observatory will carry out collaborative research and data analytics, identify good practices linked to laws and public policies and promote policy dialogue. The Observatory will capture policy measures taken by governments on economic security – labor market, fiscal and economic measures and social security; and gender parity in decision making in public administration. The Observatory will provide guidance for policymakers and evidence for advocates to ensure a gender-sensitive policy response. The Global Observatory will be anchored in the successful work of the UNDP and UNW COVID Global Gender Policy Tracker.

The Gender AI Early Warning System (EWS): Monitoring Hate Speech on Social Media

An AI-based early warning system to monitor hate speech on social media is currently under development as part of the Global Observatory. The Gender AI Early Warning System (pilot version) aims to track hate speech occurrence and patterns on social media, specifically Twitter, with a focus on hate speech targeting women and girls across three pilot countries. With the online proliferation of hate speech, there is an urgent need for systems that can detect harmful content on social media networks for trends and signs of specific gender related events particularly indications of backlashes relating to gender equality and women’s rights.

An artificial intelligence (AI)-based Early Warning System (EWS) has been developed to track and monitor hate speech against women and girls on social media, with a focus on Twitter. We are currently in the first phase of the project, piloting the tool in the Philippines, Colombia and Uganda.

We are looking for volunteers to support the further development and improvement of the AI EWS in the Philippines by doing manual validation and review of the hate-speech related Tweets that the AI system has identified. Tweets to be reviewed will either be in English or Filipino. This work is valuable in improving the system and training the AI to accurately identify hate speech tweets.

Volunteers will be required to attend a training session via Zoom and will be responsible for validating and entering data into an existing database. Volunteers should be available to work on this project at least until October 2022.

The gender team is looking for volunteers fluent in English and Filipino. They will be required to attend weekly meetings for training and coordination purposes. Volunteers must be able to follow detailed data entry and coding guidelines to ensure accurate and transparent data. Familiarity and expertise with gender issues are considered an asset. Familiarity with data science and AI topics are considered an asset.