CN110399462A - A kind of querying method and device of information - Google Patents
A kind of querying method and device of information Download PDFInfo
- Publication number
- CN110399462A CN110399462A CN201910682736.4A CN201910682736A CN110399462A CN 110399462 A CN110399462 A CN 110399462A CN 201910682736 A CN201910682736 A CN 201910682736A CN 110399462 A CN110399462 A CN 110399462A
- Authority
- CN
- China
- Prior art keywords
- text
- inquiry
- parsing text
- scene
- information
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
Abstract
This application discloses a kind of querying method of information and devices, wherein, method includes: reception voice inquirement, voice inquirement is the voice for the inquiry content that user sends, text resolution is carried out to the text that voice inquirement is converted to depending at least on vertical category feature, obtain parsing text, the category feature that hangs down is preset civil aviaton's Scene Semantics, determine the inquiry scene of parsing text, inquiry scene is to inquire content query service type affiliated in preset query service type, according to the inquiry scene of parsing text, the query result of inquiry parsing text.By scheme provided by the present application, more easy information inquiry may be implemented.
Description
Technical field
This application involves the querying methods and device of field of information processing more particularly to a kind of information.
Background technique
People are when taking flight, it is often necessary to obtain the multidate information of flight (for example, the security check of flight, check-in etc.
Information) and airport related service information (for example, the information such as airport weather and airport food and drink).
Currently, generally searching the multidate information of flight and the related service information on airport by PC or dedicated APP.Specifically
, the respective interface into PC or dedicated APP is found, and input is sought for indicating the text information of inquiry content into interface
Look for required information.The lookup at interface results in certain difficulty that has of information inquiry, and text input results in information inquiry
It is not convenient enough.
Therefore, easier Flight Information querying method becomes a kind of research tendency.
Summary of the invention
This application provides a kind of querying method and device, and it is an object of the present invention to provide a kind of skill of more easy information inquiry
Art scheme.
To achieve the goals above, this application provides following technical schemes:
This application provides a kind of querying methods of information, comprising:
Receive voice inquirement;The voice inquirement is the voice for the inquiry content that user sends;
Text resolution is carried out to the text that the voice inquirement is converted to depending at least on vertical category feature, obtains parsing text
This;The vertical category feature is preset civil aviaton's Scene Semantics;
Determine the inquiry scene of the parsing text;The inquiry scene is the inquiry content in preset query service
Query service type belonging in type;
According to the inquiry scene of the parsing text, the query result of the parsing text is inquired.
Optionally, further includes:
In the case where meeting default loading environment, the Flight Information that load meets preset condition is the flight letter preloaded
Breath;The default loading environment includes: the information or the Flight Information update for receiving preloading for detecting and indicating that user logs in
Information;The preset condition includes: in the trip Flight Information that user has subscribed, and/or, user is predetermined and the inquiry
In the Flight Information of the flight provided in voice, belong to the information of preset high frequency query term;
The inquiry scene according to the parsing text, inquires the query result of the parsing text, comprising:
Belong to preset personal information inquiry scene, and the flight of the preloading in the inquiry scene of the parsing text
There are in the case where the query result of the parsing text in information, from the Flight Information of the preloading, the solution is obtained
Analyse the query result of text.
Optionally, the inquiry scene according to the parsing text, inquires the query result of the parsing text, also wraps
It includes:
There is no in the case where the query result of the parsing text in the Flight Information of the preloading, from described pre-
In the Flight Information of load, at least partly conditional information in the corresponding conditional information of the parsing text is obtained;
After getting the corresponding conditional information of the parsing text, according to default pair between inquiry scene and query interface
It should be related to, by calling the corresponding query interface of inquiry scene of the parsing text, obtain and meet looking into for the parsing text
The information for asking the corresponding conditional information of scene is the query result.
Optionally, the inquiry scene according to the parsing text, inquires the query result of the parsing text, packet
It includes:
In the case where the inquiry scene of the parsing text belongs to preset public information inquiry scene, the solution is obtained
Analyse the corresponding conditional information of inquiry scene of text;
According to the default corresponding relationship between inquiry scene and query interface, by the inquiry scene for calling the parsing text
Corresponding query interface, the information that acquisition meets the corresponding conditional information of inquiry scene of the parsing text is inquiry knot
Fruit.
Optionally, the inquiry scene of the determination parsing text, comprising:
There is the height that the matching degree between the parsing text is greater than preset matching threshold value in the first default corresponding relationship
In the case where frequency inquiry parsing text, the corresponding inquiry scene of parsing text will be inquired with the high frequency, it is literary as the parsing
This inquiry scene;First corresponding relationship is default between preset high frequency inquiry parsing text and preset inquiry scene
Corresponding relationship;
There is no the matching degrees between the parsing text to be greater than described preset in preset first corresponding relationship
In the case where the high frequency inquiry parsing text of matching threshold, judge in the second corresponding relationship with the presence or absence of between the parsing text
Similarity be greater than the pre-set text of default similar threshold value;Second corresponding relationship is each default text in pre-set text set
Originally the default corresponding relationship between inquiry scene;
The similarity existed between the parsing text in second corresponding relationship is greater than the default similar threshold value
Pre-set text in the case where, will inquiry scene corresponding with the pre-set text, as it is described parsing text inquiry scene.
Optionally, the inquiry scene of the determination parsing text, further includes:
There is no the similarities between the parsing text to be greater than described preset in preset second corresponding relationship
In the case where the pre-set text of similar threshold value, preset helper is called;The helper output for user for determining
The expression way of the inquiry content provides the sentence of reference.
Optionally, it is described according to it is described parsing text inquiry scene, inquire it is described parsing text query result it
Afterwards, further includes:
The query result is converted into voice;
Play the voice.
