Specific embodiment
In order to make those skilled in the art more fully understand the technical solution in this specification, below in conjunction with this explanation
Attached drawing in book embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described
Embodiment be only this specification a part of the embodiment, instead of all the embodiments.The embodiment of base in this manual,
Every other embodiment obtained by those of ordinary skill in the art without making creative efforts, all should belong to
The range of this specification protection.
In view of existing order generation method is mechanically to respond user's operation mostly, shows to user and set in advance step by step
The order scheme of classes set determines user's to obtain and the selection operation according to user to option in order scheme of classes
Carte content.For the above method when realizing, user cannot freely go tissue, input to meet oneself according to oneself speech habits
Individual requirement carte content, and the order scheme of classes that can only passively go reception system to be showed, and need successively to exist
In each order scheme of classes to respective selection carry out selection operation come complete to single object and customization require setting (such as
Set-point single object is caffe american, it is desirable that not sugaring, not on the rocks etc.).Therefore, it is based on existing order generation method, point is single
Process is not smart enough, implements comparatively trouble, point single-action rate for a user and is relatively low.
For the basic reason for generating the above problem, this specification consideration can be provided for user a kind of permission user according to
The speech habits of oneself are freely organized, input to include the point single object of user and/or the voice point forms data of customization requirement, clothes
The available and natural language point forms data according to user of business device, by matching preset order scheme of classes, with automatically true
Point list information entrained in user speech point forms data is made, and then is quickly generated the order for meeting user's requirement.In this way
User can not have to the order scheme of classes for passively going to receive, be arranged various complexity again, but certainly according to the speech habits of itself
By, easily comprising the natural language point forms data of a little single information to complete point single for tissue.To solve existing method presence
The problem of user's point list is inconvenient, poor user experience, reach the natural language point list that can be inputted based on user's self-organization
Data automatically identify point list information entrained by the natural language point forms data of user, and then generate and meet user's requirement
The effect of order.
Based on above-mentioned thinking, this specification embodiment provides a kind of order generation method, and the order generation method is specific
Can be applied to it is a kind of include: server, the first client, the second client system framework in, can refering to fig. 1 shown in.
Above-mentioned first client, the second client can be respectively by wired or wireless modes and Coupled processors, to carry out data
Interaction.
Wherein, above-mentioned first client can be understood as a kind of end for being applied to user (or consumer, customer etc.) side
End equipment.The equipment can be used for showing the data input screen for receiving the natural language point forms data of user to user;
The natural language point forms data of the user of used above-mentioned input interface input is received, and the natural language point forms data is sent
To server;And then it can receive trade company to Confirmation of Orders information, wherein the order is server according to the natural language
Speech point forms data generates.
Specifically, above-mentioned first client can be that user uses with certain data-handling capacity, communication function
Entity electronic equipment.For example, first client can be desktop computer, tablet computer, laptop, smart phone,
Digital assistants, intelligent wearable device, shopping guide's terminal, television set with network access functions etc..Alternatively, first client
End is also possible to the software application that can be run in above-mentioned electronic equipment.For example, first client can also be outside certain
Sell app, certain app that orders etc..
Above-mentioned second client can be understood as a kind of terminal for being applied to trade company (or service, goods providers) side
Equipment.The equipment can be used for receiving order, wherein the order is that server is raw according to the natural language point forms data of user
At;The order is shown to trade company;Trade company is received to the Confirmation of Orders operation data, and by the confirmation operation data
It is sent to server.
Specifically, second client is similar with first client, and can be that trade company uses has a fixed number
According to processing capacity, the entity electronic equipment of data communication function, it is also possible to the software that can be run in above-mentioned electronic equipment
Using.
Above-mentioned server can be understood as it is a kind of be applied to platform side (such as take out platform, platform of ordering etc.) be responsible for into
The equipment of row order processing.The equipment can be used for obtaining the natural language point forms data of the user from the first client;And
Word cutting processing is carried out to the natural language point forms data, and according to word cutting treated natural language point forms data, is obtained more
A keyword;The multiple keyword is matched with preset order scheme of classes, obtains the natural language point forms data
Entrained point list information;According to point list information entrained by the natural language point forms data, order is generated;Again to the second visitor
Family end sends order generated;The trade company from the second client is received to Confirmation of Orders information, and by the trade company pair
Confirmation of Orders information is sent to the first client, to show trade company the point single object of clear user and customization to user
It is required that determining order.
Specifically, the server can be that by the Batch Processing service of the functions such as semantics recognition, data processing
Device.The service implement body can be a kind of electronic equipment with data operation, store function and network interaction function;?
It can be operate in the electronic equipment, provide the software of support for data processing, storage and network interaction.In present embodiment
In do not limit the quantity of the server specifically.The server can be a server, can also be several servers,
Alternatively, the server cluster that several servers are formed.
It, can be as shown in fig.1, by above system frame, user A be can be convenient, efficiently in a Sample Scenario
Ground uses mobile phone freely to be organized as the first client according to the speech habits of oneself, inputs comprising having a single object and customization
It is required that voice data or text data as natural language point forms data, it is single efficiently advantageously to complete point;Server then can be with
According to the first client received user natural language point forms data generate meet user requirement order;Trade company then can be with
Computer is received as the second client, shows that server point list according to entrained by the natural language of user point forms data is believed
Order generated is ceased, and is specifically handled according to user's order contents.
Specifically, user A wants sending outside for the caffe american for selecting one glass of month Bark on the net, and wants in the side user A
Seeking Truth not sugaring, big coffee not on the rocks.To input the voice data of user's A self-organization as natural language point forms data
For.The moon Bark that user A can be clicked to enter in mobile phone (i.e. the first client) sends interface outside, refering to shown in Fig. 2.It clicks
The above-mentioned voice input icon sent outside in interface, triggering mobile phone start to acquire the voice data of user.User A can be against mobile phone
Microphone directly says the carte content of self-organization: " I want to have a drink mug not sugaring the caffe american of ice is not added ".Wherein,
Above-mentioned carte content includes that the desired personalized customization for selecting single object (i.e. " coffee ") and user of user A requires (i.e.
" one glass of (quantity) ", " not sugaring ", " not on the rocks ", " American ") mobile phone acquires natural language of the above-mentioned voice data as user A
Point forms data, and the natural language point forms data is sent to send outside by cell phone network and is responsible on platform (such as X group takes out)
The server of order processing.The order generation method that user A in this way is provided by this specification embodiment can according to oneself
Speech habits freely organize to have input include carte content and multiple personal the desired voice data of customization be used as from
Right language point forms data, thus efficiently, be readily achieved point single operation.
Certainly, it should be noted that above-mentioned cited user's mode by voice input set self-organization
Comprising having a single object and the voice data of personalized customization requirement as natural language point forms data, only one kind is schematically said
It is bright.When it is implemented, user can also speech habits according to oneself, operating habit, select self-organization edit package containing a little
The text data that single object and personalized customization require is completed a little in a manner of text input as natural language point forms data
It is single.In this regard, this specification is not construed as limiting.
In server-side, refering to Fig. 3.Natural language point forms data of the server in the user A for receiving mobile phone transmission
Afterwards, word cutting processing first can be carried out to the natural language point forms data, the natural language point forms data of user A is decomposed into following
Multiple language construction units: " I ", " wanting to drink ", " one glass ", " mug ", " not sugaring ", " not on the rocks ", " caffe american ".It is right again
Above-mentioned multiple language construction units carry out semantics recognition, obtain multiple keywords.For example, available following keyword:
" one glass ", " mug ", " not sugaring ", " not on the rocks ", " caffe american ".
In turn, server can be by the spy of the classification option in obtained multiple keywords and preset order scheme of classes
Sign word is matched.Specifically, server can be according to the current key word in multiple keywords, in preset order class table
The Feature Words of classification option retrieved, filter out from the Feature Words of preset classification option and matched with current key word
The Feature Words of the classification option of (the i.e. represented semantic difference degree with current key word is less than threshold level) are closed as current
The matching result of keyword.
For example, current key word is " not sugaring ".Server can be preset in order scheme of classes to being previously stored in
It is retrieved in the Feature Words of classification option, determines that classification option corresponding with the keyword is " sugariness option ".Further,
By semantic analysis, spy is determined from the Feature Words of multiple next stage classification options under above-mentioned " sugariness option " classification option
Sign word " sugar-free " semanteme for being characterized be with the keyword " not sugaring " unanimously, determine successful match, and then can will be preset
The Feature Words " sugar-free " of classification option are determined as the Feature Words with the matched classification option of current key word " not sugaring ", and will
" sugar-free " is determined as the matching result of " not sugaring ".
