Coursehero >> Washington >> Washington >> STAT 570

Course Hero has millions of student submitted documents similar to the one below including study guides, homework solutions, papers, exam answer keys and textbook solutions.

Tokenizing Stemming & Stemming Used in many IR applications For building equivalence classes Connect Connected Connecting Connection Connections Same class; suffixes irrelevant Porter Stemmer, simple and efficient Website: http://www.tartarus.org/~martin/PorterStemmer On patas: ~/dropbox/08-09/570/porter Stemming and Performance Does stemming help IR performance? Harman 91 indicated that it hurt as much as it helped Krovetz 93 shows that stemming does help Porter-like algorithms work well with smaller documents Krovetz proposes that stemming loses information Derivational morphemes tell us something that helps identify word senses (and helps in IR) Stemming them = information loss Evaluating Performance Measures of Stemming Performance rely on similar metrics used in IR: Precision: measure of the proportion of selected items the system got right precision = tp / (tp + fp) # of correct answers / # of answers given Recall: measure of the proportion of the target items the system selected recall = tp / (tp + fn) # of correct answers / # of possible correct answers Rule of thumb: as precision increases, recall drops, and vice versa What about for other NLP tasks, such as tagging? Metrics widely in adopted Stat NLP Precision and Recall Take a given stemming task Suppose there are 100 words that could be stemmed A stemmer gets 52 of these right (tp) But it inadvertently stems 10 others (fp) Precision = 52 / (52 + 10) = .84 Recall = 52 / (52 + 48) = .52 Note: easy to get precision of 1.0. Why? Baseline Tokenizer 1 Tokenizer 2 Tokenizer 3 After After After After coming coming coming coming close close close close to to to to a a a a partial partial partial partial settlement ettlement settlement settlement s a a a a year year year year ago ago ago ago , , , , shareholders shareholders shareholders shareholders who who who who filed filed filed filed civil civil civil civil suits suits suits suits against against against against Ivan Ivan Ivan Ivan F. F F. F. Boesky . Boesky Boesky & Boesky & & Co. & Co. Co L.P. Co L.P. . Drexel . Drexel's L.P. 's L.P plaintiffs Drexel plaintiffs . ' 's ' Drexel plaintiffs 's ' plaintiffs Tokenizer 4 After coming close to a partial settlement a year ago , shareholders who filed civil suits against Ivan F. Boesky & Co. L.P. Drexel 's plaintiffs ' Precision Recall F-Measure 0.8 0.888889 0.844444444 Tokenizer 1 Tokenizer 2 0.961538 0.925926 0.943732194 Tokenizer 3 0.928571 0.962963 0.945767196 1 1 1 Tokenizer 4 1

Find millions of documents here - Study Guides, Homework Solutions, Papers, Exam Answer Keys and more. Course Hero has millions of course related materials that will enable you to learn better, faster and get an A in all your courses.
Below is a small sample set of documents: