Author: Marina Santini

Computational Linguist, PhD

Lecture 9: Perceptron

Course website: Machine Learning for Language Technology Lecture 9 Perceptron from Marina Santini Outline: feature representation, main theorem, margin and separability, perceptron

Lecture 8: Decision Trees and k-Nearest Neighbors

Lecture 8: Decision Trees & k-Nearest Neighbors from Marina Santini Outline Decision trees, best split, entropy, information gain, gain ratio, k-Nearest Neighbors, distance metric. Course Website:

Lecture 7: Hidden Markov Models

Lecture 7: Hidden Markov Models (HMMs) from Marina Santini Outline: Hidden Markov Models (HMMs), Markov Assumptions, Problems for HMMs, Algorithms for HMMs, POS Tagging with HMMs, Smoothing for POS Tagging Course Website:

Lecture 5 Bayesian Classification

Lecture 5: Bayesian Classification from Marina Santini Outline: Bayesian Classification, Instance Attributes, Naive Bayes Classifiers, Naive Bayes in NLP, Spam Filtering Course Website:

Lecture 4: Statistical Inference

Lecture 4: Statistical Inference from Marina Santini Outline: stochastic variables, frequency functions, expectations, variance, entropy, joint probabilities, conditional probabilities, independence, sampling, estimation, maximum likelihood estimation (MLE), smoothing, hypothesis testing,z-test.

Spreading the Word about (Web)Genre Research

What is genre? Why is it useful to master genre conventions? Can we classify document genres automatically? Around the world, lots of researches and scholars belonging to a wide range of disciplines are trying to provide answers to these and…