This is a Machine Learning Task for the Junior Data Scientist position at AIM Technologies.

Introduction

Many countries speak Arabic; however, each country has its own dialect, the aim of this task is to build a model that predicts the dialect given the text.

Guidelines

• You are given a dataset which has 2 columns, id and dialect.
• Target label column is the “dialect”, which has 18 classes.
• The “id” column will be used to retrieve the text, to do that, you need to call this API by a POST request. https://recruitment.aimtechnologies.co/ai-tasks
• The request body must be a JSON as a list of strings, and the size of the list must NOT exceed 1000.
• The API will return a dictionary where the keys are the ids, and the values are the text, here is a request and response sample.

• The dataset and the dialect identification problem were addressed by Qatar Computing Research Institute, moreover, they published a paper, feel free to get more insights from it, https://arxiv.org/pdf/2005.06557.pdf
• You must use python.
• Choose the most suitable data pre-processing techniques.
• Train two models, a machine learning model and a deep learning model, then compare the results (You are free to choose any ML algorithm and any DL architecture)
• Use Flask or FastAPI or any suitable web framework to deploy the model locally.
• Push the source code to a GitHub repo, it’s preferred to be a well-documented repo.

Deliverables

1. A GitHub repo link, which has the following structure:

• a. Data fetching script/notebook
• b. Data pre-processing script/notebook
• c. Model Training script/notebook
• d. Deployment script/notebook
• e. Any additional files/documentation you need

2. A PowerPoint presentation summarizing your approach, data pre-processing, model architecture, evaluation metrics and results. (Max 8 slides)

Notes

• Make sure that the GitHub repo is public.
• In the assessment phase, you’ll be asked to run your models locally, furthermore, you’ll be asked in any technical decision/implementation you’ve made, so be well prepared, and avoid over-complicated approaches you don’t fully grasp.
• The deadline is 12 days from the day you’ve received the task, specific dates are written in the email.
• Early submission doesn’t affect your grade, take your time.
• To submit the task, reply to the email you’ve received with the deliverables.