Loading your learn pack…
Read the summary, then start training!
QA systems aim to answer natural-language questions by using either structured knowledge bases or unstructured document collections.
Question answering (QA) is a computer science discipline within information retrieval and natural language processing (NLP) focused on building systems that automatically answer questions posed by humans in natural language. A QA system may generate answers by querying structured knowledge bases (e.g., curated databases) or, more commonly, by extracting answers from unstructured collections of natural language documents such as reference texts, news articles, Wikipedia, and other web pages. The scope of QA includes answering many kinds of questions (e.g., fact, list, definition, how/why, hypothetical, and cross-lingual) and supporting different interaction and reasoning needs. QA can be implemented in various settings such as closed-book (memorized facts without explicit context), closed-domain (restricted to a specific subject area), and open-domain (answers about nearly anything using general knowledge and large document collections). Modern QA research and systems also extend to specialized domains like mathematics, temporal/geospatial questions, definitions/terminology, biographies, and multimodal content (e.g., audio, images, and video), often using architectures such as retriever-reader pipelines or end-to-end transformer-based models.
QA systems aim to answer natural-language questions by using either structured knowledge bases or unstructured document collections.
QA scope spans multiple question types and settings, including closed-book, closed-domain, and open-domain QA, as well as specialized and multimodal domains.
A field of computer science that builds systems to automatically answer human questions expressed in natural language.
A structured repository of information that QA systems can query to construct answers.
QA that answers questions about nearly anything, typically relying on general knowledge and large document collections.
QA restricted to a specific subject area, often leveraging domain-specific knowledge such as ontologies.
QA where the system answers from facts learned during training without being given explicit external context at inference time.
A QA architecture where a retriever finds relevant documents and a reader infers the answer from those documents.
“Can you explain what "QA systems aim to answer natural-language questions by using either structured knowledge bases or unstructured document collections." means in simple terms?”