CN108876527A - Method of servicing and service unit, using open platform and storage medium - Google Patents
Method of servicing and service unit, using open platform and storage medium Download PDFInfo
- Publication number
- CN108876527A CN108876527A CN201810574020.8A CN201810574020A CN108876527A CN 108876527 A CN108876527 A CN 108876527A CN 201810574020 A CN201810574020 A CN 201810574020A CN 108876527 A CN108876527 A CN 108876527A
- Authority
- CN
- China
- Prior art keywords
- commerce
- slot value
- intention
- service
- service request
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0623—Item investigation
- G06Q30/0625—Directed, with specific intent or strategy
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure proposes a kind of method of servicing and service unit, using open platform and storage medium, is related to computer field.Method of servicing therein includes:Obtain the service request that user is proposed in the form of natural language to E-business applications;The performance information and service logic of E-business applications are transferred, the performance information of E-business applications includes:The markup information of preset e-commerce words art and its intention and slot value;According to the performance information of E-business applications, the intention and slot value of service request are determined;According to the intention and slot value of service request, executes in service logic and handle logic accordingly;Implementing result is sent to user.The disclosure allows user to carry out information exchange using natural language and application, to reduce the requirement to user, improves user experience.
Description
Technical field
This disclosure relates to computer field, in particular to a kind of method of servicing and service unit, using open platform and deposit
Storage media.
Background technique
With the development of computer technology, there is miscellaneous application.User needs the side according to each application requirement
Formula and application carry out information exchange.By taking shopping application as an example, user needs to input search key according to the requirement of shopping application,
Shopping application is retrieved according to search key, and returns to the search result of product.If the search key of user's input
Inappropriate, then shopping application can not be supplied to the desired search result of user.The interactive mode of this user and application, to user
Requirement it is relatively high, user experience is bad.
Summary of the invention
In consideration of it, the disclosure is intended to propose the new interactive mode of a kind of user and application, user is allowed to use natural language
Information exchange is carried out with application, to reduce the requirement to user, improves user experience.
According to one aspect of the disclosure, a kind of method of servicing is proposed, including:
Obtain the service request that user is proposed in the form of natural language to E-business applications;
Transfer the performance information and service logic of the E-business applications, the performance information packet of the E-business applications
It includes:The markup information of preset e-commerce words art and its intention and slot value;
According to the performance information of the E-business applications, the intention and slot value of the service request are determined;
According to the intention and slot value of the service request, executes in the service logic and handle logic accordingly;
Implementing result is sent to the user.
Optionally, the e-commerce words art includes in the reference information of intention type and the reference information of slot value type
At least one of;
The markup information of the intention of the e-commerce words art is the reference letter of the intention type of the e-commerce words art
Cease the intention type represented;
The markup information of the slot value of the e-commerce words art is the reference letter of the slot value type of the e-commerce words art
Cease the slot value type represented;
Wherein, the reference information association of every kind of intention type represents one or more description informations of the intention type, often
The reference information association of kind slot value type represents one or more slot values of the slot value type.
Optionally, the e-commerce words art includes the reference information of slot value type;
The slot value type includes at least one in classification type, brand styles or the modifier type of product;
Wherein, the classification that the classification type of product includes is determined according to the classification in e-commerce purchases platform;
Or, the brand that the brand styles of product include is determined according to the brand in e-commerce purchases platform;
Or, description information of the modifier type of the product modifier that includes according to the product in e-commerce purchases platform
It determines.
Optionally, the determination method of modifier includes:
The product that the search key of record user's input and user click from search result, described search is the result is that benefit
It is searched for described search keyword and to be obtained from product description information;
If the usage frequency of described search keyword is more than preset value, described search keyword is determined as being clicked
The modifier of product.
Optionally it is determined that the intention and slot value of the service request include:
The service request is compared with each e-commerce words art, to select and the service request phase
Art is talked about like highest e-commerce is spent;
The markup information of intention of the highest e-commerce words art of the similarity and the markup information of slot value are determined respectively
For the intention and slot value of the service request.