Present invention also provides a kind of inquiry units of information, comprising:
Receiving module, for receiving voice inquirement;The voice inquirement is the voice for the inquiry content that user sends;
Parsing module, for carrying out text solution to the text that the voice inquirement is converted to depending at least on vertical category feature
Analysis obtains parsing text;The vertical category feature is preset civil aviaton's Scene Semantics;
Determining module, for determining the inquiry scene of the parsing text;The inquiry scene is that the inquiry content exists
Query service type belonging in preset query service type;
Enquiry module inquires the query result of the parsing text for the inquiry scene according to the parsing text.
Optionally, further includes:
Loading module, in the case where satisfaction default loading environment, the Flight Information that load meets preset condition to be
The Flight Information of preloading;The default loading environment includes: the information or receive preloading for detecting and indicating that user logs in
Flight Information update information;The preset condition includes: in the trip Flight Information that user has subscribed, and/or, Yong Huwei
In the Flight Information of the flight provided in the predetermined and described voice inquirement, belong to the information of preset high frequency query term;
The enquiry module inquires the inquiry knot of the parsing text for the inquiry scene according to the parsing text
Fruit, comprising:
The enquiry module inquires field specifically for belonging to preset personal information in the inquiry scene of the parsing text
Scape, and in the Flight Information of the preloading there are it is described parsing text query result in the case where, from the preloading
In Flight Information, the query result of the parsing text is obtained.
Optionally, the enquiry module is also used to that the parsing text is not present in the Flight Information of the preloading
Query result in the case where, from the Flight Information of the preloading, obtain in the corresponding conditional information of the parsing text
At least partly conditional information;
After getting the corresponding conditional information of the parsing text, according to default pair between inquiry scene and query interface
It should be related to, by calling the corresponding query interface of inquiry scene of the parsing text, obtain and meet looking into for the parsing text
The information for asking the corresponding conditional information of scene is the query result.
Optionally, the enquiry module inquires the parsing text for the inquiry scene according to the parsing text
Query result, comprising:
The enquiry module, specifically for belonging to preset public information inquiry field in the inquiry scene of the parsing text
In the case where scape, the corresponding conditional information of inquiry scene of the parsing text is obtained;
According to the default corresponding relationship between inquiry scene and query interface, by the inquiry scene for calling the parsing text
Corresponding query interface, the information that acquisition meets the corresponding conditional information of inquiry scene of the parsing text is inquiry knot
Fruit.
Optionally, the determining module, for determining the inquiry scene of the parsing text, comprising:
The determining module, specifically for there is the matching degree between the parsing text in the first default corresponding relationship
It, will inquiry corresponding with high frequency inquiry parsing text in the case where high frequency inquiry parsing text greater than preset matching threshold value
Scene, the inquiry scene as the parsing text;First corresponding relationship is for preset high frequency inquiry parsing text and in advance
If inquiry scene between default corresponding relationship;
There is no the matching degrees between the parsing text to be greater than described preset in preset first corresponding relationship
In the case where the high frequency inquiry parsing text of matching threshold, judge in the second corresponding relationship with the presence or absence of between the parsing text
Similarity be greater than the pre-set text of default similar threshold value;Second corresponding relationship is each default text in pre-set text set
Originally the default corresponding relationship between inquiry scene;
The similarity existed between the parsing text in second corresponding relationship is greater than the default similar threshold value
Pre-set text in the case where, will inquiry scene corresponding with the pre-set text, as it is described parsing text inquiry scene.
Optionally, the determining module is also used to be not present and the parsing in preset second corresponding relationship
In the case that similarity between text is greater than the pre-set text of the default similar threshold value, preset helper is called;It is described
Helper output is for determining that the expression way of the inquiry content provides the sentence of reference for user.
Optionally, further includes:
Conversion module inquires the parsing text for the inquiry scene in the enquiry module according to the parsing text
After this query result, the query result is converted into voice;
Playing module, for playing the voice.
In querying method and device described herein, voice inquirement is received, wherein voice inquirement is looking into for user's transmission
Ask the voice of content, i.e., the application it is received be speech form inquiry content, i.e. user can directly say the inquiry of oneself
Content.Voice inquirement is converted to text and carried out depending at least on vertical category feature to the text that voice inquirement is converted to by the application
Parsing obtains parsing text, determines the inquiry scene of parsing text, according to the inquiry scene of parsing text, inquiry parsing text
Query result.
Firstly, since therefore the inquiry content that the application is suitable for speech form can be convenient user query.Secondly, this
Voice inquirement is converted to text by application, and is parsed depending at least on vertical category feature to text, and due to hanging down, category feature is default
Civil aviaton's Scene Semantics so that the semanteme of parsing text that parsing obtains meets civil aviaton's scene, so that parsing obtained solution
It is more acurrate to analyse text;Finally, the inquiry scene of parsing text is determined, according to the inquiry scene of parsing text, inquiry parsing text
Query result, therefore, obtained query result is the query result to match with the inquiry content of user, and query result has more
Targetedly, required information is found on the page without user oneself decorrelation, also, compared to the existing solution to text
Analysis method, the accuracy for parsing text parsed using the application is higher, therefore, is based on the higher parsing text of accuracy
The accuracy of obtained query result is higher.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of application for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of flow chart of the querying method of information disclosed in the embodiment of the present application;
Fig. 2 is the process schematic for analytically determining inquiry scene disclosed in the embodiment of the present application in text;
Fig. 3 is the disclosed inquiry scene according to parsing text of the embodiment of the present application, the query result of inquiry parsing text
Process schematic;
Fig. 4 is a kind of structural schematic diagram of the inquiry unit of information provided by the embodiments of the present application.
