ANLP 2017 Course Revision Guide

Disclaimers:

  1. This page provides a list of concepts you should be familiar with and questions you should be able to answer if you are thoroughly familiar with the material in the course. It is safe to assume that if you have a good grasp of everything listed here, you will do well on the exam. However, we cannot guarantee that only the topics mentioned here, and nothing else, will appear on the exam. In a few cases (mainly formulas) we do specify what *will not* be required for the exam, but otherwise we make no guarantees.

Exam rubric

The exam will have 3 questions with the following rubric:

Answer QUESTION 1 and ONE other question.
Question 1 is COMPULSORY.
If both QUESTION 2 and QUESTION 3 are answered, only QUESTION 2 will be marked.
All questions carry equal weight.
CALCULATORS MAY NOT BE USED IN THIS EXAMINATION.

Past papers

Previous versions of this course were called Advanced Natural Language Processing through the 2014-15 academic year. You can look up past exam papers if you wish.

However, about 25% of the material in this course has changed since last year, and there was a previous major revision and change of both instructors in 2014-15. Exams prior to 2014-15 therefore have a significant amount of material that we no longer cover, and a somewhat different question style. These are not recommended as study guides.

Since last year's course, the main topics we have eliminated are active chart parsing, details of advanced smoothing methods, and nearly all topics related to discourse. The main topics we have added are dependency parsing, neural networks, and more material on distributional semantics, data analysis, experiments, and evaluation.

In working through papers from the past three years, you should not expect to be able to answer the following questions:

In addition, the rubric of the exam is different this year, as noted above.

Generative probabilistic models

We have discussed the following generative probabilistic models:

For each of these, you should be able to

Discriminative probabilistic models

We have discussed the following discriminative probabilistic model:

For this model, you should be able to

Other formulas

In addition to the equations for the models listed above, you should know the formulas for the following concepts, what they may be used for, and be able to apply them appropriately. Where relevant you should be able to discuss strengths and weaknesses of the associated method, and alternatives.

Algorithms and computational methods

For each of the following algorithms, you should be able to explain what each of these computes (its input and output), what it is used for, and be able to hand simulate each one. Some of these algorithms are naive, or solve problems that more naive algorithms face. You should be able to explain what those problems are and how the better algorithms solve them.

For each of the following methods, we haven't discussed algorithms at the level of data structures or implemention, but you should still be able to explain what each method computes (its input and output), what it is used for, and be able to hand simulate each one.

For each of the following methods, you should be able to explain what it computes (its input and output), what it is used for, and be able to describe how it works in some detail.

Additional Mathematical and Computational Concepts

Overarching concepts:

In addition, for the following concepts you should be able to explain each one, give one or two examples where appropriate, and be able to identify examples if given to you. You should be able to say what NLP tasks these are relevant to and why.

Linguistic and Representational Concepts

You should be able to explain each of these concepts, give one or two examples where appropriate, and be able to identify examples if given to you. You should be able to say what NLP tasks these are relevant to and why.

Tasks

You should be able to explain each of these tasks, give one or two examples where appropriate, and discuss cases of ambiguity or what makes the task difficult. In most cases you should be able to say what algorithm(s) or general method(s) can be used to solve the task, and what evaluation method(s) are typically used.

Resources

You should be able to describe what linguistic information is captured in each of these resources, and how it might be used in an NLP system.

You should also be able to identify legal and ethical issues in the creation and collection of linguistic resources.

Evaluation concepts and methods

For each of the following, you should be able to explain what each of the specific methods measures, what tasks it would be appropriate for, and why.

In addition:


Home : Teaching : Courses : Anlp