Wherein, it should be noted that the Feature Words of the classification option in above-mentioned preset order scheme of classes specifically can be
Trade company, according to provided commodity or the specific feature of service, has been previously set, and be uploaded to service by the second client
Device forms corresponding order scheme of classes by server, and is stored in storage equipment, to make when subsequent progress Keywords matching
With.
For example, can multiple classifications first be arranged for the commodity of this larger classification of coffee on a second client in certain cafe
The Feature Words of option, such as: " Blue Mountain Coffee ", " Ma Qiduo coffee ", " caffe american ", " espresso coffee " etc..Wherein, above-mentioned every
The Feature Words of one classification option are corresponding with a type of coffee provided by the cafe.Further, in order to full
The classification of multiple next stage can also be arranged in the personalized customization requirement of sufficient user, the cafe for the coffee of each type
Option.Such as be also provided with sugariness option, temperature option, cup type option etc..Wherein, in the classification option of above-mentioned next stage
The Feature Words of multiple specific classification options are respectively separately included again.By taking sugariness option as an example, the Feature Words of sugariness option can
To include: " sugar-free ", " three points of sugar ", " five points of sugar ", " seven points of sugar ", " very sugared " etc..Wherein, the feature of each sugariness option
Word is all corresponding with a kind of sugariness option type.The above-mentioned setting to classification option that second client can acquire the cafe is grasped
Make data, and the setting operation data of above-mentioned classification option is sent to server.Server is received from the second client
Classification option setting operation data after, the Feature Words of corresponding classification option can be generated according to above-mentioned data, and formed
Or more new order scheme of classes, in the database for subsequent calls by the storage of the above order scheme of classes.
According to the matching way similar with above-mentioned identification, the Feature Words for the type option for determining keyword " not sugaring ", clothes
Business device can match each keyword in multiple keywords with the Feature Words of preset classification option respectively, obtain more
The matching result of a keyword.For example, available: the matching result of keyword " one glass " is " one glass ", keyword " mug "
Matching result be " mug ", the matching result of keyword " not sugaring " is " sugar-free ", the matching knot of keyword " caffe american "
Fruit is " caffe american ", and the matching result of keyword " not on the rocks " is " no ice ".
Then, server can combine the matching result for multiple keywords that above-mentioned matching obtains, as including user
Point single object and customization require user natural language point forms data entrained by point list information, and then it is subsequent can basis
Above-mentioned point list information generates the corresponding user's order for meeting user's requirement.
Specifically, server can by the matching result " one glass " of obtained multiple keywords, " mug ", " sugar-free ",
" caffe american ", " no ice " are combined according to syntax logic, obtain the first character group understood convenient for people, such as pass through combination
Available first character group is " one glass of mug sugar-free is without Iced CaféAmericano ".Wherein, above-mentioned first character group can be accurately
The single object of selecting for symbolizing user A is that coffee and user A are: quantity to the multiple personalization customization requirement of the single object
It is one glass, cup type is mug, and sugariness is sugar-free, and temperature is no ice, and type of coffee is caffe american.Further, server can
To generate the order of user A according to above-mentioned first character group, wherein the order includes the first character group.The order is led to again
It crosses network and is sent to the second client for being set to moon Bark side.Server can be provided by this specification embodiment in this way
Order generation method accurately have identified user A self-organization setting natural language point forms data entrained by point list
Information, and according to above-mentioned point list information generate can accurate characterization go out the point single object of user A and personalized customization requirement is ordered
It is single.
In this example, it is also necessary to explanation carries out word cutting processing in natural language point forms data of the server to user,
It is subsequent to invalid language in order to reduce when carrying out semantics recognition to multiple language construction units after obtaining multiple language construction units
The processing of adopted information improves treatment effeciency, can first be deleted according to the recognition result of semantics recognition only comprising invalid semantic information
The language construction unit of (i.e. content requires unrelated with point single object and/or for the customization of single object).For example, language construction
Unit " I ", " wanting to drink " can determine the letter that above-mentioned two language construction unit is included according to the recognition result of semantics recognition
Breath is all invalid semantic information, can first reject above-mentioned two language construction unit, i.e., subsequent not to upper two language construction lists
Member carries out the acquisition of keyword.
In moon Bark side, the moon, the staff of Bark can receive server hair by computer (i.e. the second client)
The order of the user A sent.Also, computer can extract the first character group from the order of user A, in the exhibition of computer display screen institute
The order shown shows the particular content for showing above-mentioned first character group in interface to the staff of moon Bark.
For example, refering to Fig. 4.The prompt information of new order can be shown on computer screen.The moon, the staff of Bark infused
It anticipates to after the prompt information, can show interface into order information by clicking this prompt information.It is shown in order information
In interface, there are phone number, the address location of user A in addition to showing, and outside the essential informations such as " caffe american ", " one glass ",
Also show that include the first character group content customization requirement, i.e., " customer requirement: one glass of mug sugar-free is without Iced CaféAmericano ".
And then the staff of moon Bark accurately can be ready to meet user more according to the point list information of above-mentioned user A for user A
The caffe american that kind personalized customization requires.Trade company month Bark can be generated by the order that this specification embodiment provides in this way
Method in addition to can obtain relatively simple, basic user select list information (such as caffe american, one glass etc.) other than, moreover it is possible to it is accurate
Ground gets the more complicated personalized customization requirement of user A, more so as to accurately be that user A is ready to meet user A
The coffee that kind customization requires.
By above-mentioned Sample Scenario as it can be seen that this specification provide order generation method, by allow user according to oneself
The input of speech habits self-organization meets natural language point forms data (including voice data and the textual data of oneself speech habits
According to), and the keyword in the natural language point forms data of user is obtained to match with preset order scheme of classes, with determination
Point list information entrained by the natural language point forms data of user out, and then corresponding user's order is automatically generated, to solve
The problem of point of user present in existing method list inconvenience, poor user experience, reaching can be inputted based on user's self-organization
Natural language point forms data, automatically identify user point single object and/or customization require, generate meet user requirement order
Single effect.
In a Sample Scenario, after server generates corresponding order according to the natural language point forms data of user A,
Before order is sent to trade company by server, order can be first sent to the first client by server.First client is receiving
To after the order, order can be shown in the first client.The order interface that user can be shown by the first client
In particular content confirmation server order generated whether accurately given expression to the order information oneself being intended by.Such as
Fruit user determines that the order that server generates is the order for meeting oneself requirement, and user A can show interface by clicking order
In the icon that confirms an order the order is confirmed.First client can receive user and indicate the Confirmation of Orders, and
User is sent to server to the confirmation instruction of the above order.Server is receiving user to Confirmation of Orders instruction
Afterwards, then by the order it is sent to the second client.If the user feels that the order inaccuracy that server generates, at this moment can click
Order shows the resetting order icon in interface, and then natural language point forms data can be re-entered by the first client,
So that server regenerates order.
In another Sample Scenario, server is obtained to the natural language point forms data voluntarily inputted based on user A
Multiple keywords carry out Keywords matching when, in determining the preset order scheme of classes be not present and the keyword
In the case where the Feature Words for the classification option matched, it can determine whether that the Keywords matching is unsuccessful, and the keyword is labeled as " to true
It is fixed ".
In the present embodiment, the keyword of successful match has corresponding matching result, i.e., with the class of the Keywords matching
The Feature Words of mesh option.And to unsuccessful keyword is matched, server does not make an amendment to the content of keyword, is only arranged one
A data label " to be determined " is to distinguish the keyword of successful match.Again by the matching result of multiple keywords, and mark
It is keyword to be determined together as point list information entrained by natural language point forms data.
Further, subsequent when generating order, the first character group can be generated according to the matching result of multiple keywords,
In, first character group is used to characterize the point single object of user's requirement, and/or, customization requirement of the user for point single object;
According to keyword to be determined is labeled as, the second character group is generated, wherein second character group is for characterizing user for point
The customization to be determined of single object requires (i.e. server is based on the unrecognized customization requirement of preset order scheme of classes);According to
Above-mentioned first character group and the second character group generate corresponding user's order, are sent to the second client.Server will be upper in this way
The customization that stating can not determine requires corresponding keyword not make an amendment, and directly passes through user's order in the form of the second character group
It is sent to trade company, determination is voluntarily identified according to knowledge, the experience of industry by trade company, avoids the something lost required the customization of user
Leakage.
Certainly, it should be noted that it is above-mentioned it is cited will match unsuccessful keyword be labeled as it is to be determined, only one
Kind schematically illustrates.When it is implemented, can also be labeled in other way, such as can also will match unsuccessful
Keyword is labeled as " can not identify ", " can not determine " etc., as long as can distinguish with successful keyword is identified.It is right
It is not construed as limiting in specific notation methods this specification.
