jahn-da/final_project

MY PROJECT:

PISTACHIO CLASSIFICATION

Using dataset: https://www.kaggle.com/datasets/amirhosseinmirzaie/pistachio-types-detection

My project is to create the best machine learning model for Pistachio classification.

The column I want to predict is "Class", which has 2 values: Kirmizi Pistachio and Siirt Pistachio.

I will create 2 datasets:

1. Using correlation matrix
2. Analysing plots and graphs

I will use the following Scalers:

1. MinMacScaler
2. StandardScaler

I will create the following models:

1. KNN
2. Neural Networks
3. Random Forest
4. Decision Tree
5. Gaussian Naive Bayes
6. Support Vector Machines
7. XGBoost
8. Gradient Boosting
9. Gradient Boosting Decision Trees
10. Logistic Regression
11. ADA Boost

I will use the following metrics to find the best model:

1. Cohen Kappa Score
2. F1 Score
3. Accuracy Score

Findings:

Dataset 1 is the best dataset to use! Random Forest is the best model to use! For all 3 measure metrics! Cohen Kappa dataset 1: 0.9474 F1 dataset 1: 0.9737 Accuracy dataset 1: 0.9737