Optionally, it is described by the service request and the e-commerce words art be compared including:
Word segmentation processing is carried out respectively to the service request and each e-commerce words art;
Each participle of the service request is compared with each participle of e-commerce words art, to select
Art is talked about with the highest e-commerce of the service request similarity;
Wherein, the method for word segmentation processing is:The sentence for treating participle carries out first word segmentation processing, if first word segmentation processing
The continuous multiple participles obtained afterwards appear in preset e-commerce dictionary as a phrase, by continuous multiple participles
Cutting is a participle.
Optionally, it executes and handles logic in the service logic accordingly and include:
If the service request is intended that shopping, according to the slot value of the service request instruction product classification,
At least one of in brand or modifier, execute the product search logic in the service logic;
If the service request is intended that inquiry order, the inquiry for executing user's order in the service logic is patrolled
Volume.
Optionally, method of servicing further includes:
According to the intention and slot value of the service request, corresponding query-response is issued to the user, and obtain institute
It states user and is directed to the response that the query-response is made;
Wherein, the slot value type of the query-response and the slot value type of the service request are configured as difference;
Also, logic is handled accordingly in the execution service logic and includes:According to the intention and slot of the service request
Value and the response, execute in the service logic and handle logic accordingly.
Optionally, the performance information of the E-business applications and the service logic pass through front end interface by application and development
Personnel's typing application open platform, and be registered to by application developer by registration interface described using open platform.
According to another aspect of the disclosure, a kind of service unit is proposed, including:
Route service module, the service request proposed in the form of natural language to E-business applications for obtaining user;
Craft services module, for transferring the performance information and service logic of the E-business applications;According to the electricity
The performance information of sub- business application determines the intention and slot value of the service request;Wherein, the technical ability of the E-business applications
Information includes:The markup information of preset e-commerce words art and its intention and slot value;
Business logic modules execute corresponding in the service logic for the intention and slot value according to the service request
Processing logic;Implementing result is sent to the user.
Optionally, the e-commerce words art includes in the reference information of intention type and the reference information of slot value type
At least one of;
The markup information of the intention of the e-commerce words art is the reference letter of the intention type of the e-commerce words art
Cease the intention type represented;
The markup information of the slot value of the e-commerce words art is the reference letter of the slot value type of the e-commerce words art
Cease the slot value type represented;
Wherein, the reference information association of every kind of intention type represents one or more description informations of the intention type, often
The reference information association of kind slot value type represents one or more slot values of the slot value type.
Optionally, the e-commerce words art includes the reference information of slot value type;
The slot value type includes at least one in classification type, brand styles or the modifier type of product;
Wherein, the classification that the classification type of product includes is determined according to the classification in e-commerce purchases platform;
Or, the brand that the brand styles of product include is determined according to the brand in e-commerce purchases platform;
Or, description information of the modifier type of the product modifier that includes according to the product in e-commerce purchases platform
It determines.
Optionally, the determination method of modifier includes:
The product that the search key of record user's input and user click from search result, described search is the result is that benefit
It is searched for described search keyword and to be obtained from product description information;
If the usage frequency of described search keyword is more than preset value, described search keyword is determined as being clicked
The modifier of product.
Optionally, craft services module is used for:
The service request is compared with each e-commerce words art, to select and the service request phase
Art is talked about like highest e-commerce is spent;
The markup information of intention of the highest e-commerce words art of the similarity and the markup information of slot value are determined respectively
For the intention and slot value of the service request.
Optionally, it is described by the service request and the e-commerce words art be compared including:
Word segmentation processing is carried out respectively to the service request and each e-commerce words art;
Each participle of the service request is compared with each participle of e-commerce words art, to select
Art is talked about with the highest e-commerce of the service request similarity;
Wherein, the method for word segmentation processing is:The sentence for treating participle carries out first word segmentation processing, if first word segmentation processing
The continuous multiple participles obtained afterwards appear in preset e-commerce dictionary as a phrase, by continuous multiple participles
Cutting is a participle.
Optionally, when the business logic modules handle logic accordingly in executing the service logic, including:
If the service request is intended that shopping, according to the slot value of the service request instruction product classification,
At least one of in brand or modifier, execute the product search logic in the service logic;
If the service request is intended that inquiry order, the inquiry for executing user's order in the service logic is patrolled
Volume.
