1. Field of the Invention
The present invention relates to a system and method of providing a natural language interface to a database such as the Internet.
2. Introduction
Recently, a large number of web based search engines, such as Yahoo® and Google®, have become widely available making it easier to find information over the Internet. The operation of these search engines is shown in
These search engines came to organize the billions of pages of information on the WWW in such a way as to make them accessible and useful for those who submit queries into the search engine input field. The key technical innovation in the search engines lies in the ability to index all words and phrases in the WWW and use an efficient information retrieval algorithm along with many computers to be able to instantly find information per the user's request.
Finding information about a subject is as simple as typing keywords into an input field. This is illustrated in
Users typically Type keywords such as Hurricane-speed and then manually search through the relevant documents for the answer to the original question. Finding the answer typically requires the user to search manually through each page and look for the appropriate answer. When an answer cannot be found, users would normally provide alternative words in the hope of getting a different ranking of Website.
Another search engine example will further illustrate the problem of the types of responses delivered from the search query. This example relates to a user who desires to find a specific type of information on a technical computer question or error message. On the WWW, many websites have developed to provide information technology (IT) solutions to specific error messages. These websites house a growing database of content generated by users submitting questions and a technical community providing answers. In this manner, threads of information develop on specific IT problems. The websites then may present a large number of these threads, where each thread includes a particular question or technical issue and the various responses and attempted solutions to the problem. Therefore, when others receive the same error message, they can find the solution to the problem.
For example, a user may have the Microsoft® Word application crash with the error message “Winword.exe requested the runtime termination in an unusual way”. The user may submit keywords from this error message into the Google search engine and get a response with a number of webpages that contain multiple threads dealing with various Microsoft Word problems. The user is then left to search multiple threads of information to find the particular thread that relates to the particular error message. In some cases, there may be hundreds of threads with no mechanism to find the specific answer. Furthermore, access to the threads may be obtained only via a password and subscription to the website. The user can get to the certain website wherein the answer to the query is found but then be faced with multiple interactive threads in which one of the threads holds the answer to the question. But the user cannot identify which is the appropriate thread within the website that contains the answer to the question.
This illustrates a basic problem with search engines in that they do not provide specific answers to user's questions and are based on non-natural language keyword searching. In this regard the standard search engines in some cases are not useful because they only identify Websites containing many links to other webpages and may not provide controlled access to a deeper layer of information which may hold a specific answer to the user's question.
Accordingly, what is needed in the art is a system and method of enabling a user to query a data base such as the WWW and obtain a precise answer to a question.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The features and advantages of the invention may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. These and other features of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth herein.
A system, method and tangible computer-readable media for providing a natural language interface to a database or the Internet are disclosed. The method provides a response from a database to a natural language query. The method comprises receiving a user query, extracting key data from the user query, submitting the extracted key data to a data base search engine to retrieve a top n pages from the data base, processing of the top n pages through a natural language dialog engine and providing a response based on processing the top n pages.
In order to describe the manner in which the above-recited and other advantages and features of the invention can be obtained, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:
Various embodiments of the invention are discussed in detail below. While specific implementations are discussed, it should be understood that this is done for illustration purposes only. A person skilled in the relevant art will recognize that other components and configurations may be used without parting from the spirit and scope of the invention.
The present invention relates to systems, methods and computer-readable media for providing a natural language dialog to a database such as the WWW, a corporate database, a private database or any other database. As can be appreciated, the primary benefit of the invention will be to search the WWW but there is no restriction unless specifically claimed to that as the searched database.
As shown in
The invention provides for a natural language dialog approach to searching a database. In one aspect of the invention, the computing device includes a microphone 306 for receiving voice input from the user. An automatic speech recognition (ASR) engine or module 308 converts the speech into text 310. While the ASR engine is shown as part of the computing device 302, there is no restriction on where any particular hardware or software component resides. For example, the microphone may be separate from the computing device 302 and the ASR module may run on the dialog server 304 instead of the compute device 302. Those of skill in the art will understand the variations that are possible and contemplated while maintaining the core principles of the invention. It is preferably that a large vocabulary speech recognizer be used as the front-end to the user's request.
