Automated Email Classification Engine

SVM spam/ham classifier with custom parsing and zero train/test leakage.

Macro F10.97
Accuracy97%

Repository

Python
Scikit-Learn
SVM
NLP

What it does

Classifies emails as spam or ham with a Support Vector Machine.

Custom email parsing and cleaning.
Vocabulary building and count vectorization.
Linear SVM classifier: 0.97 macro F1 and 97% accuracy, with no train/test leakage.

Workflow

Solid lines are the training emails; dashed lines are the test emails, which go through the same steps but are only used to evaluate the trained SVM. Tap or click the diagram to open it at full size.

"""Automated Email Classification Engine

SVM spam/ham classifier with custom parsing and zero train/test leakage.
"""

from portfolio import Metric, Project


class EmailClassifier(Project):
    status = "complete"
    featured = True
    stack = ["Python", "Scikit-Learn", "SVM", "NLP"]
    metrics = [Metric("Macro F1", "0.97"), Metric("Accuracy", "97%")]
    links = {
        "repo": "https://github.com/maaz7409/email_classifier",
    }
    demo = None

    highlights = [
        "Custom email parsing and cleaning.",
        "Vocabulary building and count vectorization.",
        "Linear SVM classifier: 0.97 macro F1 and 97% accuracy, with no train/test leakage.",
    ]

Automated Email Classification Engine

What it does

Workflow

Open Editors

Best on a bigger screen

What it does

Workflow

email_classifier.py source

Open Editors