Textbook Chapter 8

In Chapter 8, there is a very good summary of research work on stemming. The main point is that a great deal of space can be saved, which makes it worthwhile. Some studies have found stemming to improve retrieval effectiveness, or to behave about the same as truncation. In addition, a good implementation of the Porter stemming algorithm is given, along with code. Note, however, that studying the code is not necessary - only the major phases of the algorithm and key data structures should be studied.

• p. 132, Figure 8.1, don't worry about things not really covered in this chapter such as the right 3 methods
• p. 132, middle, note that at Virginia Tech we are studying morphological analysis as an alternative - there is still an open question regarding its value
• p. 136, near top, note that ``12'' might need to be scaled
• p. 141, bottom: study question: why do we need m greater than zero?
• p. 143, after first paragraph - ignore through end of the ``One of ...'' par. on page 144