Skip to content

Latest commit

 

History

History
63 lines (52 loc) · 3.71 KB

README.md

File metadata and controls

63 lines (52 loc) · 3.71 KB
Raw

Diamond Price Prediction Repository

Description

This is a Mini-Project for SC1015 (Introduction to Data Science and Artificial Intelligence) which uses a diamond dataset from Kaggle.

Contributors

School of Computer Science and Engineering Nanyang Technological University Lab: A136 Group: 8

Name GitHub Acount Main Contributions
Juan Frederick @dreonic Data Resampling, Machine Learning, Hyperparameter Tuning
Hilarius Jeremy @HilariusJeremy Data Visualisation, Problem Definition, Data Driven Insight
Poh Jing Ting, Felicia @feliciapjt Exploratory Data Analysis, Machine Learning, Problem Definition

Problem Statement

  • Building an accurate model for estimation and prediction diamond price
  • Determine related factors to diamond price to suggest further action from buyers and sellers.

Workflow

  1. Identifying Data and Data Cleaning
  2. Exploratory Data Analysis
  3. Machine Learning
  4. Hyperparametric Tuning and Optimization
  5. Data Driven Insights

Models Used

  • Linear Regression > A simple model to predict the dependent variable based on the independent variable by fitting a line
  • Random Forest > A classification algorithm that involves many decision trees, using bagging and feature randomness to build each tree
  • XGBoost > An implmentation of gradient-boosting decision trees. It uses a gradient descent algorithm to minimize the loss function and it includes regularization techniques to avoid overfitting.

Conclusion

  • Price of diamonds can reliably be predicted using their attributes
  • XGBoost is the best model we found to predict price
  • Main variables that affect price: carat weight, cut, lab certificate
  • Color, culet size, and cut quality is irrelevant to predict diamond price

What We Have Learnt From this Project

  • How to handle missing values using imputation
  • Types of visualizations to better understand relationship between variables
  • Encoding numerical variables to get better results
  • Handling random forest and XGBoost models along with their parameters
  • Hyperparametric tuning and how to choose the parameters
  • How to use GitHub and markdown syntax

References

  1. 123dartist. (n.d.). Luxury Shinning Diamond Balanced on Rippled Water Surface with Reflected Sky Background, Diamond Wallpaper, [Image]. Adobe Stock. https://stock.adobe.com/sg/images/luxury-shinning-diamond-balanced-on-rippled-water-surface-with-reflected-sky-background-diamond-wallpaper/452760672?prev_url=detail
  2. How to Deal with Missing Data. (n.d.). Master’s in Data Science. https://www.mastersindatascience.org/learning/how-to-deal-with-missing-data/
  3. Upgrade Your Diamond. (n.d.). Long Jewelers. https://www.longjewelers.net/Upgrade-Your-Diamond/3001108/EN
  4. HROKR. (2023). The largest diamond dataset currently on Kaggle. Kaggle. https://www.kaggle.com/code/hrokrin/kaggle-s-largest-dataset-of-diamonds-eda
  5. Kulkarni, M. (n.d.). Machine Learning Notebook. GitHub. https://github.com/maykulkarni/Machine-Learning-Notebooks/tree/master/02.%20Regression
  6. Yiu, T. (2019). Understanding Random Forest. Towards Data Science. https://towardsdatascience.com/understanding-random-forest-58381e0602d2
  7. What is linear regression? (n.d.). IBM. https://www.ibm.com/topics/linear-regression#:~:text=Resources-,What%20is%20linear%20regression%3F,is%20called%20the%20independent%20variable.
  8. Brownlee, J. (2016). A Gentle Introduction to XGBoost for Applied Machine Learning. Machine Learning Mastery. https://machinelearningmastery.com/gentle-introduction-xgboost-applied-machine-learning/