CB.

Computationele linguïstiek (KI2V13007)

Completed: 04-07-2023 | 7.5 EC | Universiteit Utrecht

What I Learned

In (KI2V13007), I applied computational methods to linguistic analysis and modeling, combining theory and practice through lectures and Python-based practica, assessed via a final exam and two assignments. Below is a breakdown of the topics covered:

Language Modeling Foundations

Edit Distance: Learned to measure string similarity for text processing.

N-grams: Studied probabilistic language models and smoothing techniques.

Regular Expressions: Explored pattern matching in text corpora.

 

Linguistic Analysis Techniques

Part-of-Speech Tagging: Gain familiarity on tagging with hidden Markov models.

Information Retrieval: Analyzed documentEIA document ranking and retrieval methods.

Speech Synthesis: Investigated techniques for generating spoken language.

 

Probabilistic and Classification Methods

Bayesian Statistics: Covered fundamentals for probabilistic modeling.

Classification: Learned naive Bayes and logistic regression for text tasks.

Gaussian Mixture Models: Explored clustering with GMMs and introduced to the EM algorithm.

 

Speech Technology

Automatic Speech Recognition: Studied methods for converting speech to text.

Corpus Analysis: Worked with text and speech datasets.

 

Practical Application

Assignment 1: Corpus Analysis: Wrote a program for text searching and N-gram modeling, using regex and file I/O.

Assignment 2: Tagging and Classification: Developed a POS tagger and classifier, integrating HMMs and Bayesian methods using NLTK.