Optionally, the craft services module includes:
Unit is transferred, for transferring the performance information and service logic of the E-business applications;
Semantic resolution unit determines that the business is asked for the performance information according to the E-business applications transferred
The intention and slot value asked;
Dialog engine issues corresponding for the intention and slot value according to the determining service request to the user
Query-response obtains the user and is directed to the response that the query-response is made;Wherein, the slot value type of the query-response with
The slot value type of the service request is configured as difference;
Also, when handling logic accordingly in the business logic modules execution service logic, for according to
The intention and slot value and the response of service request, execute in the service logic and handle logic accordingly.
According to another aspect of the disclosure, propose it is a kind of using open platform, including:
Front end interface is configured to supply the performance information of E-business applications and the input function of service logic;
Registration interface is configured to supply the performance information of E-business applications and the registering functional of service logic;And
Any one aforementioned service unit;
Wherein, the performance information of the E-business applications includes preset e-commerce words art and its is intended to and slot value
Markup information.
According to another aspect of the disclosure, a kind of service unit is proposed, including:
Memory;And
It is coupled to the processor of the memory, the processor is configured to the finger based on storage in the memory
It enables, executes any one aforementioned method of servicing.
According to another aspect of the disclosure, a kind of computer readable storage medium is proposed, be stored thereon with computer journey
Sequence, the program realize any one aforementioned method of servicing when being executed by processor.
The disclosure allows user to carry out information exchange using natural language and application, to reduce the requirement to user, mentions
High user experience.
Detailed description of the invention
Attached drawing needed in embodiment or description of Related Art will be briefly described below.According to following ginseng
According to the detailed description of attached drawing, the disclosure can be more clearly understood,
It should be evident that the accompanying drawings in the following description is only some embodiments of the present disclosure, skill common for this field
For art personnel, without any creative labor, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow diagram of some embodiments of the method for servicing of the disclosure.
Fig. 2 is the flow diagram of other embodiments of the method for servicing of the disclosure.
Fig. 3 is the schematic diagram of some embodiments using open platform of the disclosure.
Fig. 4 is the schematic diagram of some embodiments of the craft services module of the disclosure.
Fig. 5 is the flow diagram of some embodiments of the method for servicing based on application open platform of the disclosure.
Fig. 6 is the structural schematic diagram of some embodiments of the service unit of the disclosure.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present disclosure, the technical solution in the embodiment of the present disclosure is carried out clear, complete
Site preparation description.
Fig. 1 is the flow diagram of some embodiments of the method for servicing of the disclosure.
As shown in Figure 1, the method for servicing of the embodiment includes:Step 110~150.
In step 110, the service request that user is proposed in the form of natural language to E-business applications is obtained.
Wherein, service request for example can be shopping, inquiry order, connection customer service etc., but be not limited to examples cited.With certainly
The purchase request that right linguistic form proposes for example can be:" I will buy milk ".According to the requirement of conventional electrical business application, then
User is needed to input the search key of " milk ".
In step 120, the performance information and service logic of E-business applications are transferred.
Wherein, the performance information of E-business applications includes:The mark of preset e-commerce words art and its intention and slot value
Infusing information can also include optionally the relevant dictionary of slot value.
Wherein, the intention of e-commerce words art for example can be shopping, inquiry order, connection customer service etc..E-commerce words
The slot value of art for example can be classification, brand or modifier of product etc..But it is not limited to examples cited.
For example, I will buy milk->It is intended to:Shopping;
Slot value:The classification-of product>Milk.
I will look into order->It is intended to:Inquire order.
The intention and slot value of service request are determined according to the performance information of E-business applications in step 130.
In some embodiments, service request is compared with each e-commerce words art, is asked with selecting with business
Seek similarity highest e-commerce words art, and by the markup information and slot of the intention of the highest e-commerce words art of the similarity
The markup information of value is identified as the intention and slot value of service request.
Wherein, it for example can be and service request exact matching with the highest e-commerce words art of service request similarity
Art is talked about in e-commerce, if the e-commerce words art not exactly matched, chooses and the immediate e-commerce of service request
Talk about art.