Specific embodiment
The inquiry unit of information provided by the embodiments of the present application, can be separately provided, and also can integrate in existing boat
In class trip information on services inquiry unit such as APP, it may operate on portable personal terminal, such as mobile phone, can also transport
Row is in the common equipment with speech recognition and speech-sound synthesizing function, for example, on the intelligent robot of airport setting.The application is real
The querying method for applying the information of example offer is realized by the inquiry unit of above- mentioned information, it is therefore intended that, a kind of information of simplicity is provided
Inquiry mode.
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of embodiments of the present application, instead of all the embodiments.It is based on
Embodiment in the application, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall in the protection scope of this application.
Fig. 1 is a kind of querying method of information provided by the embodiments of the present application, comprising the following steps:
S101, voice inquirement is received.
Voice inquirement in this step refers to the inquiry content that user is sent with speech form, for example, user needs inquiry " bright
Which its flight from Beijing to Guangzhou has ", then this step receives voice inquirement is the voice that user says: " tomorrow from
Which the flight in Beijing to Guangzhou has ".
S102, in the case where meeting default loading environment, the Flight Information that load meets preset condition be preloading
Flight Information.
Wherein, default loading environment may include: the information or the boat for receiving preloading for detecting and indicating that user logs in
The information of class's information update.For example, detecting preset (the speech polling side provided by the embodiments of the present application APP of user's login
The APP that case is installed) in the case where, load meets the Flight Information of preset condition.
Wherein, preset condition includes: in trip Flight Information that user has subscribed, and/or, user is predetermined and inquires language
In the Flight Information of the flight provided in sound, belong to the information of preset high frequency query term.High frequency query term can be according to history
Query information counts to obtain, for example, flight number, departure place, destination, journey time, safety check information, check-in information and from
Port information etc..
Wherein, it includes following two that load, which meets the mode of the Flight Information of preset condition:
The first, used identity information when being logged in by user, inquiring the user whether there is the trip subscribed
Flight, and if it exists, then load in the trip Flight Information that the user has subscribed, belong to the information of preset high frequency query term.
Second, when being logged in by user, used identity information inquires the user and the trip subscribed is not present
In the case where flight, using the flight identified from the voice inquirement that user sends as the interested trip flight of user, and
It loads in interested trip Flight Information, belongs to the information of preset high frequency query term.
S103, voice inquirement is converted into text.
In this step, voice inquirement is converted to the specific implementation process of text is the prior art, and which is not described herein again.
S104, text resolution is carried out to text depending at least on vertical category feature, obtains parsing text.
In this step, the purpose for being parsed to obtain parsing text to text is: carrying out semantic parsing to text, obtains
For indicating the information of text semantic, for convenience, it is known as the information for being used to indicate text semantic to parse text.
Specifically, due to being applied to civil aviaton's class scene in the embodiment of the present application, it can be depending at least on vertical category feature pair
Text is parsed, wherein the category feature that hangs down is preset civil aviaton's Scene Semantics, specifically, preset civil aviaton's Scene Semantics include
But it is not limited to: flight number and the default word for referring to civil aviaton's scene.For example, flight number is usually 4-digit number, it is assumed that user is defeated
The text that the voice entered is converted to be " 1502 ", do not consider hang down category feature in the case where, the meaning of " 1502 " may include it is a variety of,
And in the case where considering vertical category feature, final parsing result is " 1502 flight ".
Referring to the default word of civil aviaton's scene includes but is not limited to: flight, departure time etc., another example is: voice is converted to
Text be " which tomorrow has from Beijing to Guangzhou ", in the case where not considering vertical category feature, meaning is unclear, and considers to hang down
In the case where category feature, final parsing result be " which the flight from Beijing to Guangzhou have tomorrow ".
S105, the inquiry scene for determining parsing text.
In this step, the inquiry scene inquiry clothes affiliated in preset query service type for the inquiry content of user
Service type.In the present embodiment, inquiry scene can be any for flight inquiring, airport weather, airport food and drink and safety check inquiry etc.
One, certainly, in practice, scene is inquired other than this step is enumerated, can also include other content, the present embodiment is not
The particular content of inquiry scene is limited.
Specifically, the process of inquiry scene is analytically determined in this step in text, as shown in Fig. 2, may include: step
A1~step A5:
A1, judge to be greater than preset matching threshold with the presence or absence of the matching degree between parsing text in preset first corresponding relationship
The high frequency inquiry parsing text of value, if it is, A2 is executed, if it is not, then executing A3.
In this step, the first corresponding relationship is corresponding between preset high frequency inquiry parsing text and preset inquiry scene
Relationship.High frequency inquiry parsing text refer to by high frequency inquire parsing text, for example, " which the flight from A to B have tomorrow ".
Specifically, in the present embodiment, in advance according to historical query information, setting a variety of high frequency inquiry parsing texts, wherein be set
Fixed every kind of high frequency inquiry parsing text setting inquiry scene, obtains the corresponding pass that high frequency is inquired parsing text with inquired between scene
System.
It is the prior art in the specific deterministic process of this step, which is not described herein again.
A2, the corresponding inquiry of high frequency inquiry parsing text that the matching degree between parsing text is greater than to preset matching threshold value
Scene, the inquiry scene as parsing text.
A3, judge to be greater than default similar threshold value with the presence or absence of the similarity with parsing text in preset second corresponding relationship
Pre-set text, if it is, execute A4, if it is not, then execute A5.
In the present embodiment, the second corresponding relationship is the default corresponding relationship between pre-set text set and inquiry scene.Its
In, each pre-set text in pre-set text set is for indicating a kind of text for inquiring scene, also, pre-set text set
Including text representation inquiry scene, including all possible inquiry scene.Specifically, to the pre- of any one inquiry scene
If text may be set according to actual conditions, the present embodiment is not made the particular content of the pre-set text of any one inquiry scene
It limits.