For example, trade company month Bark is previously stored in ordering for server in advance that temperature classification option is not arranged for coffee
There is no the Feature Words of corresponding classification option to match in the Feature Words of classification option in single scheme of classes with keyword " not on the rocks ".
Therefore, it is based on existing preset order scheme of classes, can only obtain the matching result of other keywords, it may be assumed that keyword " one glass "
Matching result be " one glass ", the matching result of keyword " mug " is " mug ", and the matching result of keyword " not sugaring " is
" sugar-free ", the matching result of keyword " caffe american " are " caffe american ".For keyword " not on the rocks ", due to can not match
Function, server do not deal with the keyword, and the keyword is only labeled as " to be determined ".Again by above-mentioned multiple successful match
Keyword matching result and be labeled as entrained by the natural language point forms data of keyword to be determined together as user
Point list information, to be subsequently generated corresponding order.
When specifically generating user's order, server can be pressed the matching result of the keyword of above-mentioned multiple successful match
It is combined according to syntax logic, obtains the first character group;The keyword for being labeled as to be determined is combined, the second character is generated
Group;User's order is generated according to above-mentioned first character group and the second character group;User's order is sent to the second client again.It needs
It is noted that above-mentioned user's order not only include server it is identified determine be recorded in user in the first character group
The point single object of A and part customization require, and determining be recorded in the second character group can not be identified by further comprising server at present
User A another part customize require.
In moon Bark side.Computer is after receiving the order of user A of server generation, in addition to showing the first character group
The content that is characterized of information outside, can also detect in user's order with the presence or absence of the second character group, determine that the user orders
There are after second character group in list, second character group is extracted, and show in interface and open up in the order that computer is shown
The related content of second character group is shown.
For example, can be refering to Fig. 5.The computer of month Bark is showing the customization requirement based on the first character group content, i.e.,
While " customer requirement: one cup of mug sugar-free caffe american ", can also near above-mentioned customer requirement, in a manner of remarks,
Being shown with the shallower box of color can not be identified in the second determining character group based on preset order classification list server
Hold, i.e., " requirement to be determined: not on the rocks ".Although can not identify determination based on existing preset order classification list server in this way
The specific requirement that second character group is characterized out, but keyword included by second character group can directly be transmitted and be showed
Trade company.Trade company can be based on domain knowledge and experience, voluntarily identify, determine that the user to be characterized in the second character group wants
It asks.It in this example, can be by user although server can not be matched to the Feature Words of classification option corresponding with " not on the rocks "
Natural language point forms data in the keyword that can not determine the staff of moon Bark is presented directly to as the second character group,
In this way the moon Bark staff can be based on the requirement to be determined shown, it is known that user A is in addition to requiring " one glass of mug sugar-free
Without Iced CaféAmericano " outside, it is claimed as well " not on the rocks ", and then can accurately prepare the coffee for meeting user's requirement for user A.
In another Sample Scenario, server can also obtain trade company by the second client send to classification option
Operation data is set.Further according to the trade company to the setting operation data of classification option, generation includes the feature of classification option
The new order scheme of classes of word, or to being updated in existing order scheme of classes.And then when subsequent progress Keywords matching,
The Feature Words that can use the classification option in updated order scheme of classes carry out corresponding Keywords matching.Wherein, above-mentioned
At least one of can specifically include to the update of order scheme of classes: increasing the feature of the classification option in order scheme of classes
Word, the Feature Words for deleting classification option in order scheme of classes, the Feature Words of classification option etc. in modification order scheme of classes.When
So, it should be noted that above-mentioned cited the update to order scheme of classes to be intended merely to this specification reality is better described
Apply mode.When it is implemented, can also requirement according to trade company and specific application scenarios, introduce other kinds of update side
Formula.In this regard, this specification is without limitation.
For example, certain trade company increases the new taste option of fried chicken: plum taste can be shown on a second client
It modifies in the set interface of classification option to the Feature Words of the classification option in the taste option of fried chicken.Specifically, can be with
The Feature Words for increasing new classification option are clicked under the taste option of fried chicken, and input the Feature Words of newly-increased taste option:
Plum taste.Second client can acquire trade company and operate to the setting operation data of the taste option of fried chicken, and by above-mentioned setting
Data are sent to server.Server can increase after receiving above-mentioned setting operation data according to above-mentioned setting operation data
Add the Feature Words of the classification option in existing order scheme of classes.Fried chicken taste option i.e. in corresponding order scheme of classes
Increase the Feature Words of " plum taste " as new fried chicken taste option in Feature Words.
Further, spy that can targetedly to the classification option in existing order scheme of classes in order to facilitate trade company
Sign word is modified, and the second client, can be first from preset time period after receiving trade company to the setting operation of classification option
In user's order of (such as in nearest one week), the keyword of the second phrase is extracted, and by the key of extracted second phrase
Word shows user.Above-mentioned second phrase is often spy of the server based on the classification option in existing default order scheme of classes
User's requirement that sign word can not be identified, be determined.Trade company can be according to the keyword of provided second phrase, targetedly in this way
Ground modifies setting to the Feature Words of original classification option, so that server can be based on modifying updated order classification
The Feature Words of classification option in table more accurately identify, determine the customization requirement of user.
In another Sample Scenario, in order to guide user organize as far as possible input is effective, server can be more compared with subject to
The natural language point forms data for really identifying, determining.In user side, input operation of first client in response user, acquisition
Before the natural language point forms data of user, preset leading question first can also be shown to user.Wherein, the leading question is used for
User is guided to input satisfactory natural language point forms data, wherein the satisfactory natural language point forms data is
It include the natural language point forms data with the preset matched keyword of order scheme of classes.Specifically, above-mentioned leading question can be with
It include the Feature Words etc. of the classification option in the preset order scheme of classes.Wherein, the particular content of above-mentioned leading question can
To be to be arranged by trade company by the second client self-organization, the feature of classification option can also be set according to trade company by server
Word and provided commodity and the specific feature of service automatically generate.
For example, refering to Fig. 6.User A can click to enter on mobile phone the moon Bark when sending interface outside, send outside on interface
A plurality of prompt information can first be popped up as leading question, determining include to guide user A to input server as far as possible and can identify
There is the natural language point forms data of effective information.Specifically, following prompt information can be popped up in the dialog as guidance
Language " joyous drink presence moon Bark, to give me a cup of coffee ", " small shop manager special recommendation Blue Mountain Coffee, caffe american ", " parent, you like
It is joyous which kind of? ", " sugar-free is strong, and five points of sugar are palatable, very sugared sweetness ", " parent, which kind of sugariness is suitble to you? ".By in user
Before input point forms data, show that above-mentioned leading question can effectively guide user to organize input that can accurately identify determination as far as possible
, the natural language for including effective information point forms data.Meanwhile it also playing and providing the reference letter of a list for user's point list
Trade company's commodity are carried out the effect of appropriate promotional, further increase user experience by breath.
By above-mentioned Sample Scenario as it can be seen that the order generation method that this specification provides, defeated by permission user's self-organization
Enter to meet the natural language point forms data of oneself speech habits, and obtains the keyword in the natural language point forms data of user and come
It is matched with preset order scheme of classes, to determine point list information entrained by the natural language point forms data of user, into
And corresponding user's order is automatically generated, so that it is inconvenient, poor user experience to solve the point of user present in existing method list
Problem is reached the natural language point forms data that can be inputted based on user's self-organization, automatically identifies the point single object of user
And/or customization requires, and generates the effect for meeting the order of user's requirement;In server based on the class in existing order scheme of classes
The situation that the content that the Feature Words of mesh option characterize Partial key word can not determine will be unable to determining keyword as
Two character groups feed back to trade company together with the first character group successfully determined together, so that trade company is according to domain knowledge, warp
It tests and voluntarily identifies that user to be determined requires, avoid the omission that the system fixed to user requires;Further, also allow trade company to class
Mesh option is configured, and according to trade company to the classification option in the setting operation data more new order scheme of classes of classification option
Feature Words improve the accuracy that server determines the single information of point entrained by the natural language point forms data of user, further mention
High user experience.
Refering to Fig. 7, this specification embodiment provides a kind of order generation method, wherein this method is applied particularly to put down
Platform server-side.When it is implemented, this method may include the following contents:
S71: the natural language point forms data of user is obtained;
S72: word cutting processing is carried out to the natural language point forms data, and according to word cutting treated natural language point list
Data obtain multiple keywords;
S73: the multiple keyword will be matched with preset order scheme of classes, and will obtain the natural language point list
Point list information entrained by data;
S74: according to point list information entrained by the natural language point forms data, order is generated.
In the present embodiment, the order generation method that this specification embodiment provides specifically can be applied to platform side
Responsible order processing server.