In some embodiments, word segmentation processing is carried out respectively to service request and each e-commerce words art;Business is asked
The each participle asked is compared with each participle of e-commerce words art, to select and the highest electricity of service request similarity
Sub- commercial affairs words art.
Wherein, the method for word segmentation processing is:For example, by using CRF (Conditional Random Field, condition random
), the sentence for treating participle carries out first word segmentation processing, if the continuous multiple participles obtained after first word segmentation processing are as one
A phrase appears in preset e-commerce dictionary, is a participle by continuous multiple participle cuttings.
For example, first word segmentation result is:" I ", " buying ", " three ", " only ", " squirrel ", " ", " happy ", " fruit ", benefit
It is with the result that e-commerce dictionary segments again:" I ", " buying ", " three squirrels ", " ", " American pistachios ".
It executes in service logic according to the intention and slot value of service request in step 140 and handles logic accordingly.
For example, if service request is intended that shopping, according to the classification of the product of the slot value of service request instruction, brand
At least one of or in modifier, execute the product search logic in service logic.If the inquiry that is intended that of service request is ordered
It is single, execute the query logic of user's order in service logic.
In step 150, implementing result is sent to user.
For example, sending the product information searched or the order information inquired to user.
Above-described embodiment allows user to carry out information exchange using natural language and application, wants to reduce to user
It asks, improves user experience.
In some embodiments, the performance information and service logic of E-business applications is opened by front end interface by application
Hair personnel's typing application open platform, and be registered to by application developer using open platform by registration interface, thus complete
At application and development.Public service logic involved in each step in method of servicing such as obtains user's request, calls application registration letter
Intention and slot value, the execution service logic etc. for ceasing, determining user's request, can be provided by application open platform is unified, without answering
It is individually developed again with developer, is conducive to the quick exploitation of application.
It further includes under normal circumstances the description information of slot value that e-commerce words art, which includes the description information being intended to,.Some spies
Different e-commerce words art can only include the description information of intention.For example, " I will look into order " only includes the description letter of intention
Breath.
In some embodiments, the description information of the intention of the e-commerce words art of setting for example can be intention type
Reference information, it is intended that the reference information of type is associated with several description informations of the intention type.The intention of art is talked about in e-commerce
Markup information is the intention type that the reference information of the intention type of e-commerce words art represents.Similar, the electronics quotient of setting
The description information of the slot value of business words art for example can be the reference information of slot value type, and the reference information of slot value type, which is associated with, to be represented
Several slot values of the slot value type.Several slot values of slot value type are also referred to as the dictionary of the slot value type.Art is talked about in e-commerce
The markup information of slot value is the slot value type that the reference information of the slot value type of e-commerce words art represents.
It is set as example, the e-commerce for buying product can be talked about art by application developer:
I {@do shopping intention type } { classification of@product },
Wherein ,@is the symbol for indicating " to quote ", and markup information is:It is intended to:Shopping;Slot value:The classification of product.
Definition:The description information of " shopping " intention type includes:It buys, buy;The slot value of " classification of product " slot value type
Including:Milk, clothes, mobile phone etc..
In this way, can simplify the development of application developer, the quantity for the e-commerce words art for needing to be arranged is saved.
In some embodiments, slot value type include in classification type, brand styles or the modifier type of product extremely
One item missing.The classification that the classification type of product includes is determined according to the classification in e-commerce purchases platform.The brand class of product
The brand that type includes is determined according to the brand in e-commerce purchases platform.The modifier that the modifier type of product includes according to
The description information of product in e-commerce purchases platform determines.
The information of needs is grabbed from ready-made e-commerce purchases platform, so that application developer does not need to build certainly
Oneself crawler and assembly line obtain these information, can save the development of application developer.
In some embodiments, it according to the description information of the product in e-commerce purchases platform, can use with lower section
Method determines the modifier that the modifier type of product includes:
Method 1:Using TextRank algorithm, extracted from the description information of the product in e-commerce purchases platform crucial
Word rejects commodity word, brand word, stop-word (such as:, very) obtain the modifier of the product.
The modifier for the product that this method obtains is more complete.