Specifically, any default text in parsing text and pre-set text set can be calculated using Simase LSTM algorithm
This similarity, specific calculating process are the prior art, and which is not described herein again.
A4, the corresponding inquiry scene of the default pre-set text of similar threshold value will be greater than with the similarity of parsing text, as
Parse the inquiry scene of text.
A5, preset helper is called.
There is no the similarities between parsing text to be greater than the pre- of default similar threshold value in preset second corresponding relationship
If in the case where text, i.e., in the case where that can not determine the inquiry scene of parsing text, executing this step.
Specifically, in this step, helper exports example sentence, wherein example sentence refers to: for being determined in inquiry to user
The expression way of appearance provides the sentence of reference.The example sentence exported by helper, can make user be looked into according to needed for oneself
The content of inquiry determines the expression way of inquiry content according to the expression way of example sentence.To, return execute A1 so that according to
Family adjusts the voice after expression way, determines the inquiry scene of the parsing text of the inquiry content of voice.
The inquiry scene of S106, foundation parsing text, the query result of inquiry parsing text.
In the present embodiment, all preset inquiry scenes are divided into two classes, one kind is that personal information inquires scene, that is, is used
In the inquiry scene for obtaining information relevant to personal information.Another kind of is public information inquiry scene, i.e., for obtaining and public affairs
The inquiry scene of the relevant information of co-information.In the present embodiment, inquiry included by personal information inquiry scene is previously set
Which scene has, such as setting personal information inquiry scene includes: flight inquiring, safety check inquiry etc., public information inquiry scene
Which included inquiry scene has, and obtains preset personal information inquiry scene and preset public information inquiry scene, example
Such as airport weather, airport food and drink.
In this step, the inquiry scene according to parsing text, the process of the query result of inquiry parsing text, such as Fig. 3
It is shown, may include step B1~step B7:
B1, judge whether the inquiry scene for parsing text belongs to preset personal information inquiry scene, if it is not, then executing
B2, if it is, executing B4.
B2, the corresponding conditional information of inquiry scene for obtaining parsing text.
In the present embodiment, querying condition has all been correspondingly arranged for every kind of inquiry scene to get inquiry scene and institute is arrived
Default corresponding relationship between the querying condition needed.Wherein, scene is inquired for any one, for the corresponding inquiry of inquiry scene
Condition refers to: at least one condition used needed for the query result for inquiring parsing text.In this step, from inquiry field
In default corresponding relationship between scape and required querying condition, determines the corresponding querying condition of inquiry scene of parsing text, be
Querying condition needed for parsing text.The particular content of querying condition needed for the parsing text being met, as obtaining
Conditional information needed for parsing the query result of text.
Specifically, the acquisition modes of the corresponding conditional information of inquiry scene of parsing text may include:
The first, it is analytically obtained in text.Specifically, can determine the condition letter in parsing text using technologies such as NLP
Breath, specific implementation process are the prior art, and which is not described herein again.For example, parsing text is " flight of the tomorrow from Beijing to Guangzhou
Which is there " parse text inquiry scene be " flight inquiring ", " flight inquiring " corresponding querying condition set include: " when
Between ", " origin " and " destination ".In this step, parse text present in querying condition be " time ", " origin " and
" destination ".The particular content of querying condition " time " is " tomorrow ", and the particular content of querying condition " origin " is " Beijing ",
The particular content of querying condition " destination " is " Guangzhou ".I.e. parse text conditional information include: " tomorrow ", " Beijing " and
" Guangzhou ".
The second, the conditional information got in analytically text is the corresponding conditional information of inquiry scene for parsing text
In partial information in the case where, can pass through that call preset query interface to obtain remaining based on the conditional information obtained
Conditional information.
Third, the conditional information got in analytically text are the corresponding conditional information of inquiry scene for parsing text
In partial information in the case where, remaining conditional information can be obtained by way of puing question to user.
B3, meet that parse text corresponding with the corresponding query interface of the parsing inquiry scene of text, acquisition by calling
The information of conditional information is query result.
In the present embodiment, corresponding query interface is both provided with for every kind of inquiry scene, wherein any one inquiry
The corresponding query interface of scene, for providing information relevant to the inquiry scene, and from information provided by query interface,
Obtain query result of the information for meeting the corresponding conditional information of parsing text as parsing text.Detailed process may include:
The corresponding conditional information of text will be parsed and be sent to corresponding query interface, receive the query result that query interface is sent.
With the presence or absence of the query result of parsing text in the Flight Information that B4, judgement preload, if it is, B5 is executed,
If it is not, then executing B6.
In this step, judge that the process of the query result in the Flight Information preloaded with the presence or absence of parsing text can be with
Include:
The default high frequency item according to involved in the Flight Information of preloading in advance, sets in the Flight Information of preloading and includes
The parsing text of query result, and be default parsing text by set parsing text.In turn, it is by judgement parsing text
It is no to belong to default parsing text, judge in the Flight Information preloaded with the presence or absence of the query result of parsing text.Certainly, this
It is a kind of implementation, other modes can also be used in practice, the present embodiment does not limit concrete implementation mode.
B5, from the Flight Information of preloading, obtain parsing text query result.
In the case where there is the query result of parsing text in the Flight Information of preloading, this step is executed, i.e., from pre-
In the Flight Information of load, the query result of parsing text is obtained.
B6, from the Flight Information of preloading, obtain parsing text the corresponding conditional information of inquiry scene at least
Partial condition information.
