original:https://class.coursera.org/nlp/lecture/preview
Video Lectures
Week 1 – Course Introduction
Week 1 – Basic Text Processing
- Regular Expressions (11:25)
- Regular Expressions in Practical NLP (6:04)
- Word Tokenization (14:26)
- Word Normalization and Stemming (11:47)
- Sentence Segmentation (5:31)
Week 1 – Edit Distance
- Defining Minimum Edit Distance (7:04)
- Computing Minimum Edit Distance (5:54)
- Backtrace for Computing Alignments (5:55)
- Weighted Minimum Edit Distance (2:47)
- Minimum Edit Distance in Computational Biology (9:29)
Week 2 – Language Modeling
- Introduction to N-grams (8:41)
- Estimating N-gram Probabilities (9:38)
- Evaluation and Perplexity (11:09)
- Generalization and Zeros (5:15)
- Smoothing: Add-One (6:30)
- Interpolation (10:25)
- Good-Turing Smoothing (15:35)
- Kneser-Ney Smoothing (8:59)
Week 2 – Spelling Correction
- The Spelling Correction Task (5:39)
- The Noisy Channel Model of Spelling (19:30)
- Real-Word Spelling Correction (9:19)
- State of the Art Systems (7:10)
Week 3 – Text Classification
- What is Text Classification? (8:12)
- Naive Bayes (3:19)
- Formalizing the Naive Bayes Classifier (9:28)
- Naive Bayes: Learning (5:22)
- Naive Bayes: Relationship to Language Modeling (4:35)
- Multinomial Naive Bayes: A Worked Example (8:58)
- Precision, Recall, and the F measure (16:16)
- Text Classification: Evaluation (7:17)
- Practical Issues in Text Classification (5:56)
Week 3 – Sentiment Analysis
- What is Sentiment Analysis? (7:17)
- Sentiment Analysis: A baseline algorithm (13:27)
- Sentiment Lexicons (8:37)
- Learning Sentiment Lexicons (14:45)
- Other Sentiment Tasks (11:01)
Week 4 – Discriminative classifiers: Maximum Entropy classifiers
- Generative vs. Discriminative Models (7:49)
- Making features from text for discriminative NLP models (18:11)
- Feature-Based Linear Classifiers (13:34)
- Building a Maxent Model: The Nuts and Bolts (8:04)
- Generative vs. Discriminative models: The problem of overcounting evidence (12:15)
- Maximizing the Likelihood (10:29)
Week 4 – Named entity recognition and Maximum Entropy Sequence Models
- Introduction to Information Extraction (9:18)
- Evaluation of Named Entity Recognition (6:34)
- Sequence Models for Named Entity Recognition (15:05)
- Maximum Entropy Sequence Models (13:01)
Week 4 – Relation Extraction
- What is Relation Extraction? (9:47)
- Using Patterns to Extract Relations (6:17)
- Supervised Relation Extraction (10:51)
- Semi-Supervised and Unsupervised Relation Extraction (9:53)
Week 5 – Advanced Maximum Entropy Models
- The Maximum Entropy Model Presentation (12:14)
- Feature Overlap/Feature Interaction (12:51)
- Conditional Maxent Models for Classification (4:11)
- Smoothing/Regularization/Priors for Maxent Models (29:24)
Week 5 – POS Tagging
- An Intro to Parts of Speech and POS Tagging (13:19)
- Some Methods and Results on Sequence Models for POS Tagging (13:04)
Week 5 – Parsing Introduction
- Syntactic Structure: Constituency vs Dependency (8:46)
- Empirical/Data-Driven Approach to Parsing (7:11)
- The Exponential Problem in Parsing (14:30)
Week 5 – Instructor Chat
Week 6 – Probabilistic Parsing
- CFGs and PCFGs (15:29)
- Grammar Transforms (12:05)
- CKY Parsing (23:25)
- CKY Example (21:56)
- Constituency Parser Evaluation (9:45)
Week 6 – Lexicalized Parsing
- Lexicalization of PCFGs (7:03)
- Charniak’s Model (18:23)
- PCFG Independence Assumptions (9:44)
- The Return of Unlexicalized PCFGs (20:53)
- Latent Variable PCFGs (12:07)
Week 6 – Dependency Parsing (Optional)
- Dependency Parsing Introduction (10:25)
- Greedy Transition-Based Parsing (31:05)
- Dependencies Encode Relational Structure (7:20)
Week 7 – Information Retrieval
- Introduction to Information Retrieval (9:16)
- Term-Document Incidence Matrices (8:59)
- The Inverted Index (10:42)
- Query Processing with the Inverted Index (6:43)
- Phrase Queries and Positional Indexes (19:45)
Week 7 – Ranked Information Retrieval