Method 2:The product that the search key of record user's input and user click from search result, the search result
It is to search for obtain from product description information using search key, if the usage frequency of search key is more than default
Value, the modifier for the product that search key is determined as being clicked.
The modifier for the product that this available current comparison of method is popular.
For example, a product description information is " Hong Kong tide board poplar power is with money design of scattered small flowers and plants chiffon half body skirt ", it is " fragrant according to method 1
The modifiers such as port tide board " " the same money of poplar power " " design of scattered small flowers and plants " " chiffon " " half body skirt " may be all extracted, according to method 2, " poplar power
The hot topic modifier such as same money " may be extracted.
Fig. 2 is the flow diagram of other embodiments of the method for servicing of the disclosure.
As shown in Fig. 2, the method for servicing of the embodiment includes:Step 210~270.
In step 210, the service request that user is proposed in the form of natural language to E-business applications is obtained.
In step 220, the performance information and service logic of E-business applications are transferred.
Wherein, the performance information of E-business applications includes:The mark of preset e-commerce words art and its intention and slot value
Infusing information can also include optionally the relevant dictionary of slot value.
The intention and slot value of service request are determined according to the performance information of E-business applications in step 230.Determining
Specific method is with reference to aforementioned.
Corresponding query-response is issued the user with according to the intention and slot value of service request in step 240.
Wherein, the slot value type of query-response and the slot value type of service request are configured as difference.For example, user proposes
Service request be " I will buy milk ", slot value type is the classification of product, then what brand query-response can be " to buy
", slot value type is the brand of product.
In step 250, obtains user and be directed to the response that query-response is made.
In step 260, according to the intention of service request and slot value and response, executes to handle accordingly in service logic and patrol
Volume.
For example, if service request is intended that shopping, according to the class for the product that service request and the slot value of response indicate
At least two in mesh, brand or modifier, execute the product search logic in service logic.
In step 270, implementing result is sent to user.
Above-described embodiment pointedly further inquires user according to the request of user, and being conducive to precise positioning user needs
It asks, accurate service is provided.
Fig. 3 is the schematic diagram of some embodiments using open platform of the disclosure.
As shown in figure 3, including using open platform 300:Front end interface 310, registration interface 320 and service unit 330.
Front end interface 310 is configured to supply the performance information of E-business applications and the input function of service logic.It answers
The performance information of E-business applications and service logic can be entered by front end interface 310 using opening with developer
Platform.
Registration interface 320 is configured to supply the performance information of E-business applications and the registering functional of service logic.It answers
The performance information of E-business applications and service logic can be registered to by registration interface 320 using opening with developer
Platform.After registration, the application is just can be used in user.
Service unit 330 is configured as acquisition user and is asked in the form of natural language to the business that E-business applications propose
It asks;Transfer the performance information and service logic of E-business applications;According to the performance information of E-business applications, the business of determination is asked
The intention and slot value asked;According to the intention and slot value of service request, executes in service logic and handle logic accordingly;It sends and executes
As a result user is given.
Public service logic involved in service unit 330 such as obtains user's request, calls and apply registration information, determination
The intention and slot value of user's request execute service logic etc., can be provided by application open platform is unified, be not necessarily to application and development people
Member's individually exploitation again, is conducive to the quick exploitation of application.
As shown in figure 3, service unit 330 includes module 331~333.
Route service module 331 is asked in the form of natural language to the business that E-business applications propose for obtaining user
It asks.
Craft services module 332, for transferring the performance information and service logic of E-business applications;According to e-commerce
The performance information of application determines the intention and slot value of service request.
Business logic modules 333 are executed and are handled accordingly in service logic for the intention and slot value according to service request
Logic;Implementing result is sent to user.
In some embodiments, as shown in figure 4, craft services module 332 includes:Transfer unit 3321, semantic parsing list
Member 3322, further can also include dialog engine 3323.
Unit 3321 is transferred, the performance information and business for calling and obtaining user request respective application (such as E-business applications)
Logic.
Semantic resolution unit 3322 determines service request for the performance information according to the E-business applications transferred
Intention and slot value.
