Chapter 8, Problem 13E

Explanation of Solution

Concept of frequency

• Frequency is the logarithm of number of occurrences of the term that is divided by the number of terms in the document.
• The concept of term frequency does not make sense when it comes to small databases with few attributes with each containing only a few words.
• The relevance of any term does not depend on the number of occurrences of the term.
• The logarithmic increase in the term frequency is not a good indicator when the domain is very small...

Explanation of Solution

Similarity of two tuples

• The similarity of two tuples can be measured by cosine similarity metric.
• The only difference is that the values that...