Homework 

For due dates, please see the schedule.

Homework 1 (hw1) is an exploration of the difficulties of evaluating relevance of a query.  It requires you to judge the relevance of some Web pages as part of the TREC 2007 Million Query track.  It was worth 20 points.

Homework 2 (hw2) covers questions about evaluation, statistics of text, and term weighting.  It includes 5 problems, each worth 20 points.

Homework 3 (hw3) touches on statistics of text, compression, relevance feedback, and clustering.  It includes 4 problems, each worth 20 points.

Homework 4 (hw4) is not yet assigned.

Homework 5 (hw5) is not yet assigned.