In the Flight Information of preloading, there is no the query results of parsing text, for example, parsing text is that " I am present
Can be with safety check ", as preloading Flight Information in there is no the user flight belonging to security check currently whether open
Therefore the information put is unable to get the query result of the parsing text from the Flight Information of preloading.But to obtain
The query result of the parsing text needs to know the corresponding conditional information of inquiry scene of the parsing text.The parsing text
Inquiring scene is " safety check inquiry ", may include: the particular content of " flight number " for " safety check inquiry " corresponding conditional information
The particular content of " date ", wherein which day " date " refer to.At this point it is possible to from the Flight Information of preloading
Middle acquisition corresponding conditional information of parsing text.
It should be noted that in practice, the inquiry scene of parsing text can be obtained from the Flight Information of preloading
At least partly conditional information in corresponding conditional information.The corresponding condition of inquiry scene of some parsing texts is believed
Breath, can obtain from the Flight Information of preloading, can only for the corresponding conditional information of inquiry scene of some parsing texts
The conditional information of enough fetching portions from the Flight Information of preloading, the condition in Flight Information for being not present in preloading
The acquisition modes of information (for convenience, referred to as residue condition information), can be with reference to the acquisition conditional information provided in B2
Mode, which is not described herein again.In the case where the corresponding conditional information of inquiry scene for parsing text is all got, execute
B7。
B7, in the case where getting the corresponding conditional information of inquiry scene of parsing text, pass through and call and parsing text
This corresponding query interface of inquiry scene, the information that acquisition meets the corresponding conditional information of parsing text is query result.
The implementation procedure of this step can refer to B3, and which is not described herein again.
S107, output query result.
In this step, query result is converted to by voice using speech synthesis technique, and to the voice being converted into
Row plays.
The present embodiment has the advantages that
Beneficial effect one,
In the present embodiment, the Flight Information for meeting preset condition is loaded in the case where meeting default loading environment, wherein
Preset condition includes: in the Flight Information for the trip that user has subscribed, and/or, what user did not provided in predetermined and voice inquirement
In the Flight Information of flight, belong to the information of preset high frequency query term.So that belonging to for the inquiry scene for parsing text pre-
If personal information inquire scene, query result or the parsing that parsing text can be obtained from the Flight Information of preloading are literary
At least partly conditional information needed for this inquiry scene, it can thus be avoided presetting query interface by calling to be solved
At least partly conditional information needed for analysing the query result of text or parsing the inquiry scene of text.Also, the boat preloaded
Class's information is related to preset high frequency query term, and therefore, the inquiry scene for belonging to the inquiry scene of preset personal information is high frequency
Therefore the scene of appearance passes through the inquiry of the query result or parsing text of the Flight Information inquiry parsing text of preloading
At least partly conditional information needed for scene can largely reduce the number for calling relevant inquiring interface, in turn, improve inquiry speed
Degree, to improve user experience.
Beneficial effect two,
In the present embodiment, the Flight Information of preloading include: in the flight for the trip that user has subscribed and/or user not
In the Flight Information of the flight provided in reservation and voice inquirement, belong to the information of preset high frequency query term, wherein default height
Frequency item includes flight number, and therefore, the user is when inquire the trip flight subscribed, in the voice inquirement of transmission
Flight can not included, i.e. user is not necessarily to say the flight of inquiry, so that it may be inquired, and then user experience can be improved.
Beneficial effect three,
In the present embodiment, judge default corresponding between each pre-set text and preset inquiry scene in pre-set text set
In relationship, if during existing and parsing the pre-set text that the similarity between text is greater than default similar threshold value, determine solution
The similarity in text and pre-set text set between any one default file is analysed, using the depth based on Simase LSTM
Spend learning algorithm, due to the deep learning algorithm based on Simase LSTM calculate two sentences similarity on have it is higher
Accuracy, therefore, in the present embodiment, determine parsing text and pre-set text between similarity accuracy with higher.
Beneficial effect four,
In the present embodiment, during determining the inquiry scene of parsing text, first by judging that default high frequency is looked into
It askes parsing text and inquires in the default corresponding relationship of scene, if exist and be greater than preset matching threshold with the matching degree of parsing text
The high frequency inquiry parsing text of value, in default high frequency inquiry parsing text and is inquired in the default corresponding relationship of scene, exist with
In the case where high frequency inquiry parsing text of the matching degree greater than preset matching threshold value for parsing text, by the matching with parsing text
Degree is greater than inquiry scene of the corresponding inquiry scene of high frequency inquiry parsing text of preset matching threshold value as parsing text, due to
High frequency query text refers to the more parsing text of inquiry times, therefore, first inquires parsing text and inquiry from preset high frequency
In the default corresponding relationship of scene, determine that the inquiry scene of parsing text, greater probability can determine looking into for parsing text
Scene is ask, in turn, the speed for determining the inquiry scene of parsing text can be improved.
It in default high frequency inquiry parsing text and inquires in the default corresponding relationship of scene, there is no with parsing text
In the case where the high frequency inquiry parsing text for being greater than preset matching threshold value with degree, by that will parse in text and pre-set text set
Pre-set text between carry out similarity calculation, judge whether there is and parse similarity between text and be greater than default similar threshold value
Pre-set text, in turn, the similarity inquiry scene corresponding greater than the default pre-set text of similar threshold value between parsing text is made
For the inquiry scene for parsing text, summary, two ways combines the inquiry scene for determining parsing text, can accomplish guaranteeing
Under the premise of the accuracy of definitive result, the speed for determining inquiry scene is improved.
Fig. 4 is a kind of inquiry unit of information provided by the embodiments of the present application, comprising: receiving module 401, parsing module
402, determining module 403 and enquiry module 404.Wherein, receiving module 401, for receiving voice inquirement, voice inquirement is user
The voice of the inquiry content of transmission.Parsing module 402, the text for being converted to depending at least on vertical category feature to voice inquirement
Text resolution is carried out, parsing text is obtained, the category feature that hangs down is preset civil aviaton's Scene Semantics.Determining module 403, for determining solution
The inquiry scene of text is analysed, inquiry scene is to inquire content query service type affiliated in preset query service type.
