Case Study

Design the inputs and outputs for a document-based support question system.

Case context: Your team is building a system for a municipal utility office. Staff members need to search internal policy documents and answer citizen questions quickly.

Question: Using the basic ideas behind end-to-end question answering, decide what information must be supplied to the system as input and what form the system should produce as output.

Sample answer: The system should receive two linked inputs: the reference document text and the user’s question about that document. Together, these form a text-question pair. The system’s output should be the answer text itself, written to directly address the question using information from the document.

Key points:

  • Input consists of a text and a question.
  • The text is the source document.
  • The question is the query to answer.
  • Output is the answer text generated by the system.

Rubric: Award credit if the learner correctly identifies the input as a document plus question pair and the output as the generated answer text.

0

1

Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI