Exam review topics

What is, discuss, true or false, choose
Text segmentation

Word segmentation
Sentence splitting
Word segmentation
Segmentation problems
Sentence segmentation
Automatic segmentation approaches
, part-of-speech tagging (POS tagging or POST),grammatical tagging, word-category disambiguation
POS Principle
Use of Hidden Markov Models HMMs
CLAWS
Dynamic Programming methods
Unsupervised taggers
Other taggers and methods, issues
Parsing
Traditional methods vs. computational methods
Parser, types

Top-down parsers
Some of the parsers that use top-down parsing include:
• Recursive descent parser
• LL parser (Left-to-right, Leftmost derivation) • Earley parser
Bottom-up parsers
Some of the parsers that use bottom-up parsing include:
• Precedence parser
• Operator-precedence parser
• Simple precedence parser
• BC (bounded context) parsing
• LR parser (Left-to-right, Rightmost derivation)


Parser development software
Some of the well known parser development tools include the following. Also see comparison of parser generators.
• ANTLR
• Bison
• Coco/R
• GOLD
• JavaCC

Lookahead
Finite-state machine: first 4 paragraphs
Concepts and vocabulary
Transducers
Discourse analysis
Topics of interest
Perspectives
Dialogical analysis definition

major: 25%
summary sheets: 20
summary presentation: 15

_
60
Name
Major
Research
Final Presentationj
Total/60
Nora AlAngary
25
20
12
57
AlJohara AlSobai
0
20
11
32
Sara AlSaud
0
20
10
30
Sita
25
20
15
60
Aseel
25
20
12
57
Review Questions:
1- What is bayes rule?
2- Why studying Bayes Rule is important?
3- What is the advantage of the Baysian Approach? Slide 18
4- What is it used for? Slide 18
5- What are the elements of the Baysian Reasoning? Slide 19
6- Mention the bayes rule formula?
7- A problem:
Sue has written 100 essays of 5000 words in 3 years. She made spelling mistakes for the word determine which was misspelled for ditermine 100 times. Tomorrow, she has a TOEFL exam. What are the chances that she would make mistakes for the word determine tomorrow considering the fact that in the first 2 years she made 90 mistakes and in the third year she corrected her mistakes which became 10 only?
In order to answer this please review this website: http://stattrek.com/probability/bayes-theorem.aspx
- First you have to mention the formula which is half of the grade.
8- MLE
8- Definition
9- Purpose
10- Formula
11- A question: Divide the number of occurences of a single word on the total number of a corpus.Example, The word THE Occured 100 times in a corpus of 10000 words so the MLE for THEis: 100/10000= 0.01
Bayes rule ppt p 5, 8, 18, 19

OUR GENERATED BOOK FOR THE COURSE







http://nlp.stanford.edu/links/statnlp.html#Taggers




Todays Class 18th Feb, 2013 on MLE: Maximum Likelihood estimation
Divide the number of occurences of a single word on the total number of a corpus.
Example, The word THE Occured 100 times in a corpus of 10000 words so the MLE for THEis: 100/10000= 0.01