Enquiry module 404, for the inquiry scene according to parsing text, the query result of inquiry parsing text.
Optionally, device further include: loading module 405, in the case where meeting default loading environment, load to meet
The Flight Information of preset condition is the Flight Information preloaded, and default loading environment includes: the letter for detecting and indicating that user logs in
The information that the Flight Information of preloading updates is ceased or receives, preset condition includes: in the trip Flight Information that user has subscribed,
And/or user is predetermined and the voice inquirement in the Flight Information of flight that provides, belong to preset high frequency query term
Information.
Enquiry module 404 inquires the query result of the parsing text, packet for the inquiry scene according to parsing text
Include: enquiry module 404 inquires scene, and pre-add specifically for belonging to preset personal information in the inquiry scene of parsing text
In the case where the query result that there is parsing text in the Flight Information of load, from the Flight Information of preloading, parsing text is obtained
This query result.
Optionally, enquiry module 404 are also used to be not present the query result of parsing text in the Flight Information of preloading
In the case where, from the Flight Information of preloading, obtain at least partly conditional information in the corresponding conditional information of parsing text.
After getting the corresponding conditional information of parsing text, according to the default corresponding relationship between inquiry scene and query interface, pass through
The corresponding query interface of inquiry scene of parsing text is called, the corresponding conditional information of inquiry scene for meeting parsing text is obtained
Information be the query result.
Optionally, 404 enquiry module, for the inquiry scene according to parsing text, inquiry parses the query result of text,
It include: enquiry module 404, specifically for the case where the inquiry scene of parsing text belongs to preset public information inquiry scene
Under, the corresponding conditional information of inquiry scene of parsing text is obtained, according to inquiry scene and the default corresponding pass between query interface
System, by calling the corresponding query interface of inquiry scene of parsed text, the inquiry scene that acquisition meets parsing text is corresponding
Conditional information information be query result.
Optionally, determining module 403, for determining the inquiry scene of parsing text, comprising: determining module 403 is specific to use
Parsing is inquired in there is high frequency of the matching degree between parsing text greater than preset matching threshold value in the first default corresponding relationship
In the case where text, the corresponding inquiry scene of parsing text will be inquired with high frequency, as the inquiry scene of parsing text, first pair
It should be related to the default corresponding relationship inquired between parsing text and preset inquiry scene for preset high frequency.
There is no the height that the matching degree between parsing text is greater than preset matching threshold value in preset first corresponding relationship
In the case that frequency inquires parsing text, judges to be greater than in the second corresponding relationship with the presence or absence of the similarity between parsing text and preset
The pre-set text of similar threshold value, the second corresponding relationship are default between each pre-set text in pre-set text set and inquiry scene
Corresponding relationship.
There is the similarity between parsing text in the second corresponding relationship and is greater than the pre-set text for presetting similar threshold value
In the case of, will inquiry scene corresponding with pre-set text, as parsing text inquiry scene.
Optionally, determining module 403 are also used to the phase being not present between parsing text in preset second corresponding relationship
In the case where the pre-set text for being greater than default similar threshold value like degree, call preset helper, helper output for for
User determines that the expression way of inquiry content provides the sentence of reference.
Optionally, device further include: conversion module 406 and playing module 407.Wherein, conversion module 406, for looking into
It askes inquiry scene of the module according to parsing text and query result is converted into voice after inquiry parses the query result of text.
Playing module 407, for playing the voice.
If function described in the embodiment of the present application method is realized in the form of SFU software functional unit and as independent production
Product when selling or using, can store in a storage medium readable by a compute device.Based on this understanding, the application is real
The part for applying a part that contributes to existing technology or the technical solution can be embodied in the form of software products,
The software product is stored in a storage medium, including some instructions are used so that a calculating equipment (can be personal meter
Calculation machine, server, mobile computing device or network equipment etc.) execute each embodiment the method for the application whole or portion
Step by step.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), with
Machine accesses various Jie that can store program code such as memory (RAM, Random Access Memory), magnetic or disk
Matter.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (14)
1. a kind of querying method of information characterized by comprising
Receive voice inquirement;The voice inquirement is the voice for the inquiry content that user sends;
Text resolution is carried out to the text that the voice inquirement is converted to depending at least on vertical category feature, obtains parsing text;Institute
Stating vertical category feature is preset civil aviaton's Scene Semantics;
Determine the inquiry scene of the parsing text;The inquiry scene is the inquiry content in preset query service type
In belonging to query service type;
According to the inquiry scene of the parsing text, the query result of the parsing text is inquired.
2. the method according to claim 1, wherein further include:
In the case where meeting default loading environment, the Flight Information that load meets preset condition is the Flight Information preloaded;
The default loading environment includes: the information or receive the letter that the Flight Information of preloading updates for detecting and indicating that user logs in
Breath;The preset condition includes: in the trip Flight Information that user has subscribed, and/or, user is predetermined and the voice inquirement
In in the Flight Information of flight that provides, belong to the information of preset high frequency query term;
The inquiry scene according to the parsing text, inquires the query result of the parsing text, comprising:
Belong to preset personal information inquiry scene, and the Flight Information of the preloading in the inquiry scene of the parsing text
It is middle there are in the case where the query result of the parsing text, from the Flight Information of the preloading, it is literary to obtain the parsing
This query result.
