- Draw a tree structure for query (A or (B and (C or
D) and (E or F)) or G).
- The extended Boolean schemes discussed all:
- a)
- can be used to obtain a ranking of documents.
- b)
- return a similarity in the range [0,1].
- c)
- return a similarity that considers the document
weight of each query term.
- d)
- all of the above.
- e)
- a and b.
- f)
- b and c.
- g)
- a and c.
- h)
- none of the above.
- In FAST-INV, the reason for multiple loads is to save on
requirements for:
- a)
- computation time.
- b)
- primary memory.
- c)
- secondary memory.
- d)
- all of the above.
- e)
- none of the above.
Briefly explain your answer.
- What type of query (e.g., Boolean, ...)
is likely to be used when
the inverted file stores the document numbers and
weights for each term? What type of (more precise) query
would be possible if it stored the precise location
of each term occurrence?
- For implementing Boolean operations, hashing is slower
than methods using bit vectors, but has what two main
advantages over the bit vector scheme?