Natural Language Processing with Classification and Vector Spaces Course Syllabus

Overview: This course provides a comprehensive introduction to foundational Natural Language Processing (NLP) techniques, focusing on classification, vector space models, and semantic similarity. Through hands-on assignments and practical examples, you'll learn to implement sentiment analysis classifiers, build word vectors, and develop a simple machine translation system. The course spans approximately 33 hours, designed for flexible learning that fits working professionals. Each module combines theory with coding exercises to reinforce understanding and application.

Module 1: Sentiment Analysis with Logistic Regression

Estimated time: 9 hours

• Extract features from text into numerical vectors
• Build a binary classifier for tweets using logistic regression
• Understand preprocessing steps in NLP pipelines
• Apply feature extraction techniques for sentiment analysis

Module 2: Sentiment Analysis with Naïve Bayes

Estimated time: 8 hours

• Learn the theory behind Bayes’ rule and conditional probabilities
• Apply Naïve Bayes to classify tweet sentiment
• Implement a Naïve Bayes classifier from scratch
• Compare performance with logistic regression models

Module 3: Vector Space Models

Estimated time: 8 hours

• Create word vectors that capture word dependencies
• Use PCA for dimensionality reduction and visualization
• Explore semantic meaning in vector spaces
• Analyze relationships between words using cosine similarity

Module 4: Machine Translation and Document Search

Estimated time: 8 hours

• Transform word vectors for cross-lingual alignment
• Use locality-sensitive hashing for approximate nearest neighbor search
• Implement a simple English-to-French translation algorithm
• Apply vector techniques to document search by semantic similarity

Prerequisites

• Familiarity with Python programming
• Basic understanding of machine learning concepts
• Knowledge of linear algebra and probability fundamentals

What You'll Be Able to Do After

• Analyze sentiment in social media text using classification models
• Construct and interpret word vectors using vector space models
• Reduce and visualize high-dimensional text data with PCA
• Build a basic machine translation system using pre-computed embeddings
• Perform semantic document search using vector similarity techniques