Wherein, platform applied by the above order generation method specifically can be shopping platform (such as X is precious), send platform outside
(such as certain group takes out), the pre- fixed platform of travelling etc. are also possible to the platform of ordering used inside trade company, reservation platform etc..Specifically
When implementation, user can be completed by above-mentioned platform on line to desired purchase or the selection of scheduled commodity, service, and be arranged
Corresponding requirements complete point single operation;Platform Server can according to the user's choice with require to generate corresponding order, and will order
Single-shot gives trade company;Trade company can receive order by platform, and be that user is ready to meet user's requirement according to order contents
Commodity or service, complete user's order.
Certainly, it should be noted that above-mentioned cited Platform Type is intended merely to that this specification implementation is better described
Mode.When it is implemented, can also as the case may be and scene requirement, the order generation side that this specification embodiment is provided
Method promotes the use of other kinds of scene platform.In this regard, this specification is not construed as limiting.
In the present embodiment, above-mentioned natural language point forms data be specifically understood that for by user according to oneself language
Habit, freely organize, input comprising have a single object and/or for single object customization require point forms data.Its
In, above-mentioned natural language point forms data specifically can be voice data, be also possible to text data.For above-mentioned natural language point
The concrete form of forms data, this specification are not construed as limiting.
In the present embodiment, the natural language point forms data for obtaining user can specifically refer to that server receives first
The natural language point forms data of client (such as mobile phone, laptop etc.) acquisition and the user's self-organization input sent.
When it is implemented, user can directly input self-organization by the first client according to oneself speech habits
Voice data as natural language point forms data.For example, user can directly say the demand of oneself against mobile microphone:
" I wants a super peppery shredded pork with green pepper portioned rice bowl, to add that of meat ".Mobile phone can acquire the above-mentioned voice data of user,
And using above-mentioned voice data as the natural language of user point forms data, certain service for taking out platform is sent to by cell phone network
Device.To realize that server receives the natural language point forms data of the user from the first client.
In the present embodiment, above-mentioned keyword can be understood as wanting the commodity obtained or service for describing user
The vocabulary of content (i.e. point single object) and attributive character.For example, corresponding keyword may include milk tea for milk tea
This noun itself, and the vocabulary for describing milk tea attributive character specifically can be related with sugariness vocabulary (such as nothing
Sugar, three points of sugar, five points of sugar etc.), vocabulary related with content (such as plus coconut plus pearl plus red bean plus double-skin milk etc.),
Vocabulary (such as cuvette, middle cup, mug etc.) related with component or vocabulary related with temperature (such as room temperature, without ice, on the rocks
Deng) etc..Certainly, above-mentioned cited keyword is intended merely to that this specification embodiment is better described.When it is implemented, can
Other kinds of vocabulary is arranged as above-mentioned keyword according to the specific features of specific application scenarios and commodity or service.
In this regard, this specification is not construed as limiting.
In the present embodiment, above-mentioned that word cutting processing is carried out to the natural language point forms data, and handled according to word cutting
Natural language point forms data afterwards, obtains multiple keywords, specifically can be understood as server to the natural language point odd number
According to word cutting processing is carried out, multiple language construction units are obtained;Semantics recognition is carried out respectively to the multiple language construction unit, with
Obtain the multiple keyword.Wherein, it may include selection of the user to single object in above-mentioned multiple keywords, it can also be with
It include the requirement of the attributive character of selected single object of user, i.e. the customization requirement for point single object.
In the present embodiment, word cutting processing is carried out in natural language point forms data of the server to user, obtained multiple
It is subsequent to invalid semantic information in order to reduce when carrying out semantics recognition to multiple language construction units after language construction unit
Processing, improve treatment effeciency, can first be deleted according to the recognition result of semantics recognition only comprising invalid semantic information (i.e. content and
Point single object and/or require unrelated word for the customization of single object) language construction unit, or deletion language construction list
Invalid word in member, then semantics recognition is carried out to remaining language construction unit, to extract corresponding keyword.In this example,
When it is implemented, can be by interjection, modal particle, conjunction etc. as the invalid word for characterizing invalid language message.
Specifically, for example, server in the natural language of the user received selects forms data, " I wants a super peppery green pepper
Shredded meat portioned rice bowl, to add meat that " after, can first to natural language point forms data carry out word cutting processing, obtain multiple languages
Say structural unit, comprising: " I ", " wanting ", " portion ", " super peppery shredded pork with green pepper portioned rice bowl ", " that add meat ".Root again
According to the semanteme of each language construction unit, according to preset rules to the language construction unit or language for only including invalid semantic information
Invalid word in speech structural unit is deleted.It can first delete following invalid word or structural unit in this way: " I ",
" wanting ", " ", " that ".Again speech recognition is carried out to the language construction unit after the invalid word of rejecting respectively, obtains multiple keys
Word, respectively " portion ", " super peppery ", " shredded pork with green pepper portioned rice bowl ", " adding meat ".
In the present embodiment, the Feature Words of the classification option in above-mentioned preset order scheme of classes specifically can be trade company
According to the spy of the corresponding classification option of the attributive character sets itself of provided commodity or service content and commodity or service
Word is levied, and is uploaded to server, is stored in the storage equipment such as database by server, so as to the tune when carrying out Keywords matching
With.
For example, certain cafe can on the second client (such as trade company use computer) first for coffee this compared with major class
The Feature Words of multiple classification options are arranged in purpose commodity, such as: " Blue Mountain Coffee ", " Ma Qiduo coffee ", " caffe american ", " meaning
Formula coffee " etc..Wherein, the Feature Words of each above-mentioned classification option and a type of coffee pair provided by the cafe
It answers.Further, in order to meet the personalized customization requirement of user, which can also be directed to the coffee of each type
The classification option of multiple next stage is arranged in coffee again.Such as be additionally provided with sugariness option, temperature option, cup type option etc..Wherein, on
State in the classification option of next stage and separately include the Feature Words of multiple specific classification options.By taking sugariness option as an example, sweet tea
The Feature Words for spending option can include: " sugar-free ", " three points of sugar ", " five points of sugar ", " seven points of sugar ", " very sugared " etc..Wherein, often
The Feature Words of one sugariness option are all corresponding with a kind of sugariness type.Second client can acquire the above-mentioned to class of the cafe
The setting operation data of mesh option, and the setting operation data of above-mentioned classification option is sent to server, server can root
New order scheme of classes is generated according to above-mentioned setting operation data or existing order scheme of classes is updated.
In the present embodiment, point list information entrained by the point forms data of the natural language of above-mentioned user specifically can wrap
Personalized customization requirement containing selected single object of user and user to single object.Above-mentioned single object can be coffee
The commodity such as coffee, milk tea or service.Above-mentioned personalized customization requires to can be sugar-free, on the rocks etc..
In the present embodiment, above-mentioned to match the multiple keyword with preset order scheme of classes, obtain institute
Point list information entrained by the natural language point forms data of user is stated, when it is implemented, current in multiple keywords to match
It, can be by the classification option in the current key word and preset order scheme of classes in the multiple keyword for keyword
Feature Words carry out semantic matches, will be determined as current key word with the Feature Words of the matched classification option of the current key word
Matching result;The matching result of the multiple keyword is determined as point entrained by the natural language point forms data of the user
Single information.
When specifically being matched, can be examined in the Feature Words of preset classification option according to current key word
Rope is to determine whether that semantic or semantic difference degree consistent with current key word is less than the spy of the classification option of threshold level
Levy word.In the determining Feature Words existed with the semantic congruence of current key word or approximate classification option, successful match is determined,
And the Feature Words of the classification option are as the Feature Words with the matched classification option of current key word.
In the manner described above, to following multiple keywords: " portion ", " super peppery ", " shredded pork with green pepper portioned rice bowl ", " adding meat "
The matching for carrying out classification option feature word respectively, obtaining corresponding matching result, to be respectively as follows: " one bowl ", " Samsung is peppery " (most high
Grade it is peppery), " shredded pork with green pepper portioned rice bowl ", " adding meat ";Using above-mentioned matching result as the single information of point.
In the present embodiment, the above-mentioned point list information according to entrained by the natural language point forms data generates order,
When it is implemented, can be according to multiple in the single information of point entrained by the point forms data of natural language described in syntax logical combination
The matching result of keyword, to generate the first character group, wherein first character group is used to characterize the point single object of user
And/or the customization requirement for single object;Further according to first character group, user's order is generated.And then it can incite somebody to action
It include that user's order of the first character group is sent to trade company, so that trade company can be according to the first character group institute in user's order
The content of characterization determines selected single object of user, and/or requires for the personalized customization of single object, so as to
It is ready to meet commodity or the service of user's requirement for user.