The text is transmitted to a dialog server 304 that extracts key data from the user query. Various modules may be programmed to perform the steps set forth herein for practicing the invention. Those of skill in the art will understand the programming language and means for creating these various modules. Reference is also made to
Once the key data is extracted from the user query, the data (words and/or phrases) are used as input to at least one search engine such as Google (406). In one aspect of the invention, the data is submitted to a plurality of search engines simultaneously or in a particular order or a random order. The results from the search engine search (or search from multiple search engines) identify a top n pages.
The top n pages may be organized or pre-processed by the server 304 to prepare the n pages for processing through a natural language dialog engine (408). One example of such preprocessing would be to automatically manage entry of a user's login name and password to entry fields associated with a website (such as the IT help website discussed above) which enable the system to process the pages viewable by registration only or where any kind of security, restrictions or access steps need to be taken for viewing. This may be an entry based on a previous manual registration by the user or may be an automatic negotiation between the server 304 and the particular website unknown to the end user or as part of a subscription plan in which the user may be automatically charged a certain fee to obtain the one-time access to the data in the website. Websites with restricted or registration-only access may need to be modified to enable negotiation based on a pre-registration or dynamically at the time of the search and response. In other words, some of the top n pages may require some password or other steps necessary to view and gain access to that data for processing and preparing the response.
A dialog engine as used in the invention may be based on, for example, question and answer technology and is able to process the pages and provide a natural language response to the question asked. Those of skill in the art are aware of such engines in that they are available from such companies as IBM and AT&T Corp. The natural language response may be based, for example, on particular text found within the top n pages. Taking the IT support example above where the user is presented with hundreds of threads of conversations, a benefit of processing the webpages and links from those webpages as well is that a natural language result can be presented to the user without the need for the user to cull through so many threads to find their answer.
From the dialog engine processing, a natural language response is generated (410). At this point, there are a number of technologies for presenting that response to the user. For example, the response may be generated as text in which case a text-to-speech module running on any computing device within the architecture can “speak” the answer to the user. Or, the user may be presented with the text of the answer in a browser with further organized options to click on the underlying pages where the information was retrieved or to view other websites and so forth.
In the hurricane example, the user may be presented with the answer either as text or audibly: “The top speed of Hurricane Ivan was 155 miles per hour, more information may be found at this link.”
The invention may operate in both a text-based mode and a speech-based mode. It is further contemplated that a multi-media presentation may be used to enhance the interaction. For example, a virtual agent may be the interactive means by which the user speaks to provide the query and from which the response received. In this regard, an aspect of the invention is that it can operate in a dialog mode or an instant messaging mode where the user can chat with a virtual agent to get information or to provide subsequent relates questions on the same topic. The invention improves the user's ability to receive a specific response to a natural language query that is not possible from the known search engine approach that presents a plurality of webpages that may or may not contain the answer to the user's query.
In another aspect of the invention, a machine learning algorithm continues to find related documents from the database related to the input query and can expand the search accordingly. For example, after the system response is provided, the machine learning process can continue to find related information and can then present the user with an option to review the related information separate from the first provided response. The option to search through these additional documents may be presented to the user via a graphical user interface such as the browser or through a virtual agent.
Embodiments within the scope of the present invention may also include non-transitory computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer. Computer-readable media comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures. When information is transferred or provided over a network or another non-transitory communications connection to a computer, the computer properly views the connection as a computer-readable medium. Thus, any such connection is properly termed a computer-readable medium. Combinations of the above should also be included within the scope of the computer-readable media.
Computer-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Computer-executable instructions also include program modules that are executed by computers in stand-alone or network environments. Generally, program modules include routines, programs, objects, components, and data structures, etc. that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps.
