Skip to content

stratforge/ml-code-challenge

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

ML code challenge

⏰ Time Required: 60 - 100 hours


👋 Introduction

For this challenge, you need to develop a Text Classification model.

We will assess the following:

  1. Coding Standards: Adherence to clean, readable, and well-documented code.
  2. Best Practices: Effective implementation of machine learning principles and robust methodology.
  3. Complexity: Balance between simplicity and performance; ensuring the model is both effective and efficient.
  4. Performance: Evaluation of the model's effectiveness and accuracy based on performance metrics.

⚔️ Challenge

Important!

You are tasked with developing a Text Classification Model. Below is a detailed breakdown of the task.

Description

Build a classification model to categorize online product reviews.

Dataset

Provided dataset of online product reviews, each labeled with one or more product feature category that are given below:

  • Camera
  • Battery
  • Performance
  • Design
  • Screen
  • Price
  • Build Quality

The dataset includes raw text reviews, some of which might be poorly written, contain irrelevant information (e.g., shipping complaints), or be misclassified. Some reviews might belong to multiple categories.


Guide:

  • Text Preprocessing and Feature Engineering: Clean and preprocess the review text. This might involve handling noise, removing irrelevant information, and extracting relevant features. Explore various text representation techniques.

  • Classification Handling: Implement a strategy to handle reviews that belong to multiple categories. This might involve implementing classification ML algorithms or neural network model. Experiment building multiple models to showcase its advantages over the other and derive to final optimal model.

  • Model Evaluation: Evaluate the performance of your model using appropriate metrics

  • Error Analysis: Perform a detailed error analysis to identify common misclassifications and areas for improvement. Discuss potential strategies to address these issues.

  • Code and Documentation: Submit well-documented code and a comprehensive report detailing your approach, including data preprocessing steps, feature engineering techniques and model building.

About

ML Code Challenge ⚔️

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published