Text Analytics DATA-SCIENCE-FOUNDATIONS Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What is the primary difference between text classification and topic modeling?

   • AText classification extracts named entities, while topic modeling extracts sentiment
   • BText classification works on structured data, while topic modeling works on unstructured text
   • CText classification is unsupervised, while topic modeling is supervised
   • DText classification uses predefined categories, while topic modeling discovers categories from the data
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which text preprocessing technique reduces words to their base or root form by removing prefixes and suffixes without considering the word's meaning?

   • AStemming
   • BTokenization
   • CLemmatization
   • DStop-word removal
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A news organization has a large archive of articles and wants to automatically discover the main themes covered over the past decade. They do not have predefined categories. Which technique should they use to identify latent themes in the corpus?

   • AA supervised text classification model trained on labeled news categories
   • BLatent Dirichlet Allocation (LDA)
   • CNamed Entity Recognition (NER)
   • DSentiment analysis with a lexicon
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   An e-commerce company wants to automatically route customer emails to the correct department (billing, technical support, or sales). They have a large labeled dataset of past emails. Which approach should they use to build this system?

   • AUse topic modeling to discover clusters of emails
   • BTrain a supervised text classification model on the labeled emails
   • CUse NER to extract product names from the emails
   • DApply sentiment analysis to determine the urgency of each email
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A social media monitoring company needs to analyze millions of tweets to determine public opinion about a new product launch. They need a fast, scalable approach that does not require labeled training data. Which approach should they use?

   • ALexicon-based sentiment analysis
   • BNamed Entity Recognition
   • CTopic modeling with LDA
   • DSupervised machine learning with a custom-trained classifier
   Reveal Solution Discussion Select an answer first