CN101120341A - Method and equipment for performing mobile information access using natural language - Google Patents
Method and equipment for performing mobile information access using natural language Download PDFInfo
- Publication number
- CN101120341A CN101120341A CNA2005800476958A CN200580047695A CN101120341A CN 101120341 A CN101120341 A CN 101120341A CN A2005800476958 A CNA2005800476958 A CN A2005800476958A CN 200580047695 A CN200580047695 A CN 200580047695A CN 101120341 A CN101120341 A CN 101120341A
- Authority
- CN
- China
- Prior art keywords
- answer
- message
- user
- information
- mobile
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
This patent describes a method for mobile information access, executed in a networked computer system comprising at least a mobile information access server and one or a plurality of information retrieval systems, comprising the steps of: receiving a message from a mobile communication device; analyzing the received message; forming one or a plurality of queries based on the message analysis; obtaining documents based on the one or the plurality of queries; extracting candidate answers from the documents; validating candidate answers; composing an answer summary; sending the answer summary back to the mobile communication device, wherein the answer summary is limited to a predetermined size. The patent also describes an apparatus for mobile information access.
Description
Technical field
The present invention relates to a kind of method and apparatus that uses mobile communications device to carry out mobile information access.More specifically, it relates to a kind of method and apparatus that uses small-sized mobile communications device to carry out mobile information access, and this small-sized mobile communications device has the ability of restricted reception and output access information.
Background technology
Mobile communications device is except that the actual means that are provided for individual, human communication, and it is further used is to reside in a remote information source on the machine by visit to have the ability that satisfies his or his informational needs for the user provides.
An important example is to use the mobile communications device search information.
To mobile communications device for this purpose the important restrictions of the efficient of search information be:
Because restricted channel capacity causes the transmission mass data slow;
It is restricted that being used between mobile communications device and informant sends the form that the messaging protocol of information adopts, for example 160 character limit of short-text message (SMS);
Little and the inconvenient use of the keyboard of mobile communications device when input text;
At output terminal, mobile phone or other device have only very little display usually, are used for the display message to the user.
The restriction of transmission capacity and message, keyboard and display size requires search inquiry and replys also to be subjected to size restrictions usually.About search inquiry, the user must be able to provide very short and simple and clear inquiry.About replying, answering system must can produce very simple and clear with relevant replying.
Simple and clear inquiry and reply and can realize by the particular range system, wherein the particular range system only allow very restricted, defined the inquiry of scope, and provide visit with the predetermined structure data.This is enough in upper/lower positions/scope, and wherein possible inquiry group is known, and inquiry has known structure.If inquiry is known, searching in cache memory or special database engine enough is used for retrieving correct with relevant replying usually so.An example of this method is the inquiry train time schedule, can and be decomposed into a simple template (starting point, destination, time) its identification, and can retrieve information again exactly from structured database.
The particular range information retrieval system is open below: Gallwitz, F., M.Aretoulaki, M.Boros, J.haas, S.Harbeck, R.Huber, H.Niemann, with E.N th " Erlangen speech conversational system EVAR: state-of-the-art information retrieval system " (in international symposium in 1998 about in the minutes of speech dialogue (delivering in 98 years), the 19-26 page or leaf, Australia, Sydney, in November, 1998, yellow-study east, " the spoken processing: theoretical direction; algorithm and system development " of Alex Acero and Hsiao-Wuen Hon (2001), Prentice Hall PTR and Young, S. (2002) " statistical method of speech conversational system design " in the technical report CUED/F-INFENG/TR.433 of Cambridge University's engineering department.
Yet domain-specific approach is not really flexible aspect manageable inquiry usually.In addition, the high and very limited scope of its common only covering of the cost of realization.
In other words, and if especially the inquiry scope (domain) be not known, can use second method so.This is range of opening or more generally be to mix (particular range with range of opening) method normally: have inquiry and when understanding seldom to the knowledge of its scope, it is still attempted by relevant the replying of analysis result retrieval.
Recently, this second method has obtained more attentions, because a large amount of free informations that exist with the form of hypertext document are arranged on the Internet.In that, can also can conduct interviews by any search engine of global network system (WWW) visit by the mobile phone that uses WAP (radio access agreement). Mobile ' retrieval service (http://mobile.yahoo.com/search).
Yet, if input and output are not to be specially adapted to portable environment, so with very inconvenient use.In portable space search information current be two step process: the user at first must find out him or she where can the acquired information and second step, goes there and satisfy correct needs information.
On desk-top computer, the first step is used usually based on the key word of the Internet search engine (as Google or Yahoo) and is carried out, and it returns a row hyperlink address, wherein can find information itself.Because can show numerous results, so the big molded dimension of desktop computer screen is feasible efficient on desk-top computer based on the retrieval of key word.Even when not all result was appropriate, this processing was still worked.User's appropriate website on the surface of manually clicking the mouse, and if the website as if comprise the information of being sought, they browse to find it, because it comprises numerous further steps thereby very complicated so.
On the other hand, in the scheme of a movable type, mouse and screen are not littler owing to having, and therefore navigation is difficult more, and it needs more manual navigation step (for example roll, skip or the like).Therefore, it is not enough only imitating desktop mechanism on mobile communications device.
Therefore, when visit during based on the inquiry of webpage-answering system, for example during search engine, must adjust inquire about and answer-back unit being more suitable for mobile subscriber's needs.
J.-D.Ruvini " on midget plant, being adapted to user's retrieve strategy " (: described a kind of method the 284-286 page or leaf about intelligent user interface in the 8th boundary's international conference on information processing that the Miami of Florida, US in 2003 holds), it shows that to the Google search engine front end provides network browsing for mobile phone.
Another kind method is the GoogleSMS retrieval (http://www.***.com/sms) based on key word by SMS.
Here, coverage is big than in the top-down method usually, yet, because the not structuring character of bottom data, be difficult to realize relevant and reply accurately.As a result, may must send some short messages, need on mobile communications device, improve memory capacity, and user's inconvenience is read to guarantee to comprise appropriate replying.Yet, even like that, be not naturally can receive relevant replying yet.
Summary of the invention
A target of the present invention is adaptive composition of replying, and make it satisfy the resource limit of current mobile device, and the high correlation that keeps simultaneously/obtain and reply promptly guarantees to comprise the high probability at the correct option of problem.
Another target of the present invention is the availability that increases portable inquiry-answering system.
According to the present invention, by being used for the method for mobile information access and, realizing these targets by a kind of equipment that is used for mobile information access according to claim 15 according to a kind of of claim 1 independently.In appended claims, define preferred embodiment.
By providing one will inquire about as the interface of natural language problem or language phrase and use language tool to analyze them, improve the correlativity of result for retrieval, and therefore can correspondingly reduce the size of replying, making to the user provides an appropriate answer becomes possibility, and no matter the resource limit of his mobile communications device.
By a user's profile is provided in addition, mobile communications device or user are known to the system or identification by identification number, and being used to rise to specific user further provides the relevance of replying automatically, particularly because the intrinsic knowledge of device parameter.In addition, by means of the information and the his or her mobile communications device that use about the user, user's profile is also guaranteed reliable user experience, do not need to enter again this profile information, and come the number (step 650) of the candidate answer of considering is tied to one group that more may meet user's request by means of the advantage of using this previous context knowledge.
Provide unified according to the interface of the formation natural language problem of this claim to structure and the not visit of structured data sources.
When describing in detail below the reference accompanying drawing is read, further characteristic and advantage will obviously be changed.
Description of drawings
Fig. 1 is the synoptic diagram of an example system, wherein according to the mobile information access method of carrying out on the mobile information access server of the present invention.
Fig. 2 shows the synoptic diagram according to the embodiment of mobile information access method of the present invention.
Fig. 3 shows the synoptic diagram according to a message analysis among the embodiment of mobile information access method of the present invention.
Fig. 4 is presented at the language analysis details of the problem of extracting during the analysis of the message that shows in the analysis chart 3.
Fig. 5 shows the Language Processing according to the inquiry response among the embodiment of information retrieval method of the present invention.
Fig. 6 shows another synoptic diagram according to the embodiment of mobile information access method of the present invention.
Fig. 7 shows the form of the possible user's who uses in the embodiment of the invention illustrated in fig. 2 profile contents.
Fig. 8 shows the schematic output according to mobile information access method of the present invention.
Fig. 9 shows the embodiment according to mobile information access equipment of the present invention.
Embodiment
Fig. 1 is the synoptic diagram of example system, wherein carries out according to mobile information access method of the present invention on the mobile information access server.
In Fig. 1, witness marker 100 expression mobile communications devices, for example: cell phone, smart phone, PDA(Personal Digital Assistant), wearable device or the like.
Each mobile communications device 100 is by cordless communication network 110 communications, according to the present invention, for example: telephone network or WLAN (wireless local area network), the gateway 120 with mobile information access server 160 and the Internet 140.To 140 bridge communications in the Internet, vice versa at cordless communication network 110 for gateway 120.
Mobile information access server 160 and one or more unstructured data sources 130 that unstructured data is provided, and link to each other with one or more structured data source 150 that structural data is provided.
Unstructured data sources can include but are not limited to: local index is gathered in full, intranet retrieval engines and special the Internet/web search engines.
Structural data is originated and can be included but are not limited to: (for example index resource description form (RDF) data and true simple syndication (RSS) flow for simple on-line access agreement (SOAP) network service, relational database or semi-structured XML storehouse.
Do not show among Fig. 1, but what also occur in example system is a large amount of file server that links to each other with the Internet, wherein on the mobile information access server, carry out according to the method for mobile information access of the present invention, file server for example provides the file with the form of HTML (HTML (Hypertext Markup Language)) page, by for example internet search engine 130 it is carried out index.
Fig. 2 shows the synoptic diagram according to the embodiment of mobile information access method of the present invention.
In step 200, mobile information access server 160 reception sources are from the message of mobile communications device 100.
In step 210, analyze the message that receives; Particularly from inquiry, extract problem or language phrase with natural language.The detail analysis process will be described with reference to figure 3 subsequently.Language phrase resembles ' restaurant of Edinburg ' will be taken as problem.Can distribute default problem types to them.
In step 220, construct inquiry according to the problem that from information, extracts or phrase and according to the subsequent analysis in step 210.At first, deriving a set of keyword and key phrase constitutes as the basic search engine queries.Launch these with problem types is specific with the specific key word of scope then, and consider each characteristic in the ff engine syntax.
In addition, can add the restriction operator in the described search engine query to,, and merge this theme dedicated search and general search so that it is focused on one group of theme associated documents or the network address.
In step 230, the form retrieving information of from structure and unstructured data sources, setting with the Web page, database gulp or XML. Or the structured data sources of MSN retrieval and similar database.
Because may be the file identifier that is provided by medium file search engine or list of links be regarded as the file of needs, because they usually provide relevant information together with file identifier with the form of so-called segment.Can or directly these files be made a summary thus, perhaps by the file of downloaded by the search-engine results reference, and these file in download of analysis/summary.In one embodiment of the invention, all are further handled all to go up in search engine summary segment (snippet) and carry out.In another embodiment, from the file identifier (for example hyperlink) of search engine retrieving constant number, and download related file.
After structuring never and structurized data source acquisition result, they are merged.
In step 240, use text analyzing to analyze result for retrieval so that designate candidate is replied.Use is extracted candidate answer from the information of text analyzing and the problems referred to above analysis in the file that obtains from step 230.
In step in 250, the checking candidate answer is promptly filtered and classification according to the order that the likelihood of truthlikeness/reply reduces.Arrange the likelihood that candidate answer is replied with reflection according to relevant criterion (it is better than replying b to reply a).