For example, the matching result of multiple keywords is respectively as follows: " one in the single information of point entrained by natural language point forms data
Bowl ", " Samsung is peppery " (highest level peppery), " shredded pork with green pepper portioned rice bowl ", " adding meat ";By above-mentioned matching result according to syntax logic
It is combined, the first obtained character group can be expressed as " one bowl of Samsung is peppery to add meat shredded pork with green pepper portioned rice bowl ".Wherein, the first word
" shredded pork with green pepper portioned rice bowl " in symbol group is used to characterize the point single object of user, and " one bowl ", " Samsung is peppery ", " adding meat " are respectively used to
Characterize customization requirement of the user to multiple personalizations of shredded pork with green pepper portioned rice bowl.And then server can be according to above-mentioned first character
Group generates corresponding user's order, and is sent to trade company by network.
In one embodiment, above-mentioned to match the multiple keyword with preset order scheme of classes, it obtains
Point list information entrained by the natural language point forms data, when it is implemented, may include the following contents: according to lower section
Formula determines the matching result of the multiple keyword: will be in the current key word and order scheme of classes in the multiple keyword
The Feature Words of classification option matched, will be determined as currently with the Feature Words of the matched classification option of the current key word
The matching result of keyword;The matching result of the multiple keyword is determined as entrained by the natural language point forms data
The single information of point.
In the present embodiment, by the classification option in the current key word and order scheme of classes in the multiple keyword
Feature Words matched, of current key word will be determined as with the Feature Words of the matched classification option of the current key word
With as a result, when it is implemented, may include: the classification option according to current key word, in preset order class template table
It is retrieved in Feature Words to determine whether there is semantic or semantic difference degree consistent with current key word and be less than or equal to threshold
The Feature Words of the classification option of extent value.Determining the Feature Words existed with the classification option of the semantic congruence of current key word
When, successful match is determined, and the specific word is determined as the Feature Words with the matched classification option of current key word, by this feature
Word is determined as the matching result of current key word.It should be noted that the above-mentioned spy to carry out classification option to current key word
It illustrates how to carry out Keywords matching for the matching of sign word, the matching of other keywords in multiple keywords can be joined
It reads above with respect to the matched description of current key word, this specification repeats no more.
In one embodiment, the above-mentioned point list information according to entrained by the natural language point forms data, generation are ordered
It is single, when it is implemented, may include the following contents: combining more in the single information of point entrained by the natural language point forms data
The matching result of a keyword, to generate the first character group, wherein first character group is used to characterize the point single object of user
And/or the customization requirement for single object;According to first character group, user's order is generated.
In one embodiment, it is contemplated that since the Feature Words of preset classification option are often limited, it is understood that there may be base
The Feature Words of classification option in existing order scheme of classes, for Partial key word can not find can matched classification option spy
The case where levying word, at this moment server be can not Direct Recognition go out the user that keyword characterized and require.In order to avoid omitting use
Family requires, and the above method is when it is implemented, can also include the following contents: not depositing in determining the preset order scheme of classes
In the case where the Feature Words with the classification option of the Keywords matching, determine matching it is unsuccessful, and by it is described matching not at
Keyword be labeled as it is to be determined;By the matching result of the keyword, described it be labeled as keyword to be determined and be determined as institute
State point list information entrained by natural language point forms data.
In the manner described above, unsuccessful for matching, the keyword of matching result, server are not dealt with, but
The keyword for matching unsuccessful keyword and successful match is distinguished by modes such as marks.
In present embodiment, it can will match unsuccessful keyword and be labeled as to be determined, can also will match unsuccessful
Keyword be labeled as to be identified or the information such as can not temporarily identify.For specific identification information, this specification is not limited
It is fixed.
In one embodiment, it is required in order to avoid omitting the unrecognized user of server, it can will be to be determined
Keyword and the matching result that has determined arranges together and generates user's order and be sent to trade company, so that trade company can be based on oneself
The industry experience and knowledge of body voluntarily determine that user indicated by keyword to be determined requires.
Specifically, above-mentioned according to the single information of the point, generate user's order, may include the following contents: combination is described certainly
The matching result of the keyword in single information is put entrained by right language point forms data, to generate the first character group, described first
Character group is used to characterize the point single object of user's requirement, and/or, customization requirement of the user for point single object;According to it is described from
It is put in single information entrained by right language point forms data and is labeled as keyword to be determined, generate the second character group, wherein institute
The second character group is stated for characterizing user for the customization requirement to be determined of single object;According to first character group, institute
The second character group is stated, order is generated.
In one embodiment, in order to mention keyword acquisition accuracy and efficiency, the above-mentioned nature to the user
Language point forms data carries out carrying out the natural language point forms data word cutting processing, and according to word cutting treated natural language
Point forms data, obtains multiple keywords, when it is implemented, may include the following contents: being handled by word cutting by the user's
Natural language point forms data is split as multiple language construction units;Carry out semantic knowledge respectively to the multiple language construction unit
Not, to obtain the multiple keyword.
In the present embodiment, the natural language point forms data of the user is split as above by word cutting processing multiple
Language construction unit carries out the natural language point forms data to user at fractionation according to syntactic structure when it is implemented, can be
Reason, obtains multiple language construction units.It is subsequent in this way to be carried out using language construction unit as specific semantics recognition unit
Semantics recognition improves the accuracy rate that keyword is obtained based on semantics recognition.Certainly, it should be noted that above-mentioned cited presses
The natural language point forms data for splitting user according to syntactic structure is intended merely to better illustrate this specification embodiment.It is specific real
Shi Shi can also as the case may be tear the natural language point forms data of user open using other suitable rules, feature
Point.In this regard, this specification is not construed as limiting.
In the present embodiment, in order to improve the semantics recognition efficiency to each language construction unit, reduction was identified
Interference in journey can first carry out the processing of invalid word rejecting to each language construction unit.When it is implemented, can be according to tool
The corresponding rule of rejecting of the application scenarios setting of body is used as preset rules.Wherein, it rejects in rule and defines invalid word, i.e., be not
The word of the attributive character of characterization point single object or description point single object.It can be by interjection, the tone in rule for example, rejecting
Word, conjunction are limited to invalid word to be rejected.It can first be rejected in each language construction unit according to rejecting rule in this way
Interjection, modal particle, conjunction.Then subsequent to carry out voice knowledge again to the language construction unit after invalid word is eliminated
Not, to reduce voice content to be identified, the processing load of server is reduced;And due to eliminating some nothings in advance
Imitate the interference of word, it helps improve the accuracy of speech recognition.
In one embodiment, it can also include the following contents when the method is embodied: receive trade company to classification
The setting operation data of option;According to the trade company to the setting operation data of classification option, preset order scheme of classes is updated
In classification option Feature Words.
In the present embodiment, the Feature Words of above-mentioned classification option specifically can be trade company according to oneself concrete condition,
Such as commodity or service provided by trade company particular content and commodity or service specific object feature, pass through second visitor
Family end (such as trade company's computer) is configured operation to classification option, the second client can acquire trade company to classification option
Setting operation data is simultaneously uploaded to server;Server after receiving above-mentioned trade company to the setting operation data of classification option,
New order scheme of classes can be generated according to aforesaid operations data, or existing order scheme of classes is updated, and will be updated
Order scheme of classes afterwards is stored in the storage equipment such as database, use when to carry out Keywords matching.
In one embodiment, in order to facilitating trade company to update the spy of the classification option in existing order scheme of classes
Word is levied, in order to provide server matches success rate, improves the accuracy of server identification, the method is when it is implemented, may be used also
To include the following contents: obtaining trade company to the setting operation data of classification option;Setting according to the trade company to classification option
Operation data is updated the order scheme of classes.
In the present embodiment, above-mentioned that the order scheme of classes is updated, when it is implemented, may include with down toward
It is one of few: increase the Feature Words of the classification option in order scheme of classes, the Feature Words of deleting classification option in order scheme of classes,
Modify the Feature Words etc. of the classification option in order scheme of classes.Certainly, it should be noted that above-mentioned cited update mode is only
It is in order to which this specification embodiment is better described.When it is implemented, can also introduce as the case may be except above-mentioned listed
Other update modes other than the mode of act are updated order scheme of classes.In this regard, this specification is not construed as limiting.
In the present embodiment, trade company can according to the change of own situation (such as increase new commodity or service, or
Person can provide new customization requirement etc.) and order in requirement to be determined, can by the second client to classification option into
Row setting operation (such as the Feature Words etc. for increasing new classification option feature word or some original classification option of deletion), the
Two clients can acquire trade company to the setting operation data of classification option and be uploaded to server;Server is on receiving
After trade company is stated to the setting operation data of classification option, it can be stored according to above-mentioned setting operation data to database etc. is stored in
The Feature Words of the classification option in order scheme of classes in equipment are updated, and are based on preset classification option to improve server
Feature Words carry out Keywords matching success rate.
