2.4. Machine Learning for Text Analytics with scikit-learn

Machine Learning for Text Analytics with scikit-learn

scikit-learn is a Python module integrating classic machine learning algorithms in the tightly-knit world of scientific Python packages (numpy, scipy, matplotlib).

It aims to provide simple and efficient solutions to learning problems that are accessible to everybody and reusable in various contexts: machine-learning as a versatile tool for science and engineering.

Note

This document is meant to be used with scikit-learn version 0.11+ (i.e. the current state of the master branch at the time of writing: 2012-02-13).

• 2.4.1. Tutorial setup
  • 2.4.1.1. Install scikit-learn build dependencies
  • 2.4.1.2. Build scikit-learn from source
  • 2.4.1.3. Download the datasets
• 2.4.2. Machine Learning 101: General Concepts
  • 2.4.2.1. Features and feature extraction
  • 2.4.2.2. Supervised Learning: model.fit(X, y)
  • 2.4.2.3. Unsupervised Learning: model.fit(X)
  • 2.4.2.4. Linearly separable data
  • 2.4.2.5. Training set, test set and overfitting
  • 2.4.2.6. Key takeaway points
• 2.4.3. Working with text data
  • 2.4.3.1. Downloading the data and loading it from Python
  • 2.4.3.2. Extracting features from text files
  • 2.4.3.3. Training a classifier
  • 2.4.3.4. Building a pipeline
  • 2.4.3.5. Evaluation of the performance on the test set
  • 2.4.3.6. Parameter tuning using grid search
• 2.4.4. Exercises
  • 2.4.4.1. Exercise 0: Digits recognition
  • 2.4.4.2. Exercise 1: Language identification
  • 2.4.4.3. Exercise 2: Sentiment Analysis on movie reviews
  • 2.4.4.4. Exercise 3: Unsupervised topic extraction
  • 2.4.4.5. Exercise 4: CLI text classification utility
• 2.4.5. How to select the right algorithm for the task
• 2.4.6. Finding help
  • 2.4.6.1. The project mailing list
  • 2.4.6.2. Q&A communities with Machine Learning practictioners