In step 260, answer summary is made up of n the candidate components on top, considers that restriction is output as the needs of pre-sizing.This pre-sizing may depend on the full-size of the demonstration size of output unit, single text message or each user's hobby.The number n of the candidate answer fragment that depends on pre-sizing and retrieve, wherein n surpasses minimal confidence threshold, consider and merging numerical value c=f (s, n) individual candidate answer A1, A2, AN 215, format as much as possible or with special symbol (for example separating character of a line or similar '/') separately with the formation answer summary.
Send this answer summary back to mobile unit (108).Optionally, the phonetic synthesis unit can be converted to voice to this answer summary.
Fig. 3 has shown the step 210 of carrying out message analysis more in detail.
Message analysis is used for generated query, with this inquiry inputted search engine, but this problem is referred to a wide in range known classification or problem types, and the generation key term is used for later streamline.
In step 300, determine at first whether message is initiated with spoken form by mobile communications device 100.In this case, it will be in step be subjected to automatic speech recognition (ASR) in 310.
In step 320, extraction problem or phrase from this message.In case problem or phrase separate, and will further analyze it so that can understand this problem or phrase, perhaps can on the type of replying of expectation, draw some deduction at least.
In step 330, use language issues type this problem of Model Calculation or phrase type (to seek the information of what type?).Is ' when Galileo born owing to problem? ' be the information of the time of seeking, so its to reply can not be name.The same focus (seeking the entity information-related) (being Galileo in this example) that derives this problem or phrase with which.
In step 340, question text is used for analyzing this problem from linguistics, its use one (group) comprise part of speech (POS) sign, stem, by the classification of inflection form, chunk, named entity sign, remove the language model of word meaning polysemy and place name judgement.
Fig. 4 shows the details of the language analysis of the problem of extracting during the analytical information shown in Figure 3.
Signization is decomposed into sign with problem (step 400).
Sort out the canonical form that (not shown) generates each word, for example term " are " generation " be " by the inflection form.
POS sign (step 410) is for example represented adjective term " greatly " with JJ with phraseological identifier mark sign.
The identification of named entity sign and classification proper noun, for example name or location name, date and time or the like.
Chunk is meant the identification and the classification of the synthetic words and expressions of onrecurrent, verbal phrase for example, noun phrase, proposition group.
Fig. 5 shows the Language Processing according to the result for retrieval among the embodiment of mobile information access method of the present invention.
In step 510, result for retrieval may be a standard, that is, text may must with the separated from meta-data of suitable search engine, perhaps convert plain text to from specific format (for example HTML).
In step 520, shown in Fig. 4 and foregoing description, carry out similarly and analyze, now normalized result for retrieval is described.
In step 530, the text unit of all and problem types unit compatibility (for example, " February 14 " is a date, itself and the inquiry compatibility of " when ", " and Bella " is name, itself and the problem of " who ") compatibility, verify/classification with the likelihood of this problem being replied according to them, generation is called the record of ' grade ', the language context considered in this record, by providing this language context from the result who wherein extract to answer contextual language analysis result of candidate file and language issues analytic unit.
N candidate answer with highest ranking be as the input of forming step 540 in answer summary, considers that wherein message size constraint condition and other form answer summary from the attribute that user profile table obtains.
Fig. 6 shows the further embodiment according to mobile information access method of the present invention.Below describe will concentrate on Fig. 2 on the specific differences of the method that shows.
In this embodiment, this message also comprises identifier, so that discern mobile communications device according to the message that receives, for example extracts and the telephone number of storage in step 610.
In step 620, the identifier of initiating the mobile communications device of message is used for the profile of retrieval user.
User profile table is used to consult to inquire about the knowledge whether it comprise about this mobile communications device specific properties and (includes but not limited to: the ability of display size, resolution, number of colours, graphics display capability, sound capabilities and movie), and retrieval preferred, users motif area (including but not limited to: trival matters/general knowledge, motion, film or the like, perhaps self-defined website).