In one embodiment, in order to guarantee that the order for being sent to trade company is the order for meeting user's requirement, in basis
Point list information entrained by the natural language point forms data, after generating order, the method may include: send the order
To the first client;After obtaining the confirmation instruction of the first client, the order is sent to the second client.
Specifically, being sent after server generates corresponding order according to the natural language point forms data of user by order
To trade company, the order of generation first first can be sent to the first client by server.First client is receiving the order
Afterwards, the order can be shown in the first client.User can pass through information in order interface that the first client is shown
Whether middle confirmation server order generated can accurately express the requirement of oneself.If the user determine that the order that server generates
It is the order for meeting oneself requirement, user can be by clicking the modes such as the icon that confirms an order in order displaying interface to described
Order is confirmed.First client can receive user and indicate Confirmation of Orders, and the confirmation by user to the above order
Instruction is sent to server.Server is sent to the second visitor after receiving user to Confirmation of Orders instruction, then by the order
Family end.If the user feels that the order inaccuracy that server generates, it can be by clicking the resetting order in order displaying interface
The modes such as icon indicate server order inaccuracy generated, and then user can be supplied to the first client and re-enter certainly
Right language point forms data, so that server regenerates order.
Therefore the order generation method that this specification embodiment provides, by allowing user's self-organization to input symbol
Close the natural language point forms data of oneself speech habits, and obtain the keyword in the natural language point forms data of user come in advance
If order scheme of classes matched, to determine point list information entrained by the natural language point forms data of user, so from
It is dynamic to generate corresponding user's order, thus solve the problems, such as the point of user present in existing method list inconvenience, poor user experience,
Reach can based on user's self-organization input natural language point forms data, automatically identify user point single object and/or
Customization requires, and generates the effect for meeting the order of user's requirement;In server based on the classification choosing in existing order scheme of classes
The situation that the content that the Feature Words of item characterize Partial key word can not determine, will be unable to determining keyword as the second word
Symbol group feeds back to trade company together with the first character group successfully determined together, so as to trade company according to domain knowledge, experience from
Row identifies that user to be determined requires, and avoids the omission that the system fixed to user requires;Further, also trade company is allowed to select classification
Item is configured, and according to trade company to the feature of the classification option in the setting operation data more new order scheme of classes of classification option
Word improves the accuracy that server determines the single information of point entrained by the natural language point forms data of user, further improves
User experience.
Refering to Fig. 8, this specification embodiment provides a kind of order processing method, wherein this method is applied particularly to use
The first client-side that family uses.When it is implemented, this method may include the following contents:
S81: the data input screen for receiving the natural language point forms data of user is shown;
S82: receiving the natural language point forms data that inputs by the data input screen of user, and by the nature language
Speech point forms data is sent to server;
S83: the order that server is generated according to the natural language point forms data is received, and shows the order;
S84: receiving user and indicate the Confirmation of Orders, and user is sent to clothes to Confirmation of Orders instruction
Business device.
In the present embodiment, user can be using equipment such as mobile phone, computer, smartwatch as the first client.Specifically
When implementation, above-mentioned first client can carry out data interaction with server.User can be in the first client by corresponding
Operation input self-organization comprising having a single object and the natural language of personalized customization requirement point forms data, to complete a little
It is single;First client then acquires the natural language point forms data of user, and above-mentioned natural language point forms data is sent to service
Device;Generate the order for meeting the personalized customization requirement of user according to the natural language point forms data of user by server again, and
Order is sent to corresponding trade company's processing.
In the present embodiment, user is when it is implemented, can include to have a single object and/or customization to require with self-organization
Voice data as natural language point forms data, corresponding text data can also be organized as natural language point forms data.
Concrete form this specification of natural language point forms data is not construed as limiting.The only natural language in the form of voice data below
Respective description is carried out for point forms data.Voice data shape is referred to for the natural language point forms data of form of textual data
The processing of the natural language point forms data of formula.This specification is not in this regard, repeat.
User can carry out voice input operation in the first client in the present embodiment, so as to the first client termination
Receive natural language point forms data of the data of user speech input as user.Wherein, above-mentioned voice input operation specifically can be with
It is user in the specified interface that the first client is shown (such as put single interface etc.) to the clicking operation of voice input pointer,
Or user has used specified gesture (such as upper screen sliding curtain etc.) etc. in the first client.Certainly, above-mentioned cited user
Voice input operation be intended merely to that this specification embodiment is better described.When it is implemented, can also be according to specific feelings
Condition and application scenarios use voice input operation of other modes as user.In this regard, this specification is not construed as limiting.
It in the present embodiment, can be as shown in fig.2, user can be in the click in the point list interface that mobile phone be presented
Sketch drawing mark if voice inputs is characterized, mobile phone is able to respond the aforesaid operations of user, and then starts the microphone of mobile phone to acquire
The voice data of user's self-organization is as natural language point forms data.
In the present embodiment, the first client that user uses is coupled with server by wired or wireless mode.
The natural language point forms data of user can be sent to by the first client after the natural language point forms data for collecting user
Server, the natural language point forms data of user determined according to the natural language of user point forms data so as to server entrained by
The single information of point, and generate corresponding user's order.
In the present embodiment, in order to guarantee that the order for being sent to trade company is the order for meeting user's requirement, in server
After generating corresponding order according to the natural language of user point forms data, before order is sent to trade company, server can first by
The order of generation is first sent to the first client.First client can open up after receiving the order in the first client
Show the order.User can be by confirming server order generated in information in order interface that the first client is shown
Whether oneself requirement can accurately be expressed.If the user determine that the order that server generates is the order for meeting oneself requirement, use
Family can confirm the order by clicking the modes such as the icon that confirms an order in order displaying interface.First client
It can receive user to indicate Confirmation of Orders, and user be sent to server to the confirmation instruction of the above order.Server
The second client is sent to after receiving user to Confirmation of Orders instruction, then by the order.If the user feels that server
The order inaccuracy of generation can indicate that server is given birth to by clicking the modes such as the resetting order icon in order displaying interface
At order inaccuracy, and then user can be supplied to the first client and re-enter natural language point forms data, to service
Think highly of newly-generated order.
In one embodiment, that server can accurately identify, effective in order to guide user to input as much as possible
Natural language point forms data acquires the nature of user when it is implemented, the first client inputs operation in the voice of response user
It before language point forms data, can also be executed according to following procedure: show leading question, wherein the leading question is for guiding user
Input satisfactory natural language point forms data, wherein the satisfactory natural language point forms data be include with
The natural language point forms data of the preset matched keyword of order scheme of classes.
In one embodiment, above-mentioned leading question specifically can wrap the Feature Words containing multiple classification options.It is logical in this way
Cross show include multiple classification options Feature Words leading question can effectively guide user input include with it is preset
The natural language point forms data of the keyword of the Feature Words semantic similarity of classification option, and then server can be improved and carry out key
The matched success rate of word, improves the accuracy of user's order of server generation, while can also guide user's basis as far as possible
The personalized customization requirement of oneself is reasonably arranged in existing service provided by trade company.
Therefore the order processing method that this specification embodiment provides, by supporting user according to the language of itself
Habit self-organization input is comprising having a single object and the natural language of personalized customization requirement point forms data, and by above-mentioned nature
Language point forms data is sent to server, so that server being capable of the point list according to entrained by the natural language of user point forms data
Information generate meets user requirement order, allow user it is more convenient, freely input setting carte content, improve
User experience;It also include the leading question of the Feature Words of classification option by displaying, guidance user inputs effective natural language
Point forms data, improves the accuracy of the order of the generation of server.
Refering to Fig. 9, this specification embodiment additionally provides a kind of order processing method, wherein this method is applied particularly to
The second client-side that trade company uses.When it is implemented, this method may include the following contents:
S91: order is received, wherein the order is what server was generated according to the natural language point forms data of user;
S92: the order is shown;
S93: trade company is received to the Confirmation of Orders operation data, and by the confirmation operation data and is sent to service
Device.
In the present embodiment, trade company can be using equipment such as desktop computer, tablet computer, mobile phones as above-mentioned second client
End.When it is implemented, above-mentioned second client can carry out data interaction with server.Second client can receive by servicing
User's order that device generates, and above-mentioned user's order is parsed, it extracts in user's order for characterizing the point single pair of user
As and/or for single object customization require the first character group;And then the display equipment (example of the second client can be passed through
Such as mobile phone screen) shown to trade company include the first character group order interface.Trade company can be ordered by what is shown in this way
The first character group in single interface recognizes that point single object and/or user required by user are directed to the individual character of the single object
Change customization to require, and then order confirmation operations can be carried out by the second client.Second client then can receive trade company pair
The Confirmation of Orders operation data, and the confirmation operation data are sent to server, server is then by trade company to order
Confirmation operation data be sent to the first client to notify that user trade company is really single.Trade company then can be according to required by user
Point single object and user personalized customization requirement, prepare the commodity that meet the requirements or service for user.
