|
|
|
Homework Assignments
- Assignment 1
- There was an error in the
GradeReader.java file as originally
published. Download the new GradeReader.java to complete the
assignment. To make sure you have the right file, look for a
comment at the beginning of the file; it will start with "$Id:"
and say 1.2 (dated 2008/01/28).
- For final testing, additional data files have been provided:
- A grading description is now available.
- Assignment 5
- Test data files
- Ant build file (updated 5/9/08)
- Stemming library (stem data sources)
- Index dump tool
- JavaDoc
- search.sh — a shell script to help run the search program.
To use this, save it in your project's root directory, and then
run it from whatever directory you want the index file to be in,
passing it the same command line arguments your program
expects. You'll probably need to make it executable.
- Sample results. All these are against the data sets given to
you above.
- files-mainly.txt — shell transcript for a
search in the
files/ data set.
- lemuel.txt — output for a
match of ``Lemuel'' in books/
- river-match.txt — output for a
match of ``river'' in books/
- river-search.txt — output for a
search of ``river'' in books/
- lake-match.txt — output for a
match of ``lake'' in books/
- lake-search.txt — output for a
search of ``lake'' in books/
- tis.txt — output for a
match of ``tis'' in books/
- burns.txt — output for a
search of ``burns'' in books/
- wp-stats.txt — output of
stats in wikipedia/
- produced.txt — output for a
search of ``produced'' in wikipedia/
- pascal.txt — output for a
match of ``pascal'' in wikipedia/
- insertion.txt — output for a
match of ``insertion'' in
wikipedia/ (particularly pay attention to the score of
heapsort.txt in this one)
- insert.txt — output for a
search of ``insert'' in wikipedia/
- suns.txt — output for a
match of ``Sun's'' in wikipedia/
- There is a slight problem with the stemmer we have provided.
The database we're using for stem data was designed for spell
checking, not for searching, so in some cases it will return a
stem which is not optimal for computing meaningful search
results. For example, rather than saying "river" is the stem of
"rivers", it says that "rive" is the stem of both "river" and
"rivers". Therefore, searching for "river" in
books returns a
score of 43 for richardiii when it seems like it should be 129.
At this point, correcting the problem is difficult. The "lake"
search added above will be a better test ("lake" stems
correctly).
|
|
