Probability < Main

---+ Probability for Linguists

Probability is the branch of mathematics concerned with describing the relative likelihood of things.  Probabilitic methods have proven fruitful in characterizing patterns in natural language.  Here are some starting points for teaching yourself the math.

   * One of the best books on the subject of probability and natural language is [[http://www.amazon.com/exec/obidos/tg/detail/-/0262133601/qid=1106342854/sr=8-1/ref=sr_8_xs_ap_i1_xgl14/002-1238136-9957629?v=glance&s=books&n=507846][Foundations of Statistical Natural Language Processing]].

   * [[http://humanities.uchicago.edu/faculty/goldsmith/][John Goldsmith]] has a good tutorial called [[http://humanities.uchicago.edu/faculty/goldsmith/Industrial/Probability.htm][Probability for Linguists]] that uses examples from natural language.

   * The _[[%ATTACHURL%/two_dice.pdf][Probability Tutorial Using Dice]]_ explains the mathematical concepts using a familiar real world example.

   * The _[[%ATTACHURL%/mmic.pdf][Maximum Mutual Information Criterion Tutorial]]_ explains how to use mutual information in feature reduction.  A [[%ATTACHURL%/mmic.py.txt][sample script]] shows how this works.

-- Main.BillMcNeill - 21 Jan 2005

Attachments

Topic attachments
I	Attachment	History	Action	Size	Date	Who	Comment
pdf	mmic.pdf	r2 r1	manage	35.8 K	2005-04-13 - 19:08	UnknownUser	Maximum Mutual Information Criterion Tutorial
txt	mmic.py.txt	r4 r3 r2 r1	manage	5.3 K	2005-04-12 - 23:46	UnknownUser	Example script. Save as mmic.py
pdf	two_dice.pdf	r1	manage	57.3 K	2005-01-25 - 05:42	UnknownUser	Two dice

This topic: Main > WebHome > Probability
Topic revision: r6 - 2005-04-13 - BillMcNeill