ML.NET.Classifier

Overview

ML.NET.Classifier is a .NET Windows Forms application that utilizes the ML.NET library to demonstrate binary and textual data classification process using relevant metrics and visual charts.

ML.NET.Classifier binary classification example

Binary classification overview

The ML.NET.Classifier applies logistic regression and averaged perceptron algorithms to the binary classification of data, using the Pima Indians Diabetes Database, Kaggle.com. to demonstrate how machine learning can categorize data points into two mutually exclusive groups, such as predicting the likelihood of diabetes based on features like blood glucose levels, age, and BMI.

ML.NET.Classifier textual classification example

Textual classification overview

For textual classification, decision tree and naive Bayes algorithms are utilized, using the SMS Spam Collection Dataset, Kaggle.com to distinguish spam from ham messages.

Algorithms and Evaluation Methods

Binary Classification:

• Algorithm model - logistic regression and averaged perceptron.
• Model metric - accuracy, precision, recall.
• Performance metric - performance evaluation of predictions, receiver operating characteristic curve.

Textual Classification:

• Algorithm model - decision tree and naive Bayes classifier.
• Model metric - macro accuracy, micro accuracy, log Loss.
• Performance metric - confusion matrix evaluation, cumulative gains chart.

Dataset

• Binary classification dataset: Pima Indians Diabetes Database, Kaggle.com.
• Textual classification dataset: SMS Spam Collection Dataset, Kaggle.com.

Libraries

• .NET (v8.0)
• ML.NET (v3.0.1)
• Microsoft.ML.LightGbm (v3.0.1)
• Microsoft.ML.CpuMath (v3.0.1)
• WinForms.DataVisualization (v1.9.2)
• CsvHelper (v32.0.3)
• Microsoft.NET.Test.Sdk (v17.6.0)
• NUnit (v3.13.3)