3. according to the method described in claim 2, it is characterized in that, the inquiry scene according to the parsing text, inquiry
The query result of the parsing text, further includes:
In the case where the query result of the parsing text is not present in the Flight Information of the preloading, from the preloading
Flight Information in, obtain at least partly conditional information in the corresponding conditional information of the parsing text;
After getting the corresponding conditional information of the parsing text, according to inquiry scene and the default corresponding pass between query interface
System obtains the inquiry field for meeting the parsing text by calling the corresponding query interface of inquiry scene of the parsing text
The information of the corresponding conditional information of scape is the query result.
4. the method according to claim 1, wherein the inquiry scene according to the parsing text, inquiry
The query result of the parsing text, comprising:
In the case where the inquiry scene of the parsing text belongs to preset public information inquiry scene, the parsing text is obtained
This corresponding conditional information of inquiry scene;
According to the default corresponding relationship between inquiry scene and query interface, by calling the inquiry scene of the parsing text corresponding
Query interface, obtain meet it is described parsing text the corresponding conditional information of inquiry scene information be the query result.
5. the method according to claim 1, wherein the inquiry scene of the determination parsing text, comprising:
The matching degree existed between the parsing text in the first default corresponding relationship is looked into greater than the high frequency of preset matching threshold value
In the case where asking parsing text, the corresponding inquiry scene of parsing text will be inquired with the high frequency, as the parsing text
Inquire scene;First corresponding relationship is default corresponding between preset high frequency inquiry parsing text and preset inquiry scene
Relationship;
There is no the matching degrees between the parsing text to be greater than the preset matching in preset first corresponding relationship
In the case where the high frequency inquiry parsing text of threshold value, judge in the second corresponding relationship with the presence or absence of the phase between the parsing text
It is greater than the pre-set text of default similar threshold value like degree;Second corresponding relationship be pre-set text set in each pre-set text with
Inquire the default corresponding relationship between scene;
The similarity existed between the parsing text in second corresponding relationship is greater than the pre- of the default similar threshold value
If in the case where text, will inquiry scene corresponding with the pre-set text, as it is described parsing text inquiry scene.
6. according to the method described in claim 5, it is characterized in that, the inquiry scene of the determination parsing text, is also wrapped
It includes:
In preset second corresponding relationship there is no it is described parsing text between similarity be greater than it is described preset it is similar
In the case where the pre-set text of threshold value, preset helper is called;The helper output is for being described in user determines
The expression way for inquiring content provides the sentence of reference.
7. the method according to claim 1, wherein being looked into the inquiry scene according to the parsing text
After the query result for asking the parsing text, further includes:
The query result is converted into voice;
Play the voice.
8. a kind of inquiry unit of information characterized by comprising
Receiving module, for receiving voice inquirement;The voice inquirement is the voice for the inquiry content that user sends;
Parsing module is obtained for carrying out text resolution to the text that the voice inquirement is converted to depending at least on vertical category feature
To parsing text;The vertical category feature is preset civil aviaton's Scene Semantics;
Determining module, for determining the inquiry scene of the parsing text;The inquiry scene is the inquiry content default
Query service type in belonging to query service type;
Enquiry module inquires the query result of the parsing text for the inquiry scene according to the parsing text.
9. device according to claim 8, which is characterized in that further include:
Loading module, in the case where meeting default loading environment, the Flight Information that load meets preset condition to be pre-add
The Flight Information of load;The default loading environment includes: the information or the boat for receiving preloading for detecting and indicating that user logs in
The information of class's information update;The preset condition includes: in the trip Flight Information that user has subscribed, and/or, user does not make a reservation for
And in the Flight Information of the flight provided in the voice inquirement, belong to the information of preset high frequency query term;
The enquiry module inquires the query result of the parsing text, packet for the inquiry scene according to the parsing text
It includes:
The enquiry module inquires scene specifically for belonging to preset personal information in the inquiry scene of the parsing text,
And in the Flight Information of the preloading there are it is described parsing text query result in the case where, from the flight of the preloading
In information, the query result of the parsing text is obtained.
10. device according to claim 9, which is characterized in that
The enquiry module is also used to be not present the query result of the parsing text in the Flight Information of the preloading
In the case of, from the Flight Information of the preloading, obtain at least partly item in the corresponding conditional information of the parsing text
Part information;
After getting the corresponding conditional information of the parsing text, according to inquiry scene and the default corresponding pass between query interface
System obtains the inquiry field for meeting the parsing text by calling the corresponding query interface of inquiry scene of the parsing text
The information of the corresponding conditional information of scape is the query result.
11. device according to claim 8, which is characterized in that the enquiry module, for according to the parsing text
Scene is inquired, the query result of the parsing text is inquired, comprising:
The enquiry module, specifically for belonging to preset public information inquiry scene in the inquiry scene of the parsing text
In the case of, obtain the corresponding conditional information of inquiry scene of the parsing text;
According to the default corresponding relationship between inquiry scene and query interface, by calling the inquiry scene of the parsing text corresponding
Query interface, obtain meet it is described parsing text the corresponding conditional information of inquiry scene information be the query result.
12. device according to claim 8, which is characterized in that the determining module, for determining the parsing text
Inquire scene, comprising:
The determining module is greater than specifically for there is the matching degree between the parsing text in the first default corresponding relationship
It, will inquiry field corresponding with high frequency inquiry parsing text in the case where the high frequency inquiry parsing text of preset matching threshold value
Scape, the inquiry scene as the parsing text;First corresponding relationship is preset high frequency inquiry parsing text and presets
Inquiry scene between default corresponding relationship;
There is no the matching degrees between the parsing text to be greater than the preset matching in preset first corresponding relationship
In the case where the high frequency inquiry parsing text of threshold value, judge in the second corresponding relationship with the presence or absence of the phase between the parsing text
It is greater than the pre-set text of default similar threshold value like degree;Second corresponding relationship be pre-set text set in each pre-set text with
Inquire the default corresponding relationship between scene;
The similarity existed between the parsing text in second corresponding relationship is greater than the pre- of the default similar threshold value
If in the case where text, will inquiry scene corresponding with the pre-set text, as it is described parsing text inquiry scene.
