8 of the 41 questions in this set, with the correct answer marked and every option explained.
1. Which of the following techniques can be used for keyword normalization in NLP, the process of converting a keyword into its base form?
✗Soundex
A phonetic algorithm that encodes words by how they sound, so that Smith and Smyth collide. Useful for fuzzy name matching, but it produces a phonetic code rather than a dictionary base form.
✗Cosine Similarity
Measures the angle between two vectors to judge how alike they are. A comparison metric, not a transformation applied to a single word.
✓Lemmatization
Lemmatization helps to get to the base form of a word, e.g. are playing -> play, eating -> eat, etc.Other options are meant for different purposes.
✗N-grams
Contiguous sequences of n items from a text. A way of representing context, which increases the feature space rather than collapsing words to a base form.
2. Which of the following techniques can be used to compute the distance between two word vectors in NLP?
✓Euclidean distance
Distance between two word vectors can be computed using Cosine similarity and Euclidean Distance. Cosine Similarity establishes a cosine angle between the vector of two words. A cosine angle close to each other between two word vectors indicates the words are similar and vice a versa. E.g. cosine angle between two words "Football" and "Cricket" will be closer to 1 as compared to angle between the words "Football" and "Los Angeles"
✗N-grams
A text representation technique producing sequences of adjacent tokens. It creates features; it does not measure the distance between two vectors.
✗Lemmatization
Reduces a word to its dictionary form. A preprocessing step performed before vectorisation, not a measure of distance between vectors.
✓Cosine Similarity
3. What are the possible features of a text corpus in NLP?
✓Basic Dependency Grammar
All of the above can be used as features of the text corpus.
✓Part of Speech Tag
✓Vector notation of the word
✓Count of the word in a document
4. In NLP, Words represented as vectors are called as Neural Word Embeddings
✓True
Word2Vec, GloVe based models build word embedding vectors that are multidimensional.
✗False
Representing words as dense vectors learned by a neural network is exactly what neural word embeddings are — Word2Vec, GloVe and fastText all produce them.
5. Which one of the following are keyword Normalization techniques in NLP
✗Part of Speech
Tags each word with its grammatical role. It annotates words rather than reducing them to a common base form.
✓Lemmatization
Part of Speech (POS) and Named Entity Recognition(NER) are not keyword Normalization techniques. Named Entity help you extract Organization, Time, Date, City, etc..type of entities from the given sentence, whereas Part of Speech helps you extract Noun, Verb, Pronoun, adjective, etc..from the given sentence tokens.
✓Stemming
✗Named entity recognition
Identifies and classifies proper nouns such as people, organisations and places. An extraction task, not normalisation.
6. Which of the below are NLP use cases?
✗Detecting objects from an image
Computer vision. The input is pixels, not language.
✗Facial Recognition
Also computer vision — matching facial features against known identities.
✗Speech Biometric
Identifies a speaker from voice characteristics such as pitch and timbre. It analyses the acoustic signal rather than the meaning of the words, which places it in speech processing rather than NLP.
✓Text Summarization
Text Summarization is an NLP use case.
7. In a corpus of N documents, one randomly chosen document contains a total of T terms and the term hello appears K times. What is the correct value for the product of TF (term frequency) and IDF (inverse-document frequency), if the term hello appears in approximately one-third of the total documents?
✓K * Log(3) / T
formula for TF is K/T - formula for IDF is log(total docs / no of docs containing "data") = log(3) . Hence correct choice is Klog(3)/T
✗T * Log(3) / K
Inverts the term frequency. TF is the count of the term divided by the total terms, K/T, not T/K.
✗Log(3) / KT
Divides by both K and T. TF should multiply by K, so K belongs in the numerator.
✗KT * Log(3)
Multiplies K by T instead of dividing. Term frequency normalises by document length, so T must be the denominator: TF = K/T, IDF = log(N ÷ N/3) = log(3), giving K·log(3)/T.
8. In NLP, The process of removing words like "and", "is", "a", "an", "the" from a sentence is called as
✗Lemmatization
Reduces words to their dictionary form — 'was' becomes 'be'. It transforms words rather than deleting them.
✓Stop word
In Lemmatization, all the stop words such as a, an, the, etc.. are removed. One can also define custom stop words for removal.
✗Stemming
Crudely strips suffixes, turning 'running' into 'run'. Again a transformation, not a removal.
✗All of the above
Only stop word removal deletes words. Lemmatisation and stemming both keep every token and merely rewrite it.
33 more questions in the app
Practise the full 41-question set with a timer, scoring and progress tracking.