For example, trade company can show interface according to the order shown, it can determine the address location of a single user, use
The essential informations such as the contact method at family;Further, according to the content of the first character group shown in order interface, Ke Yigeng
Add and select single object required by user is precisely determined and be: " shredded pork with green pepper portioned rice bowl ", the customization of user require be: " one bowl
Samsung is peppery to add meat ", and then the peppery shredded pork with green pepper portioned rice bowl for adding meat of one bowl of Samsung can be got out for the user, to complete to this
It is handled under the line of order.
In one embodiment, the above method is when it is implemented, can also include the following contents:
S1: in the displaying interface of the order for showing, the first character group in order is shown, wherein described first
Character group is used to characterize the point single object of user's requirement, and/or, customization requirement of the user for point single object;
Whether S2: detecting in the order includes the second character group, wherein second character group is for characterizing user
For the customization requirement to be determined of single object;
S3: in the case where including second character group in determining the order, second character group is shown.
In the present embodiment, server, may be due to the spy of used classification option when carrying out Keywords matching
Sign word is limited, and the keyword for causing part of speech to identify can not be matched, and then cannot get corresponding matching result.But
Above-mentioned may not include the customization requirement of user by the keyword of successful match yet, in order to avoid omitting the customization of user
It is labeled as to be determined it is required that server can will match unsuccessful keyword, is distinguished with the keyword with successful match;
It will be labeled as crucial phrase to be determined again and synthesize the second character group, and user is generated according to the first character group and the second character group
Order is sent to trade company, determines to be labeled as pass to be determined in the second character group so that trade company is voluntarily identified based on industry experience
User indicated by keyword requires, to effectively prevent the omission required user.
In one embodiment, above-mentioned displaying second character group, when it is implemented, may include the following contents:
In the displaying interface in the order for showing, second character group is shown in a manner of remarks.Trade company can infuse in this way
It anticipates the keyword that can not be determined to the server in the second character group, industry experience and knowledge based on trade company itself voluntarily determine
Above-mentioned keyword characterizes customization and requires.Certainly, it should be noted that the mode of above-mentioned cited the second character group of displaying is only
It is in order to which this specification embodiment is better described.When it is implemented, other modes exhibition can also be selected as the case may be
Show the second character group.In this regard, this specification is not construed as limiting.
In one embodiment, in order to support trade company's (tool of commodity or service provided by trade company as the case may be
Appearance and commodity or the attributive character of service in vivo), the Feature Words of the classification option in default order scheme of classes are set, it is described
Method can also include the following contents when being embodied: to the set interface for showing classification option;Trade company is received in the classification
To the setting operation data of classification option in the set interface of option;Setting operation data of the trade company to classification option is sent to
Server, so as to server according to the trade company to the setting operation data more new order scheme of classes of classification option.
In the present embodiment, trade company can pass through the second client (such as trade company's computer) in advance and carry out to classification option
Setting operation, the second client can acquire trade company to the setting operation data of classification option and be uploaded to server;Service
Device can generate new order after receiving above-mentioned trade company to the setting operation data of classification option according to aforesaid operations data
Scheme of classes updates existing order scheme of classes.
In the present embodiment, the existing order scheme of classes of above-mentioned update, when specific implementation may include it is following at least it
One: increasing the Feature Words of the classification option in existing order scheme of classes, delete the classification option in existing order scheme of classes
Feature Words, the Feature Words of classification option in the existing order scheme of classes of modification etc..Certainly, it should be noted that above-mentioned institute
The update mode enumerated is intended merely to that this specification embodiment is better described.When it is implemented, can also be according to specific feelings
Other update modes that condition introduces in addition to above-mentioned cited mode are updated order scheme of classes.In this regard, this specification
It is not construed as limiting.
Therefore the order processing method that this specification embodiment provides, by supporting trade company according to the specific of itself
Situation is configured operation to the spy of classification option, and acquires the setting operation data of trade company so that server can be based on update
Order scheme of classes in the Feature Words of classification option identification determination is carried out to the natural language point forms data of acquired user, it is raw
Cheng Bingxiang trade company returns to more accurate user's order, thus guarantee that trade company can be accurately obtained the point list information of user,
It is ready to meet commodity or the service of user's requirement for user, improves user experience;Also by showing the first character group to trade company,
To show that point single object and the customization of user that server identifies require, the second character group is shown to trade company, is not had with server
There is the user to be determined identified to require, thereby may be ensured that trade company being capable of natural language that is complete, being accurately obtained user
Speech point forms data is entrained to put single information, specifies the requirement of user, provides satisfactory commodity or service for user.
This specification embodiment also provides a kind of server, and above-mentioned service implement body can be used for being responsible for the flat of processing order
Platform side can specifically include processor and the memory for storage processor executable instruction, the processing implement body
It can be according to instruction execution following steps when implementation: obtaining the natural language point forms data of user;To the natural language point list
Data carry out word cutting processing, and according to word cutting treated natural language point forms data, obtain multiple keywords;It will be the multiple
Keyword is matched with preset order scheme of classes, obtains point list information entrained by the natural language point forms data;Root
According to point list information entrained by the natural language point forms data, order is generated.
In order to more accurately complete above-metioned instruction, refering to fig. 10, this specification additionally provides another specific
Server, wherein the server includes network communications port 1001, processor 1002 and memory 1003, above structure
It is connected by Internal cable, so that each structure can carry out specific data interaction.Above-mentioned server application is in platform side.
Wherein, the network communications port 1001 obtains the natural language point forms data of user;Send order.
The processor 1002 specifically can be used for natural language point forms data progress word cutting processing, and according to
Word cutting treated natural language point forms data, obtains multiple keywords;By the multiple keyword and preset order classification
Table is matched, and point list information entrained by the natural language point forms data is obtained;According to the natural language point forms data
Entrained point list information generates order.
The memory 1003 specifically can be used for storing the natural language of the user received through network communication interface 1001
Speech point forms data, and the corresponding instruction repertorie of storage.
In the present embodiment, the network communications port 1001 can be is bound from different communication protocol, from
And the virtual port of different data can be sent or received.For example, the network communications port can be responsible for carrying out web data
No. 80 ports of communication are also possible to No. 21 ports for being responsible for carrying out FTP data communication, can also be and are responsible for carrying out mail data
No. 25 ports of communication.In addition, the network communications port can also be the communication interface or communication chip of entity.For example,
It can be mobile radio network communication chip, such as GSM, CDMA;It can also be Wifi chip;It can also be bluetooth core
Piece.
In the present embodiment, the processor 1002 can be implemented in any suitable manner.For example, processor can be with
Take such as microprocessor or processor and storage can by (micro-) processor execute computer readable program code (such as
Software or firmware) computer-readable medium, logic gate, switch, specific integrated circuit (Application Specific
Integrated Circuit, ASIC), programmable logic controller (PLC) and the form etc. for being embedded in microcontroller.This specification is simultaneously
It is not construed as limiting.
In the present embodiment, the memory 1003 may include many levels, in digital display circuit, as long as can save
Binary data can be memory;In integrated circuits, the circuit with store function of a not no physical form
Also memory, such as RAM, FIFO are;In systems, the storage equipment with physical form is also memory, such as memory bar, TF
Card etc..
This specification embodiment also provides a kind of first client, and above-mentioned first client application is in user side, specifically
It may include processor and the memory for storage processor executable instruction, the processor can root when being embodied
According to instruction execution following steps: showing the data input screen for receiving the natural language point forms data of user;Receive user
The natural language point forms data inputted by the data input screen, and the natural language point forms data is sent to service
Device;Trade company is received to Confirmation of Orders information, wherein the order is that server is raw according to the natural language point forms data
At.
This specification embodiment also provides a kind of second client, and above-mentioned second client application is in trade company side, specifically
It may include processor and the memory for storage processor executable instruction, the processor can root when being embodied
According to instruction execution following steps: receiving order, wherein the order is that server is raw according to the natural language point forms data of user
At;Show the order;Trade company is received to be sent to the Confirmation of Orders operation data, and by the confirmation operation data
Server.
This specification embodiment additionally provides a kind of computer storage medium based on above-mentioned method of payment, the computer
Storage medium is stored with computer program instructions, is performed realization in the computer program instructions: obtaining the nature of user
Language point forms data;Word cutting processing is carried out to the natural language point forms data, and according to word cutting treated natural language point
Forms data obtains multiple keywords;The multiple keyword is matched with preset order scheme of classes, obtains the nature
Point list information entrained by language point forms data;According to point list information entrained by the natural language point forms data, generation is ordered
It is single.