In question analysis process, this information from profile is used to refine query structure, so that its deflection user favored area, and similarly make candidate answer extraction and checking deflection favored area, alternatively, use the interested priority orders of expression in advance in one group of motif area.
In step 630, consider indivedual preferences, according to the problem types of determining and the key word/key phrase of extraction, can also construct a search engine inquiry or last set engine queries.For example, the user may want to be provided with his profile, (is therefore only pointing to football website and network service) with the search that limits him during the world cup in the football scope.Perhaps, he or she may only want to represent that the interest to fashion is more preferential than Financial Information, so that do not seek replying about problem from financial web site or service.
In addition, can be according to from the special search of the motif area information and executing of user profiles table search to special website.Consider the characteristic (for example the special operator of similar "+" is to guarantee must there be some word in the page "+soccer rule-law ") of the grammer of Search engine/information retrieval engine, phrase or the key word that extracts like this or form is converted to search engine or information retrieval inquiry.
In step 650, candidate answer extraction and verification step, also consider user's user preference and like: for example, user's profile reflection reaches the user who political realms is lacked interest to the strong interest in motion field, by adding or remove the inquiry composition of indication every field, seek to be used for the file in former field of query structure retrieval (step 630) and the file in field after avoiding.Thus, promote with the formation form respectively and degradation indication sports field and political realms from contextual candidate answer.
In step 660, answer summary is formed step, and the type information relevant with model of the mobile device from be stored in user profile table self obtains limiting the preliminary dimension of its output.
Based on the above information, for the only answer summary of mobile device, the user uses caller identification to send problem to identify his or her profile record.Depend on the hobby or the learned mobile device maximum information size restrictions (for example under the situation of SMS) that obtain from user's profile s, and the number n that candidate search is replied segment surpasses a minimal confidence threshold, numerical value c=f (s, n) individual candidate answer A1, A2, AN 215 forms module 216 by summary and considers and merge, and format as much as possible or with special symbol (for example line or separating character are as '/') separately is with formation answer summary 217.
In addition, the attribute that remains on the user's mobile communications device in user's the profile can be used to form answer summary, use mobile communications device ability summary to create: for example, in the possible embodiment of the present invention, if this mobile subscriber's mobile communications device possesses colored demonstration, the pith of answer summary (for example title, candidate answer phrase head) can be shown as different color so.
In addition, according to user's profile setting, answer summary can present with text (potential image and the film of comprising) or voice (quoting the voice synthesis module in this case) as a result.
At last, output is sent to mobile communications device.
Fig. 7 shows the table of the profile contents that may have portable user, comprises the possessory parameter specific to mobile device and mobile device.
The storage of user's profile is about sign, the authentication of user and his or her mobile communications device and be used for adjusting to the user data of the behavior of mobile information access server.
In the mobile information access server, discern each other with user identifier (user ID).At a profile running graphic user interface based on the user of the Internet, secret signal (password) limited subscriber oneself can only be visited its profile.Keep a row user ID feature (caller identification), include but not limited to customer call sign, Mobile Directory Number for example, when retrieve user information from user's profile, it is as key word.In storage, preserve the attribute and the ability (mobile device information) of user's mobile device, comprise whether the size of whether supporting similar color or bright apparent feature screen and resolution mobile communications device support SMS, EMS and MMS respectively, whether it be the 3G phone, whether it can merge into one with a plurality of text messages.The system action of one row preference (user preference) storage user preferences includes but not limited to: the absolute and relative order of motif area significance, the maximum quantity of response message (for example maximum quantity of SMS (mobile sms service)) require, whether with due regard to send MMS (modular memory system) and whether the user can accept adding advertisements.
