Predicting Emojis in Tweets

Hugging Face: Exploring tweet text and emoji labels to uncover patterns, insights, and key questions.

Assignment

Your task is to build a predictive model of emoji in given a piece of text. Our suggestion is to use standard vectorization methods and machine learning classifiers (of your choice). Prepare a script for conducting the experiment, an instruction to reproduce your experiment results, and a report that includes the following:

1. Experiment results, including the standard metrics e.g. precisions, accuracies, f-measures, and confusion matrices (if applicable).
2. Your findings and observations are based on the experiments, including limitations and assumptions.
3. How you choose the vectorization method and algorithms.
4. What the experiments may be useful for and how we may improve their utility of it.

Data Description

Included in this repository are two files:

1. tweets.txt, where each line includes the text of a tweet that included emoji (but the emoji has been removed);
2. emoji.txt, where each line includes the name of the emoji for the corresponding text in tweets.txt.

Practicalities

We expect candidate to demonstrate the following abilities through this assignment:

1. An ability to write well-organized and easy-to-read code in general programming
2. An ability to design an empirical research experiment
3. An ability to articulate the assumptions and limitations of research
4. An ability to think out of the box and propose alternative paths, improvements, and future work

Of course, everyone is not perfect and you have limited time :-) Our suggestion is to spend a couple of hours completing the task and let your intellectual curiosity guide you from there.