**Unformatted text preview: **(A) Increase (B) Stay the same (C) Decrease (iv) How could the dimension change if we remove a document? (A) Increase (B) Stay the same (C) Decrease (v) How could the dimension change if we add a term that is not in any of the documents? (A) Increase (B) Stay the same (C) Decrease (e) Given 100 documents, consider a term t1 that appears 2 times in document d1. What is the DocTermRank of t1 in d1 if (assume base 10 logarithm): (i) T1 appears in only d1? (ii) T1 appears in 10 documents? (iii) T1 appears in all documents? (iv) What property of TF-IDF does this pattern show? (f) Given 100 documents, consider a term t2 that appears in 10 documents. What is the Doc-TermRank of t2 in d2 if (assume base 10 logarithm): (i) T2 appears 2 times in d2? (ii) T2 appears 4 times in d2? (iii) T2 appears 6 times in d2? (iv) What property of TF-IDF does this pattern show? CS W186, Fall 2019, DIS 11 2...

View
Full Document

- Spring '14
- BRITZ,D