Optionally, semantic resolution unit 3322 includes that coupling subelement and similarity handle subelement.Coupling subelement quilt
It is configured to:Service request is matched with each e-commerce words art, if there is matching result, and the electronics quotient that will match to
The markup information of intention and the markup information of slot value of business words art are identified as the intention and slot value of service request.At similarity
Reason subelement is configured as:Such as it can be when coupling subelement have matching result, by service request and each e-commerce
Words art is compared, and art is talked about with the highest e-commerce of service request similarity to select, by the highest electronics of the similarity
The markup information of intention and the markup information of slot value of commercial affairs words art are identified as the intention and slot value of service request.
Dialog engine 3323 issues the user with corresponding query-response, obtains for the intention and slot value according to service request
It takes family and is directed to the response that query-response is made;Wherein, the slot value type of query-response and the slot value type of service request are matched
It is set to difference.At this point, business logic modules 333 are when executing service logic, for according to the intention and slot value of service request with
And response, it executes in service logic and handles logic accordingly.
When program is realized, the different progresses of work can be arranged according to different applications in craft services module 332
(worker).The work-based logic of unit 3321, semantic resolution unit 332, dialog engine 3323 etc. is transferred in operation in the progress of work.
For example, progress of work A is arranged using A, progress of work B is set using B.Craft services module 332 is if it find that user requested
It is intended to exceed the process range of current application, user can be requested by the work from current application by progress of work switching interface
Make the progress of work that process switching requests corresponding other application to user, and continues as using by the progress of work of the other application
Family service.
For example, user has found there is haze weather in the progress of work of weather, continue to request " I wants to buy haze mask ",
The then progress of work for being switched to E-business applications by the progress of work of weather with request, the progress of work of E-business applications
The search result of haze mask is sent to user.
Some public progresses of work can also be arranged in craft services module 332, and the public progress of work can be each application
The progress of work provides technical support, for example, the basic dictionary etc. of some public uses, such as the basic dictionary in province, Ke Yiyou
The public progress of work is transferred.
In some embodiments, when business logic modules 333 handle logic accordingly in executing service logic, if industry
Business request is intended that shopping, according at least one in the classification of the product of the slot value of service request instruction, brand or modifier
, execute the product search logic in service logic;If service request is intended that inquiry order, execute in service logic
The query logic of user's order.
Craft services module 332 and business logic modules 333 can for example be deployed in Storm, and (Storm is a kind of real-time number
According to stream calculation frame) in cluster, and a job can be respectively set in craft services module 332 and business logic modules 333
Process.
Fig. 5 is the flow diagram of some embodiments of the method for servicing based on application open platform of the disclosure.
As shown in figure 5, method of servicing includes:Step 510~580.
In step 510, user wakes up E-business applications.
For example, user, which sends, wakes up word (such as " please open shopping ") to route service module 331.User can be defeated in webpage
Enter and wake up word, or wake-up word is said by voice mode.For the wake-up word of voice mode, need to be converted to written form.
Route service module 331 according to wake up word identify user request be E-business applications, record the mark and knowledge of user
Not Chu Lai E-business applications mark, so as to the subsequent information from the user be transmitted to the E-business applications correlation
The progress of work handled.
In step 520, user proposes service request in the form of natural language, and route service module 331 is by the industry of the user
The progress of work for the E-business applications that business request is transmitted in craft services module 332.
For example, user proposes " I will buy milk ", route service module 331 finds corresponding electricity according to the mark of the user
The mark of sub- business application forwards the request to the progress of work of the E-business applications in craft services module 332.
Wherein, user can request in webpage incoming traffic, or say service request by voice mode.For voice side
The service request of formula needs to be converted to written form.
In step 530, craft services module 332 transfers the performance information that unit 3321 transfers E-business applications using it
And service logic, according to the performance information determination of E-business applications and service request is sent using its semantic resolution unit 3322
Intention and slot value to business logic modules 333.
Wherein, the intention of service request and the determination method of slot value are with reference to aforementioned.For example, " I will buy milk " is intended to
" shopping ", slot value are " classification-milk of product ".
According to business needs, if you do not need to and user session, then craft services module 332 by the intention of service request and
Slot value is sent to business logic modules 333, then executes step 540, if necessary to and user session, then craft services module
The intention of service request and slot value are sent to dialog engine 3323 by 332, then execute step 550.