Boolean register (location awareness flag) the storage user automated location of whether expressing consent detects, and allows to consider to send the user's of inquiry mobile communications device thus, to improve this search (location-based search).User's historical problem (question history) allows to consider that previous informational needs is to improve result for retrieval.One website liked of row and service (Cup of tea thing) allow search focus on these more may the website relevant with user's request on.(electronic mail account) for information about how to be connected with user's e-mail storage allows the personal information retrieval from the user.The information that the account balance storage is relevant with subscriber's account, for example currency in the reward scheme or virtual credit point account.
Fig. 8 shows the form of the embodiment that replys that portable information retrieval method generates according to the present invention.Answer summary comprises one group of answer candidate windows (802 to 807), and each all comprises a definite answer candidate this window, by the context of the left side (804) and the right (805) around (that is, in its found file around the text of answer candidate).
In one embodiment, answer candidate windows separates 808 by separator sign (for example, but being not limited to character "/"), with mark boundaries, prevents to obscure the user.In one embodiment, comprising the answer candidate that most probable replys inserts at initial position 806, it is without any context, with guarantee reply may very long situation under, after last answer candidate windows 807 by not causing losing best answer candidate.
Fig. 9 shows the structural drawing according to the embodiment of the server that is used for mobile information access of the present invention.
The mobile information access server comprises receiver 900 and the transmitter 901 that is used for receiving from mobile communications device message, and it is sent back to mobile communications device with message.
The input of voice recognition unit 910, problem types unit 920, question analysis unit 921 and user profile table storer 940 links to each other with the output of receiver 900.Voice recognition unit 910 also is connected with problem types unit 920, query analysis unit 921 and receiver 900.Input with query construction unit 930 is connected with query analysis unit 921 in the output of problem types unit 920.The output of this query construction unit is connected with the input of retrieval unit 950.
The output of user's profile is connected with input, classification/authentication unit 970 and the answer summary unit 980 of query construction unit 930.
The output of retrieval unit is connected with the input of candidate answer with extraction unit 960.Input with answer summary component units (980) is connected with extraction unit in the output of candidate answer.The output of answer summary component units (980) is connected with phonetic synthesis unit (911) input with transmitter (901).The output of phonetic synthesis unit also is connected with this transmitter.
Claims (15)
1. method that is used for mobile information access, it is carried out in network computer system, and this network computer system comprises at least one mobile information access server and one or more information retrieval system, and the method comprising the steps of:
Receive message from mobile communications device;
The message that analysis receives;
Form one or more inquiries according to message analysis;
Obtain file according to these one or more inquiries;
From this document, extract candidate answer;
The checking candidate answer;
Form answer summary;
Beam back this answer summary to mobile communications device,
It is characterized in that:
This answer summary is subject to predetermined size.
2. according to the method for claim 1, be characterised in that size according to following one or more condition restriction answer summary:
The maximum display size of-mobile communications device;
The maximum message size of-mobile communication protocol;
-each user's hobby.
3. according to the method for claim 1 or 2, be characterised in that the step of analyzing the message that receives comprises: the step of from message, extracting a problem or language phrase with natural language.
4. according to the method for claim 3, be characterised in that the step of analyzing the information that receives also comprises: definite this problem or the type of phrase and step of from message, extracting of linguistic property with natural language.
5. according to the method for any one claim of front, be characterised in that also to comprise step:
Whether this message of-check receives with speech form; And, if,
-be textual form with this message from speech conversion by means of automatic speech recognition.
6. according to the method for claim 4 or 5, be characterised in that forming one or more inquiries has considered whether the problem of this extraction or phrase relate to a named entity.
7. according to the method for any one claim of front, be characterised in that automatic identification user and discern the profile of retrieval user according to this.
8. according to the method for claim 7, be characterised in that and from user's profile, obtain one or more following contents:
The maximum display size of-mobile communications device;
The maximum message size of-mobile communication protocol;
-each user's hobby.
9. according to the method for claim 7 or 8, be characterised in that the step that forms one or more inquiries is also based on the information that obtains from user profile table.
10. according to one method in the claim 7 to 9, be characterised in that the step of extracting answer candidate also considered the information that obtains from user profile table.
