DM week 7

?
  • Created by: mariolarc
  • Created on: 11-05-21 15:38

1. What is the advantage of probabilistic models with respect to deterministic models for information extraction?

  • Probabilistic models are Bayesian, i.e. with high accuracy.
  • Determining models are computationally more intensive than probabilistic models.
  • Deterministic models must match expressions exactly, whereas probabilistic models are more flexible.
  • Determining models work only with characters, while probabilistic models can also handle words.
1 of 5

Other questions in this quiz

2. Which of the following is false in the context of information retrieval?

  • A precision-recall curve shows how the performance of an information retrieval approach is affected as we retrieve more and more documents relevant to a query.
  • The false positives (FP) are the non-retrieved documents, that are relevant to a query.
  • Information retrieval is similar to classification.
  • The IDF score ensures that we do not give much importance to terms that appear in too many documents.

3. Which of the following does not generally reduce the size of the term-document matrix?

  • Word embeddings.
  • Stemming
  • Tokenization
  • Inverted file (linked list) data structure.

4. Which of the following statements about n-gram models is false?

  • They can be used for text classification.
  • They are probabilistic models.
  • As n increases, they tend not to predict well.
  • Can be chosen with cross-validation.

5. Given the string s='abcccba', which of the following statements is true?

  • 'ab(c)*ba' is a regular expression for s.
  • All the other choices are correct.
  • 'abxba' is a regular expression for s.
  • 'abcba' is a sub-string of s.

Comments

No comments have yet been made

Similar Computing resources:

See all Computing resources »See all DM week 7 resources »