Those of skill in the art will appreciate that other embodiments of the invention may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, minicomputers, mainframe computers, and the like. Embodiments may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a combination thereof) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.
Although the above description may contain specific details, they should not be construed as limiting the claims in any way. Other configurations of the described embodiments of the invention are part of the scope of this invention. Accordingly, the appended claims and their legal equivalents should only define the invention, rather than any specific examples given.
The present application is a continuation of U.S. patent application Ser. No. 11/013,836, filed Dec. 16, 2004, the content of which is included herewith in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
5933822 | Braden-Harder et al. | Aug 1999 | A |
6269361 | Davis et al. | Jul 2001 | B1 |
6304864 | Liddy et al. | Oct 2001 | B1 |
6434524 | Weber | Aug 2002 | B1 |
6499013 | Weber | Dec 2002 | B1 |
6532444 | Weber | Mar 2003 | B1 |
6553347 | Tavor et al. | Apr 2003 | B1 |
6598047 | Russell et al. | Jul 2003 | B1 |
6633864 | Christensen et al. | Oct 2003 | B1 |
6658093 | Langseth et al. | Dec 2003 | B1 |
6665640 | Bennett et al. | Dec 2003 | B1 |
6691111 | Lazaridis et al. | Feb 2004 | B2 |
6708162 | Morgan et al. | Mar 2004 | B1 |
6725425 | Rajan et al. | Apr 2004 | B1 |
6765997 | Zirngibl et al. | Jul 2004 | B1 |
6792086 | Saylor et al. | Sep 2004 | B1 |
6798867 | Zirngibl et al. | Sep 2004 | B1 |
6802042 | Rangan et al. | Oct 2004 | B2 |
6819339 | Dowling | Nov 2004 | B1 |
6829603 | Wolf et al. | Dec 2004 | B1 |
7107536 | Dowling | Sep 2006 | B1 |
7725307 | Bennett | May 2010 | B2 |
7844594 | Holt et al. | Nov 2010 | B1 |
7899802 | Sailor et al. | Mar 2011 | B2 |
20020077815 | Zhang et al. | Jun 2002 | A1 |
20030028792 | Plow et al. | Feb 2003 | A1 |
20030065739 | Shnier | Apr 2003 | A1 |
20030187925 | Inala et al. | Oct 2003 | A1 |
20040073555 | Hevener | Apr 2004 | A1 |
20040199541 | Goldberg et al. | Oct 2004 | A1 |
20050033968 | Dupouy et al. | Feb 2005 | A1 |
20050114325 | Liu et al. | May 2005 | A1 |
20050154723 | Liang | Jul 2005 | A1 |
20060031214 | Solaro et al. | Feb 2006 | A1 |
20060100969 | Wang et al. | May 2006 | A1 |
20090307577 | Lee | Dec 2009 | A1 |
Entry |
---|
Feldman, Susan. NLP Meets the Jabberwocky: Natural Language Processing in Information Retrieval, Online May 1999, Copyright © Information Today, Inc. |
Chai, Joyce et al. “Natural Language Sales Assistant—A Web-based Dialog System for Online Sales,”, IBM T. J. Watson Research Center, Hawthorne, NY, 2000. |
Chai, Joyce et al. “Comparative Evaluation of a Natural Language Dialog Based System and a Menu Driven System for Information Access: a Case Study”. International Conference on Multimedia Information Retrieval (RIAO 2000), Apr. 2000, Paris, France. |
González-Bernal, J.A. et al. “Natural Language Dialog System for Information Retrieval”. National Institute of Astrophysics, Optics and Electronics. Puebla, Mexico. |
Carbonell, Jaime G. “Is There Natural Language after Data Bases?” Computer Science Department, Carnegie-Mellon University, Pittsburg, Pennsylvania. |
Number | Date | Country | |
---|---|---|---|
20110179006 A1 | Jul 2011 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 11013836 | Dec 2004 | US |
Child | 13074419 | US |