11., be characterised in that the step of checking answer candidate has also been considered from the information of user profile table acquisition according to the method for claim 7 to 10.
12., be characterised in that the step of composition answer summary has also been considered from the information of user profile table acquisition according to the method for claim 7 to 11.
13. method according to any one claim of front, be characterised in that the step of forming answer summary generates answer summary, this answer summary comprises one group of answer candidate windows, and each all comprises definite answer candidate these windows, and is centered on by the context on the left side and the right.
14., be characterised in that the step of forming answer summary generates answer summary, in answer summary, will comprise the answer candidate of replying with the highest checking grade and be inserted in without any contextual initial position according to the method for any one claim of front.
15. an equipment that is used for mobile information access comprises:
Receive the unit of message from mobile communications device;
The unit of the message that analysis receives;
Form the unit of one or more inquiries according to message analysis;
Obtain the unit of file based on one or more inquiries;
From file, extract the unit of candidate answer;
The unit of checking candidate answer;
Form the unit of answer summary, wherein answer summary is subject to predetermined size; With
Beam back the unit of answer summary to mobile communications device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2005800476958A CN101120341A (en) | 2005-02-06 | 2005-02-06 | Method and equipment for performing mobile information access using natural language |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2005800476958A CN101120341A (en) | 2005-02-06 | 2005-02-06 | Method and equipment for performing mobile information access using natural language |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101120341A true CN101120341A (en) | 2008-02-06 |
Family
ID=39055649
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2005800476958A Pending CN101120341A (en) | 2005-02-06 | 2005-02-06 | Method and equipment for performing mobile information access using natural language |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101120341A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737096A (en) * | 2011-03-31 | 2012-10-17 | 微软公司 | Conversational understanding based on position |
| CN103268315A (en) * | 2012-12-31 | 2013-08-28 | 威盛电子股份有限公司 | Natural language conservation method and system |
| US9454962B2 (en) | 2011-05-12 | 2016-09-27 | Microsoft Technology Licensing, Llc | Sentence simplification for spoken language understanding |
| US9760566B2 (en) | 2011-03-31 | 2017-09-12 | Microsoft Technology Licensing, Llc | Augmented conversational understanding agent to identify conversation context between two humans and taking an agent action thereof |
| US9842168B2 (en) | 2011-03-31 | 2017-12-12 | Microsoft Technology Licensing, Llc | Task driven user intents |
| US9858343B2 (en) | 2011-03-31 | 2018-01-02 | Microsoft Technology Licensing Llc | Personalization of queries, conversations, and searches |
| US10049667B2 (en) | 2011-03-31 | 2018-08-14 | Microsoft Technology Licensing, Llc | Location-based conversational understanding |
| US10061843B2 (en) | 2011-05-12 | 2018-08-28 | Microsoft Technology Licensing, Llc | Translating natural language utterances to keyword search queries |
| US10642934B2 (en) | 2011-03-31 | 2020-05-05 | Microsoft Technology Licensing, Llc | Augmented conversational understanding architecture |
| CN116610790A (en) * | 2023-07-18 | 2023-08-18 | 阿里健康科技(中国)有限公司 | Method, device, equipment and medium for acquiring response data |
-
2005
- 2005-02-06 CN CNA2005800476958A patent/CN101120341A/en active Pending
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10296587B2 (en) | 2011-03-31 | 2019-05-21 | Microsoft Technology Licensing, Llc | Augmented conversational understanding agent to identify conversation context between two humans and taking an agent action thereof |
| US10642934B2 (en) | 2011-03-31 | 2020-05-05 | Microsoft Technology Licensing, Llc | Augmented conversational understanding architecture |
| US10585957B2 (en) | 2011-03-31 | 2020-03-10 | Microsoft Technology Licensing, Llc | Task driven user intents |