13. device according to claim 12, which is characterized in that
The determining module is also used to be not present in preset second corresponding relationship similar between the parsing text
In the case that degree is greater than the pre-set text of the default similar threshold value, preset helper is called;The helper output
For determining that the expression way of the inquiry content provides the sentence of reference for user.
14. device according to claim 8, which is characterized in that further include:
Conversion module inquires the parsing text for the inquiry scene in the enquiry module according to the parsing text
After query result, the query result is converted into voice;
Playing module, for playing the voice.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910682736.4A CN110399462B (en) | 2019-07-26 | 2019-07-26 | Information query method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910682736.4A CN110399462B (en) | 2019-07-26 | 2019-07-26 | Information query method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110399462A true CN110399462A (en) | 2019-11-01 |
CN110399462B CN110399462B (en) | 2022-03-04 |
Family
ID=68326129
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910682736.4A Active CN110399462B (en) | 2019-07-26 | 2019-07-26 | Information query method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110399462B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111680136A (en) * | 2020-04-28 | 2020-09-18 | 平安科技(深圳)有限公司 | Method and device for matching spoken language and semantics |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120130967A1 (en) * | 2010-11-18 | 2012-05-24 | Microsoft Corporation | Classification of transactional queries based on identification of forms |
CN108491443A (en) * | 2018-02-13 | 2018-09-04 | 上海好体信息科技有限公司 | By computer-implemented and method and computer system of user session |
CN108733756A (en) * | 2018-04-11 | 2018-11-02 | 北京三快在线科技有限公司 | Data preload method, apparatus, electronic equipment and readable storage medium storing program for executing |
CN108777142A (en) * | 2018-06-05 | 2018-11-09 | 上海木木机器人技术有限公司 | A kind of interactive voice recognition methods and interactive voice robot based on airport environment |
CN108897848A (en) * | 2018-06-28 | 2018-11-27 | 北京百度网讯科技有限公司 | Robot interactive approach, device and equipment |
CN109815314A (en) * | 2019-01-04 | 2019-05-28 | 平安科技(深圳)有限公司 | A kind of intension recognizing method, identification equipment and computer readable storage medium |
-
2019
- 2019-07-26 CN CN201910682736.4A patent/CN110399462B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120130967A1 (en) * | 2010-11-18 | 2012-05-24 | Microsoft Corporation | Classification of transactional queries based on identification of forms |
CN108491443A (en) * | 2018-02-13 | 2018-09-04 | 上海好体信息科技有限公司 | By computer-implemented and method and computer system of user session |
CN108733756A (en) * | 2018-04-11 | 2018-11-02 | 北京三快在线科技有限公司 | Data preload method, apparatus, electronic equipment and readable storage medium storing program for executing |
CN108777142A (en) * | 2018-06-05 | 2018-11-09 | 上海木木机器人技术有限公司 | A kind of interactive voice recognition methods and interactive voice robot based on airport environment |
CN108897848A (en) * | 2018-06-28 | 2018-11-27 | 北京百度网讯科技有限公司 | Robot interactive approach, device and equipment |
CN109815314A (en) * | 2019-01-04 | 2019-05-28 | 平安科技(深圳)有限公司 | A kind of intension recognizing method, identification equipment and computer readable storage medium |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111680136A (en) * | 2020-04-28 | 2020-09-18 | 平安科技(深圳)有限公司 | Method and device for matching spoken language and semantics |
CN111680136B (en) * | 2020-04-28 | 2023-08-25 | 平安科技(深圳)有限公司 | Method and device for semantic matching of spoken language |
Also Published As
Publication number | Publication date |
---|---|
CN110399462B (en) | 2022-03-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110413988B (en) | Text information matching measurement method, device, server and storage medium | |
CN110347863B (en) | Speaking recommendation method and device and storage medium | |
WO2017206661A1 (en) | Voice recognition method and system | |
US10803867B2 (en) | Dialogue system and domain determination method | |
US20060235694A1 (en) | Integrating conversational speech into Web browsers | |
CN109429522A (en) | Voice interactive method, apparatus and system | |
JP2017505964A (en) | Automatic task classification based on machine learning | |
CN107844470B (en) | Voice data processing method and equipment thereof | |
CN106407198A (en) | Question and answer information processing method and device | |
CN108028044A (en) | The speech recognition system of delay is reduced using multiple identifiers | |
CN110956955B (en) | Voice interaction method and device | |
US11935524B1 (en) | Authentication of impaired voices | |
CN109003600B (en) | Message processing method and device | |
US8301452B2 (en) | Voice activated application service architecture and delivery | |
WO2023273776A1 (en) | Speech data processing method and apparatus, and storage medium and electronic apparatus | |
CN110399462A (en) | A kind of querying method and device of information | |
CN106356056A (en) | Speech recognition method and device | |
JP5185807B2 (en) | Voice search device, voice search method, and voice search program | |
JP6233867B2 (en) | Dictionary registration system for speech recognition, speech recognition system, speech recognition service system, method and program | |
CN115905497B (en) | Method, device, electronic equipment and storage medium for determining reply sentence | |
US20190179902A1 (en) | Systems and methods for task automation using natural language processing | |
CN111694884B (en) | Intelligent government affair request processing method based on big data | |
CN109002498A (en) | Interactive method, device, equipment and storage medium | |
CN114254088A (en) | Method for constructing automatic response model and automatic response method | |
CN111968630B (en) | Information processing method and device and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |