This page contains a variety of resources that I have written or gathered as part of my classwork, teaching, and research.

Fortune Cookie Corpus

I used fortune cookie fortunes as part of an assignment that I wrote for an undergraduate AI class at CMU. For the assignment, students were asked to design an algorithm that could classify a piece of text as either a fortune or not a fortune. (One of my pet peeves is opening a so-called "fortune cookie" only to find a piece of advice or compliment of some kind.) Below you will find my fortune cookie corpus, the text of 138 cookies that I have collected with the help of my family and friends. I do possess the original paper fortunes for all of the text given below. Some of the fortunes are duplicated: the text files do not include duplicates, the CSV files have a column which indicates the frequency of each fortune.

I classify text from a fortune cookie as an actual fortune if it makes a prediction or provides some information that the reader could not otherwise know.

Files, last updated 3/19/06:

• yes.txt (49 entries) Text from fortune cookies that I consider to be actual fortunes, one per line.
• no.txt (77 entries) Text from fortune cookies that I do not consider to be actual fortunes, one per line.
• yes.csv and no.csv which also include the frequency of each piece of text (representing 138 cookies total).

The Kana Project

This is a free, open-source repository consisting of two software programs designed to help beginning Japanese students learn the two Japanese phonetic alphabets, hiragana and katakana. Source code and documentation are available at the Kana Project homepage.