Information Retrieval

Question 1

What is vector space model ?

Answer 1

represent each document as a vector

Question 2

What is Dimensionality reduction ?

Answer 2

removing stop-words and very rare words, and selecting only the most distinctive terms, which is a proces known as feature selection

Question 3

What is Latent Semantic Analysis ?

Answer 3

A method for term-by-document dimensionality reduction

Question 4

Precision

Answer 4

percentage of true positives out of all returned documents

Question 5

Recall

Answer 5

percentage of true positives out of all relevant documents in the collection

Question 6

F-measure

Answer 6

weighted harmonic mean between Precision and Recall

f = 2PR/(P+R)

Question 7

what are the Types of Text Classification ? give examples

Answer 7

topic categorization
sentiment classification
authorship attribution and plagiarism detection
spam detection and e-mail classification

Question 8

Evaluation Measures for Text Classification ?

Answer 8

Macro-averaged precision, recall, and F-measure

micro-averaged precision, recall, and F-measure

Question 9

Similarity-based Text Classification ?

Answer 9

we use cosine similarity and Euclidean distance

Question 10

What is the Common N-gram(CNG) method ?

Answer 10

• creating n-gram author profile

- use Euclidean distance similarity measure