In the present embodiment, above-mentioned storage medium includes but is not limited to random access memory (Random Access
Memory, RAM), read-only memory (Read-Only Memory, ROM), caching (Cache), hard disk (Hard Disk
Drive, HDD) or storage card (Memory Card).The memory can be used for storing computer program instructions.Network is logical
Letter unit can be according to standard setting as defined in communication protocol, for carrying out the interface of network connection communication.
In the present embodiment, the function and effect of the program instruction specific implementation of computer storage medium storage, can
To compare explanation with other embodiment, details are not described herein.
Refering to fig. 11, on software view, this specification embodiment additionally provides a kind of order generating means, device tool
Body may include construction module below:
First obtains module 1101, specifically can be used for obtaining the natural language point forms data of user;
Second obtains module 1102, specifically can be used for carrying out the natural language point forms data word cutting processing, and root
According to word cutting treated natural language point forms data, multiple keywords are obtained;
Matching module 1103 specifically can be used for matching the multiple keyword with preset order scheme of classes,
Obtain point list information entrained by the natural language point forms data;
Generation module 1104 specifically can be used for the point list information according to entrained by the natural language point forms data, raw
At order
In one embodiment, above-mentioned natural language point forms data can specifically include: voice data and/or textual data
According to etc..It should be noted that above-mentioned cited voice data and text data are that one kind schematically illustrates, should not constitute
Improper restriction to this specification.
In one embodiment, above-mentioned matching module 1103 is when it is implemented, can be according to following procedure, described in determination
Point list information entrained by natural language point forms data: by the keyword respectively with the class in the preset order scheme of classes
The Feature Words of mesh option carry out semantic matches, and determining in the preset order scheme of classes whether there is and the Keywords matching
Classification option Feature Words;There is the classification option with the Keywords matching in determining the preset order scheme of classes
Feature Words in the case where, determine successful match, and the Feature Words of the classification option with the Keywords matching are determined as closing
The matching result of keyword;The matching result of the keyword is determined as the letter of point list entrained by the natural language point forms data
Breath.
In one embodiment, above-mentioned generation module 1104 is when it is implemented, may include following structural unit:
Assembled unit specifically can be used for combining the key in the single information of point entrained by the natural language point forms data
The matching result of word, to generate the first character group, first character group is used to characterize the point single object of user's requirement, and/or,
Customization requirement of the user for point single object;
Generation unit specifically can be used for generating order according to first character group.
In one embodiment, above-mentioned matching module 1103 is when it is implemented, can be also used for by the keyword
Semantic matches are carried out with the Feature Words of the classification option in the preset order scheme of classes respectively, determine the preset order
With the presence or absence of with after the Feature Words of the classification option of the Keywords matching in scheme of classes, the preset order classification is being determined
In the case where the Feature Words with the classification option of the Keywords matching are not present in table, determine that matching is unsuccessful, and will be described
Matching not at keyword be labeled as it is to be determined;By the matching result of the keyword, described it is labeled as keyword to be determined
It is determined as point list information entrained by the natural language point forms data.
In one embodiment, above-mentioned generation module 1104 is when it is implemented, can be also used for combining the natural language
The matching result of the keyword in single information is put entrained by speech point forms data, to generate the first character group, first character
The point single object that group is required for characterizing user, and/or, customization requirement of the user for point single object;According to the natural language
The keyword to be determined that is labeled as in single information is put entrained by speech point forms data, generates the second character group, wherein described the
Two character groups are used to characterize user for the customization requirement to be determined of point single object;According to first character group, described
Two character groups generate order.
In one embodiment, above-mentioned second module 1102 is obtained when it is implemented, may include following structural unit:
Word cutting unit specifically can be used for carrying out word cutting processing to the natural language point forms data, obtain multiple language
Structural unit;
Semantics recognition unit specifically can be used for carrying out semantics recognition respectively to the multiple language construction unit, to obtain
Take the multiple keyword.
In one embodiment, the above order generating means further include order scheme of classes update module, can specifically be used
In: trade company is obtained to the setting operation data of classification option;According to the trade company to the setting operation data of classification option, update
The preset order scheme of classes.
In one embodiment, the above order scheme of classes update module when it is implemented, can according to it is following at least it
The Feature Words of classification option in a pair of order scheme of classes are updated: increasing the spy of the classification option in order scheme of classes
Levy word, the Feature Words of classification option etc. in the Feature Words of deleting classification option in order scheme of classes, modification order scheme of classes.
Certainly, it should be noted that above-mentioned cited update mode is that one kind schematically illustrates.It should not constitute to this specification
Improper restriction.
It should be noted that unit, device or module etc. that above-described embodiment illustrates, specifically can by computer chip or
Entity is realized, or is realized by the product with certain function.For convenience of description, it describes to divide when apparatus above with function
It is described respectively for various modules.It certainly, can be the function of each module in same or multiple softwares when implementing this specification
And/or realized in hardware, the module for realizing same function can also be realized by the combination of multiple submodule or subelement etc..With
Upper described Installation practice is only schematical, for example, the division of the unit, only a kind of logic function is drawn
Point, there may be another division manner in actual implementation, such as multiple units or components may be combined or can be integrated into separately
One system, or some features can be ignored or not executed.Another point, shown or discussed mutual coupling or straight
Connecing coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of device or unit can be electrical property,
Mechanical or other forms.
Therefore the order generating means that this specification embodiment provides, module, which is obtained, by first obtains user's root
It is obtained according to the natural language point forms data that the speech habits self-organization of itself inputs, and by the second acquisition module and matching module
Take the keyword in the natural language point forms data of family, and by the classification option in above-mentioned keyword and preset order scheme of classes
Feature Words matched, to determine point list information entrained by the natural language point forms data of user, and then automatically generate
Corresponding user's order reaches energy to solve the problems, such as the point of user present in existing method list inconvenience, poor user experience
Enough natural language point forms datas inputted based on user's self-organization, the point single object and/or customization for automatically identifying user are wanted
It asks, generates the effect for meeting the order of user's requirement.
Although being based on routine or nothing present description provides the method operating procedure as described in embodiment or flow chart
Creative means may include more or less operating procedure.The step of enumerating in embodiment sequence is only numerous steps
One of rapid execution sequence mode does not represent and unique executes sequence.When device or client production in practice executes,
Can be executed according to embodiment or the execution of method shown in the drawings sequence or parallel (such as parallel processor or multithreading
The environment of processing, even distributed data processing environment).The terms "include", "comprise" or its any other variant are intended to
Cover non-exclusive inclusion, so that the process, method, product or the equipment that include a series of elements not only include those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, product or setting
Standby intrinsic element.In the absence of more restrictions, being not precluded is including process, method, the product of the element
Or there is also other identical or equivalent elements in equipment.The first, the second equal words are used to indicate names, and are not offered as appointing
What specific sequence.
It is also known in the art that other than realizing controller in a manner of pure computer readable program code, it is complete
Entirely can by by method and step carry out programming in logic come so that controller with logic gate, switch, specific integrated circuit, programmable
Logic controller realizes identical function with the form for being embedded in microcontroller etc..Therefore this controller is considered one kind
Hardware component, and the structure that the device for realizing various functions that its inside includes can also be considered as in hardware component.Or
Person even, can will be considered as realizing the device of various functions either the software module of implementation method can be hardware again
Structure in component.
This specification can describe in the general context of computer-executable instructions executed by a computer, such as journey
Sequence module.Generally, program module include routines performing specific tasks or implementing specific abstract data types, programs, objects,
Component, data structure, class etc..This specification can also be practiced in a distributed computing environment, in these distributed computing rings
In border, by executing task by the connected remote processing devices of communication network.In a distributed computing environment, program mould
Block can be located in the local and remote computer storage media including storage equipment.
As seen through the above description of the embodiments, those skilled in the art can be understood that this specification
It can realize by means of software and necessary general hardware platform.Based on this understanding, the technical solution of this specification
Substantially the part that contributes to existing technology can be embodied in the form of software products in other words, the computer software
Product can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer
Equipment (can be personal computer, mobile terminal, server or the network equipment etc.) execute each embodiment of this specification or
Method described in certain parts of person's embodiment.
Each embodiment in this specification is described in a progressive manner, the same or similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.This specification can be used for
In numerous general or special purpose computing system environments or configuration.Such as: personal computer, server computer, handheld device
Or portable device, laptop device, multicomputer system, microprocessor-based system, set top box, programmable electronics set
Standby, network PC, minicomputer, mainframe computer, distributed computing environment including any of the above system or equipment etc..
Although depicting this specification by embodiment, it will be appreciated by the skilled addressee that there are many become for this specification
Shape and the spirit changed without departing from this specification, it is desirable to which the attached claims include these deformations and change without departing from this
The spirit of specification.