In step 540, business logic modules 333 execute corresponding in service logic according to the intention and slot value of service request
Processing logic, by route service module 331 send implementing result to user.Process terminates.
For example, business logic modules 333 may search for the product about " milk ", and search result is returned into user.
In step 550, dialog engine 3323 according to the intention and slot value of service request, by route service module 331 to
User issues corresponding query-response.
Wherein, the slot value type of query-response and the slot value type of service request are configured as difference.For example, user proposes
Service request be " I will buy milk ", slot value type is the classification of product, then what brand query-response can be " to buy
", slot value type is the brand of product.
In step 560, user is sent to craft services by route service module 331 for the response that query-response is made
Module 332.
For example, user replys " Mongolia Ox " if query-response is " what brand bought ".
According to business needs, if no longer needed and user session, craft services module 332 is by the intention of service request
Business logic modules 333 are sent to for the response that query-response is made with slot value and user, then execute step 570, such as
Fruit needs to continue and user session, then the response of user is sent to dialog engine 3323 by craft services module 332, is then executed
Step 580.
In step 570, business logic modules 333 are directed to query-response according to the intention and slot value of service request and user
The response made, execute service logic in handle logic accordingly, by route service module 331 send implementing result to
Family.Process terminates.
For example, business logic modules 333 may search for the product about " Mongolia Ox's milk ", and search result is returned into use
Family.
In step 580, dialog engine 3323 is done according to the intention and slot value of service request and user for query-response
Response out issues the user with corresponding query-response again by route service module 331.It then proceedes to execute step 560.
For example, if classification of the slot value type in addition to product, brand, further include the place of production of product, then can inquire again
User's " buying the place of production is milk where " etc..
Above-described embodiment allows user to carry out information exchange using natural language and application, wants to reduce to user
It asks, improves user experience.Also, user pointedly can also be further inquired according to the request of user, it is precisely fixed to be conducive to
Position user demand, provides accurate service.
Fig. 6 is the structural schematic diagram of some embodiments of the service unit of the disclosure.
As shown in fig. 6, the device 600 of the embodiment includes:Memory 610 and the processing for being coupled to the memory 610
Device 620, processor 620 are configured as executing in aforementioned any some embodiments based on the instruction being stored in memory 610
Method of servicing.
Wherein, memory 610 is such as may include system storage, fixed non-volatile memory medium.System storage
Device is for example stored with operating system, application program, Boot loader (Boot Loader) and other programs etc..
Device 500 can also include input/output interface 630, network interface 640, memory interface 650 etc..These interfaces
It can for example be connected by bus 660 between 630,640,650 and memory 610 and processor 620.Wherein, input and output
The input-output equipment such as interface 630 is display, mouse, keyboard, touch screen provide connecting interface.Network interface 640 is various
Networked devices provide connecting interface.The external storages such as memory interface 650 is SD card, USB flash disk provide connecting interface.
The disclosure also proposes a kind of computer readable storage medium, is stored thereon with computer program, and the program is processed
Device realizes the method for servicing in aforementioned any some embodiments when executing.
Those skilled in the art should be understood that embodiment of the disclosure can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the disclosure
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the disclosure, which can be used in one or more,
Machine can use the meter implemented in non-transient storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of calculation machine program product.
The disclosure is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present disclosure
Figure and/or block diagram describe.It is interpreted as to be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computer journeys
Sequence instruct to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices processor with
A machine is generated, so that the instruction generation executed by computer or the processor of other programmable data processing devices is used for
Realize the dress for the function of specifying in one or more flows of the flowchart and/or one or more blocks of the block diagram
It sets.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The foregoing is merely the preferred embodiments of the disclosure, not to limit the disclosure, all spirit in the disclosure and
Within principle, any modification, equivalent replacement, improvement and so on be should be included within the protection scope of the disclosure.
Claims (20)
1. a kind of method of servicing, including:
Obtain the service request that user is proposed in the form of natural language to E-business applications;
The performance information and service logic of the E-business applications are transferred, the performance information of the E-business applications includes:
The markup information of preset e-commerce words art and its intention and slot value;
According to the performance information of the E-business applications, the intention and slot value of the service request are determined;
According to the intention and slot value of the service request, executes in the service logic and handle logic accordingly;
Implementing result is sent to the user.
