francesita,github

👋 Hi, I’m Frances Laureano, I am a first year PhD student at the University of Birmingham, UK.
👀 I’m interested in natural language processing (NLP). In particular, I am interested in multilingual word-embeddings, sentiment analysis, fake news and hatespeech detection.
🌱 I’m currently learning how to better use transformers for projects in NLP
💞️ I’m looking to collaborate on projects using NLP, as well as working on applications of NLP for social science and business.
📫 I can be reached at fxl846[@]cs.bham.ac.uk. For more information on projects visit my website: franceslaureano.com

francesita's Projects

code-mixed-probes

This repository consists of the code and data used for the COLING-LREC 2024 paper 'Code-Mixed Probes Show How Pre-Trained Models Generalise On Code-Switched Text '.

code-switch-embeddings-for-sentiment-analysis

Code for Code-switch embedding for Master's project. This was preliminary work used for SemEval and this work was presented at TechLing 2019 in Galicia, Spain.

cs-embed-semeval2020

Code and specs for CS-Embed's entry for SemEval-2020 Task-9. We present code-switched embeddings, code for code-switched bilstm sentiment classifier, and code for CS tweet collection.

Contains code used for the menopause project. The purpose was to analyze web articles and tweets to come up with a research topic dealing with menopause. This end was achieved by using different NLP techniques, including tfidf, and sentiment analysis.

msc_course

Code for some assignments from the MSc Computer Science course

mtglibrary

Program to keep track of magic collection and make decks

predictive_text_project

A predictive text dictionary. Contains code for prototype and different interations of Dictionary. The final form is a dictionary class that uses a tree structure which allows for partial signature matching so users can see part of the words that match as they type.