文件列表:

AI_Medical_Project.ipynb (163067, 2022-08-29)
ai_campers1.png (20070, 2022-08-29)
ai_campers_team.png (767448, 2022-08-29)
insurance.csv (54289, 2022-08-29)

Table of Contents

1. About The Project
   • Project Content
   • Built With
2. Model Selection
   • Best Performing Model - Polynomial Regression
3. Acknowledgments

## About The Project Hande ELEB - [Linkedin](https://www.linkedin.com/in/hande-%C3%A7elebi-5572461b8/) Anl Can ZGK - [Linkedin](https://www.linkedin.com/in/an%C4%B1l-can-%C3%B6zg%C3%B6k/) Azime ARIKAYA - [Linkedin](https://www.linkedin.com/in/azime-arikaya-69a766219/) Emre Z - [Linkedin](https://www.linkedin.com/in/emre-oz/) ### Project Content The aim of the project is to estimate the approximate cost of a person's health insurance based on the given variables. For this purpose, we built the infrastructure of the project with python. It is because python has library support and is easy to use. We performed exploratory data analysis with the data set given to us. We chose the most suitable model according to the data obtained afterwards. ### Built With This section contains all libraries used in the project. * Pandas ```sh import pandas as pd ``` * Numpy ```sh import numpy as nm ``` * Seaborn ```sh import seaborn as sb ``` * Matplot ```sh import pandas as pd ``` * Sklearn ```sh import sklearn as sk ``` ## Model Selection At this stage, we tried a total of 4 models. These are: Linear Regression, Polynomial Regression, Lasso Regression and Ridge Regression. As a result, the model with the highest R2 value was Polynomial Regression. ### Best Performing Model - Polynomial Regression * R2: 0.8411925923335958 ```sh from sklearn.preprocessing import PolynomialFeatures poly_reg = PolynomialFeatures(degree=5) X_poly = poly_reg.fit_transform(X) pol_reg = LinearRegression() poly = pol_reg.fit(X_poly, y) y_poly_pred = pol_reg.predict(X_poly) print(f"R2: {r2_score(y,y_poly_pred)}") print(f"MAE: {mean_absolute_error(y,y_poly_pred)}") print(f"MSE: {mean_squared_error(y,y_poly_pred)}") ``` ## Acknowledgments Here are the resources we used in the construction of this project. Thank you very much. * [Global AI Hub](https://globalaihub.com/ai-summer-camp-22/) * [Gülcan ündür](https://medium.com/@gulcanogundur/model-se%C3%A7imi-k-fold-cross-validation-4635b61f143c) * [Yasin Güne](https://www.linkedin.com/in/yasin-g%C3%BCne%C5%9F-3864631b5/) * [Günel Alizada](https://www.linkedin.com/in/gunelalizada/) * [Kaggle Data Set](https://www.kaggle.com/datasets/mirichoi0218/insurance)

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：