People do type questions into search engine search boxes and expect meaningful answers.
The best search results aren’t always found based upon matching the words in the questions to pages that might also contain those questions or similar questions. What strategies might be used by the search engines to provide a good user experience? Could they try to Crawling FAQs to get answers?
A paper from Google researchers on Question Answering, Statistical Machine Translation for Query Expansion in Answer Retrieval (pdf), looks at automated strategies for understanding and answering questions that people might type into a search box.
I’ve been finding answers on how to do things on the web for years, by looking for frequently asked questions (FAQs) pages, and tutorials. Usually those searches involve using those words (FAQ or tutorial) in my queries, along with a couple of words related to what I want to learn how to do.
For example, to join words together from one box in excel to another, I might type into Google the following:
concatenate excel tutorial
Less frequently, I’ll type in an actual question into a search engine box, to try to find an answer, but it seems that people do ask questions of search engines.
Can a search engine use the strategy that I employ – looking at FAQ pages by crawling FAQs, and use those frequently asked questions to try to understand relationships between words in questions, and words in answers on those pages, to understand queries in question form, and try to provide answers to those questions?
The paper explores the use of a database of FAQ pages, which they found through “intitle:faq” and “inurl:faq” queries in the search engine, to come up with question and answer pairs. The processes that they then use on those question and answer pairs, to make them useful for Question Answering are definitely worth a look.
When you write a set of questions for a Frequently Asked Questions page, how much effort do you take in trying to craft those questions, and in choosing the words used within them? If you thought a search engine might be crawling FAQ pages, would you make more of an effort?
Some posts I’ve written about patents involving question answering:
- 7/19/2007 – Search Engines Crawling FAQs to Learn How to Answer Questions?
- 9/21/2014 – Google May Use Question Answering to Populate the Knowledge Graph
- 10/12/2014 – How Google May Use Entity References to Answer Questions
- 12/30/2014 – Direct Answers – Taken from Authority Websites
- 12/31/2014 – Direct Answers – Using Query Intent Templates to Identify Answers
- 2/11/2015 – How Google was Corroborating Facts for Direct Answers
- 7/12/2015 – How Google May Answer Questions in Queries with Rich Content Results
- 9/9/2015 – When Google Started Showing Direct Answers
- 11/30/2016 – Answering Featured Snippets Timely, Using Sentence Compression on News
- 6/19/2017 – Google Extracts Facts from the Web to Provide Fact Answers
- 7/10/2019 – How Google May Handle Question Answering when Facts are Missing
Last Updated July 11, 2019.