| CN102737096A (en) * | 2011-03-31 | 2012-10-17 | 微软公司 | Conversational understanding based on position |
| CN102737096B (en) * | 2011-03-31 | 2017-08-25 | 微软技术许可有限责任公司 | Location-based session understands |
| US9760566B2 (en) | 2011-03-31 | 2017-09-12 | Microsoft Technology Licensing, Llc | Augmented conversational understanding agent to identify conversation context between two humans and taking an agent action thereof |
| US9842168B2 (en) | 2011-03-31 | 2017-12-12 | Microsoft Technology Licensing, Llc | Task driven user intents |
| US9858343B2 (en) | 2011-03-31 | 2018-01-02 | Microsoft Technology Licensing Llc | Personalization of queries, conversations, and searches |
| US10049667B2 (en) | 2011-03-31 | 2018-08-14 | Microsoft Technology Licensing, Llc | Location-based conversational understanding |
| US9454962B2 (en) | 2011-05-12 | 2016-09-27 | Microsoft Technology Licensing, Llc | Sentence simplification for spoken language understanding |
| US10061843B2 (en) | 2011-05-12 | 2018-08-28 | Microsoft Technology Licensing, Llc | Translating natural language utterances to keyword search queries |
| CN103268315B (en) * | 2012-12-31 | 2016-08-03 | 威盛电子股份有限公司 | Natural language dialogue method and system thereof |
| CN103268315A (en) * | 2012-12-31 | 2013-08-28 | 威盛电子股份有限公司 | Natural language conservation method and system |
| CN116610790A (en) * | 2023-07-18 | 2023-08-18 | 阿里健康科技(中国)有限公司 | Method, device, equipment and medium for acquiring response data |
| CN116610790B (en) * | 2023-07-18 | 2023-10-10 | 阿里健康科技(中国)有限公司 | Method, device, equipment and medium for acquiring response data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11315546B2 (en) | Computerized system and method for formatted transcription of multimedia content | |
| CN101120341A (en) | Method and equipment for performing mobile information access using natural language | |
| US20080154871A1 (en) | Method and Apparatus for Mobile Information Access in Natural Language | |
| US10162816B1 (en) | Computerized system and method for automatically transforming and providing domain specific chatbot responses | |
| JP5827380B2 (en) | Non-standard position-based text input | |
| CN102163198B (en) | A method and a system for providing new or popular terms | |
| CN1871603B (en) | System and method for processing a query | |
| US20170249309A1 (en) | Interpreting and Resolving Conditional Natural Language Queries | |
| CN101730008A (en) | Method, system, and apparatus for message generation | |
| CN102110170B (en) | System with information distribution and search functions and information distribution method | |
| US20090276500A1 (en) | Microblog search engine system and method | |
| US20070192309A1 (en) | Method and system for identifying sentence boundaries | |
| US20020007303A1 (en) | System for conducting electronic surveys | |
| CN107992513B (en) | Information processing system and method for realizing information processing | |
| US20070078814A1 (en) | Novel information retrieval systems and methods | |
| US20060184625A1 (en) | Short query-based system and method for content searching | |
| US20110246442A1 (en) | Location Activity Search Engine Computer System | |
| CN101866341A (en) | Information push method, device and system | |
| CN101112078A (en) | Information communication terminal, information communication system, information communication method, information communication program, and recording medium on which program is recorded | |
| US20130288725A1 (en) | Mobile terminal to recommend a short message recipient | |
| CN104991943A (en) | Music searching method and apparatus | |
| EP3667586A1 (en) | Electronic communication system with drafting assistant and method of using same | |
| CN101566995A (en) | Method and system for integral release of internet information | |
| CN103136192A (en) | Method and system of identifying translation demand | |
| US20100332508A1 (en) | Methods and systems for extracting and analyzing online discussions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: INFELAIT CO., LTD. Free format text: FORMER OWNER: LENGUIT CO LTD Effective date: 20081114 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20081114 Address after: British Northampton Applicant after: Yingfu Wright Co Ltd Address before: German Bart Begg Zabern Applicant before: Lenguit Co.,Ltd |
|
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |