parsingrules

parsingrules - <space> is a...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Here are some comments on what constitutes a "word" during parsing, i.e., how to decide where words start and end. From emails exchanged with a student. On Sat, 8 Oct 2005 xxx wrote: > hi, Professor Suel > i have a question on stemming. you wrote "you do not have to apply > stemming for english language pages", what do you mean? Do you mean > that I don't have to stem "eating" to "eat", "apples" to "apple"? > But for some words like "--end", "to:",. .., I think I can stem them to > "end", "to",. .., with few lines of code. I would say no stemming. Keep it simple. So "eating" stays "eating" and "eat" stays "eat" The issue of "--end" and "to:" has nothing to do with stemming. It just means that you need to make a good decision about what is a word, and what is a "separating character" that marks the beginning or end of a word. For example,
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: <space> is a separating character, and also "," or "." or ":" or "(" or ")" or "-" or many more. Decide if you should index words that include numbers, like smith007 or 7up or 7865 (I think you should). Make up reasonable rules, but the details are up to you. So "--end" should become "end" and "to:" should become "to", since "-" and ":" are separating characters.- Torsten On Sat, 8 Oct 2005 xxx wrote: > you say it is recommended to keep all stop words, so you mean I > should also store the meaningless "words" like ".", "2", "=", > "3.x",. ..? "." and "=" are not words, but separating characters. "2" is a word, and "3.x" has two words, "3" and "x". Basically, anything between separating characters is a word, unless it is within an HTML tag....
View Full Document

Ask a homework question - tutors are online