Ask Patents is a question and answer site for people interested in improving and participating in the patent system. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

AN OVERBROAD PATENT ON Q&A SERVICES - This application from Nuance Communications, Inc. seeks to patent looking up a corpus of previously asked questions in a Q&A service like Stack Exchange! 10 minutes of your time can help narrow this US patent application before it becomes a patent.

QUESTION - Have you seen anything that was published before Dec 6, 2011 that discusses:

A method for operating a frequently asked questions (FAQ)-based system, the method comprising: training a classifier based on questions stored in the FAQ-based system, each question corresponding to at least one answer; and applying the classifier to a query to return an answer.

If you've ever seen anything like this before, please submit evidence of prior art as an answer to this question -- one piece of prior art per answer. We welcome multiple answers from a single individual.


Summary: searches a user generated query in a classifier trained to search a database of questions in a set of FAQs

  • Publication Number: 20130144890
  • Assignee: Nuance Communications, Inc.
  • Prior Art Date: Seeking prior Art predating Dec 6, 2011
  • Open for Challenge at USPTO: Open through Dec 06, 2013

Claim 1 requires each and every step below:

A method for operating a frequently asked questions (FAQ)-based system, the method comprising:

  1. training a classifier based on questions stored in the FAQ-based system, each question corresponding to at least one answer; and

  2. applying the classifier to a query to return an answer.

Good prior art would be evidence of a system that did each and every one of these steps prior to Dec 6, 2011.

What is good prior art? Please see our FAQ.

Want to help? Please vote or comment on submissions below. We welcome you to post your own request for prior art on other questionable US Patent Applications.

share|improve this question

One possibility is Googles Auto Suggest feature.

If you think of the Google Search Box as the place where a user asks a question ("What are you searching for"?)
and the set of searchs that people type into the Google Search Box as the set of Frequently Asked Questions....

With Google Suggest, as you are typing your question it comes back with a list of possible suggestions.... To come up with that list Google applies its own algorithm to pick the list of suggestions it offers from its list of frequently asked questions

Its a bit 'meta' as Google is in the search business... but I think it describes the process...

Heres a blog from 2008

and from

How autocomplete works

  • Where the predictions come from

As you type, autocomplete predicts and displays queries to choose from. The search queries that you see as part of autocomplete are a reflection of the search activity of all web users and the content of web pages indexed by Google. If you're signed in to your Google Account and have Web History enabled, you might also see search queries from relevant searches that you've done in the past. In addition, Google+ profiles can sometimes appear in autocomplete when you search for a person's name. Apart from the Google+ profiles that may appear, all of the predicted queries that are shown in the drop-down list have been typed previously by Google users or appear on the web.

For certain queries, Google will show separate predictions for just the last few words. Below the word that you're typing in the search box, you'll see a smaller drop-down list containing predictions based only on the last words of your query. While each prediction shown in the drop-down list has been typed before by Google users or appears on the web, the combination of your primary text along with the completion may be unique.

  • Does Google monitor the autocomplete results?

Autocomplete predictions are algorithmically determined based on a number of factors (including popularity of search terms) without any human intervention. Just like the web, the search queries presented may include silly or strange or surprising terms and phrases. While we always strive to reflect the diversity of content on the web (some good, some objectionable), we also apply a narrow set of removal policies for pornography, violence, hate speech, and terms that are frequently used to find content that infringes copyrights.

So from this Google keeps a set of FAQ's... and its clearly adaptive/being trained by new questions being asked and its used to deliver answers....

share|improve this answer

The biggest example I can think of, ironically, is the StackExchange network itself. When asking a question, SE sites automatically suggest similar questions which have answers. The patent seems to cover simply collecting the answer(s) to the similar questions and showing them instead.

share|improve this answer
Does the SE system use a "classifier" as required by the claims? – George White Sep 9 '13 at 0:26
One would assume so - I don't see how else they can do otherwise. Questions are just text sentences so finding similar ones needs some kind of classifier - it's not something that's efficient or possible to do with a database query for example. – Sudhir Jonathan Sep 9 '13 at 13:57

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.