Natural Language Processing support in French language for Digital Fair Finance Social Listening in Morocco

Contract

This is a UNV contract. More about UNV contracts.

SDG AI Lab is developing a social listening platform, which would provide insights about people's opinion -specifically vulnerable groups- on regulatory changes to allow authorities to shape their decisions. The tool actively collects data from various social platforms, analyzes and visualizes the results. Online volunteers will support the analysis of the collected data to inform authorities about the possible issues on digital finance.

Online volunteers would benefit in various aspects such as team working, GitHub usage and knowledge sharing environment in Natural Language Processing & Data Analysis fields. Online volunteers' effort on the assigment will be directly used for the web tool which will be deployed so that it is possible to mention about your assigment in portfolios/CVs.

The online volunteers will support the ongoing project Suptech for Digital Finance with Social Listening for support authorities to reach out vulnerable groups. The volunteers will work on building a NLP pipeline for French Text Data. The followings describes the tasks:

1. French Text Data Cleaning
2. Sentiment Analysis over French Text Data
3. Topic Modelling over French Text Data
4. Keyword Matching for French Text Data
5. Keyword extraction from French Text Data
6. Existing ML model review

SDG AI Lab will support the volunteer as needed, with regular check-ins and maintaining collaboration through MS Teams, Slack and GitHub group repository. The volunteer will work under the coordination of the Technical Specialist of the IICPSD.

Bachelor’s in computer science/engineering, software engineering or related field with a strong interest in data science and web development for data visualization. Strong knowledge of

• Python: Data Cleaning for French text data, Data Analysis for French text data
• Git/GitHub

Experience on building French NLP pipeline and/or Natural Language Processing would be an asset