NLP

NLP

Learn how machines understand human language. From basic text cleaning to advanced Large Language Models, test your NLP skills today.


  1. What is the primary goal of Natural Language Processing?

    • To make computers faster
    • To enable machines to understand and process human language
    • To build better hardware
    • To create computer viruses
  2. The process of breaking a sentence into individual words is called:

    • Parsing
    • Tokenization
    • Stemming
    • Filtering
  3. Which technique removes common words like 'is', 'the', and 'a'?

    • Lemmatization
    • Stop Word Removal
    • POS Tagging
    • NER
  4. Reducing 'running' to 'run' using a crude heuristic is known as:

    • Stemming
    • Lemmatization
    • Tokenization
    • Vectorization
  5. What is a collection of text documents used for analysis called?

    • Database
    • Corpus
    • Dictionary
    • Frame
  6. What is the main advantage of Word2Vec over Bag of Words?

    • It is faster to train
    • It captures semantic relationships and meaning
    • It uses less memory
    • It only works on English
  7. In Word2Vec, the 'Skip-gram' model predicts:

    • Target word from context
    • Context words from a target word
    • The next sentence
    • The sentiment
  8. Which similarity measure is most effective for comparing word vectors?

    • Euclidean Distance
    • Cosine Similarity
    • Jaccard Index
    • Manhattan Distance
  9. Word Sense Disambiguation' is used to:

    • Find typos
    • Determine the correct meaning of a word in context
    • Count syllables
    • Translate words
  10. What does 'BLEU' score evaluate?

    • Image quality
    • Machine Translation quality
    • CPU performance
    • Database speed
  11. Which mechanism allows a Transformer to focus on different parts of a sentence?

    • Backpropagation
    • Self-Attention
    • Recursion
    • Convolution
  12. What is the core architecture in the 'Attention is All You Need' paper?

    • RNN
    • CNN
    • Transformer
    • LSTM
  13. BERT is an acronym for:

    • Binary Encoder Representation
    • Bidirectional Encoder Representations from Transformers
    • Basic Entity Recognition Tool
    • Binary Entity Relation Task
  14. What is 'Masked Language Modeling' (MLM)?

    • Hiding user identities
    • Predicting words that are hidden in a sentence
    • Translating hidden text
    • Removing noise
  15. In a Transformer, 'Positional Encoding' is needed because:

    • It stores the file path
    • The model has no inherent sense of word order
    • It encrypts data
    • It compresses the model