2. method of servicing as described in claim 1, wherein e-commerce words art include intention type reference information and
At least one of in the reference information of slot value type;
The markup information of the intention of the e-commerce words art is the reference information generation of the intention type of the e-commerce words art
The intention type of table;
The markup information of the slot value of the e-commerce words art is the reference information generation of the slot value type of the e-commerce words art
The slot value type of table;
Wherein, the reference information association of every kind of intention type represents one or more description informations of the intention type, every kind of slot
The reference information association of Value Types represents one or more slot values of the slot value type.
3. method of servicing as described in claim 1, wherein the e-commerce words art includes the reference information of slot value type;
The slot value type includes at least one in classification type, brand styles or the modifier type of product;
Wherein, the classification that the classification type of product includes is determined according to the classification in e-commerce purchases platform;
Or, the brand that the brand styles of product include is determined according to the brand in e-commerce purchases platform;
Or, the modifier that the modifier type of product includes is true according to the description information of the product in e-commerce purchases platform
It is fixed.
4. method of servicing as claimed in claim 3, wherein the determination method of modifier includes:
The product that the search key of record user's input and user click from search result, described search is the result is that utilize institute
It states search key and searches for from product description information and obtain;
If the usage frequency of described search keyword is more than preset value, described search keyword is determined as to the product being clicked
Modifier.
5. method of servicing as described in claim 1, wherein determine the intention of the service request and slot value includes:
The service request is compared with each e-commerce words art, to select and the service request similarity
Art is talked about in highest e-commerce;
The markup information of intention of the highest e-commerce words art of the similarity and the markup information of slot value are identified as institute
State the intention and slot value of service request.
6. method of servicing as claimed in claim 5, wherein described to carry out the service request and e-commerce words art
Compare including:
Word segmentation processing is carried out respectively to the service request and each e-commerce words art;
Each participle of each participle of the service request and the e-commerce words art is compared, to select and institute
State the highest e-commerce words art of service request similarity;
Wherein, the method for word segmentation processing is:The sentence for treating participle carries out first word segmentation processing, if after first word segmentation processing
To continuous multiple participles appear in preset e-commerce dictionary as a phrase, will continuously multiple participle cuttings
It is segmented for one.
7. method of servicing as described in claim 1, wherein execute and handle logic in the service logic accordingly and include:
If the service request is intended that shopping, according to the classification of the product of the slot value of service request instruction, brand
At least one of or in modifier, execute the product search logic in the service logic;
If the service request is intended that inquiry order, the query logic of user's order in the service logic is executed.
8. method of servicing as described in claim 1, further includes:
According to the intention and slot value of the service request, corresponding query-response is issued to the user, and obtain the use
Family is directed to the response that the query-response is made;
Wherein, the slot value type of the query-response and the slot value type of the service request are configured as difference;
Also, logic is handled accordingly in the execution service logic and includes:According to the intention of the service request and slot value with
And the response, it executes in the service logic and handles logic accordingly.
9. method of servicing as described in claim 1, wherein
The performance information of the E-business applications and the service logic are answered by front end interface by application developer's typing
With open platform, and it is registered to by application developer by registration interface described using open platform.
10. a kind of service unit, including:
Route service module, the service request proposed in the form of natural language to E-business applications for obtaining user;
Craft services module, for transferring the performance information and service logic of the E-business applications;According to the electronics quotient
The performance information of business application, determines the intention and slot value of the service request;Wherein, the performance information of the E-business applications
Including:The markup information of preset e-commerce words art and its intention and slot value;
Business logic modules execute and locate accordingly in the service logic for the intention and slot value according to the service request
Manage logic;Implementing result is sent to the user.
11. service unit as claimed in claim 10, wherein the e-commerce words art includes the reference information of intention type
With at least one in the reference information of slot value type;
The markup information of the intention of the e-commerce words art is the reference information generation of the intention type of the e-commerce words art
The intention type of table;
The markup information of the slot value of the e-commerce words art is the reference information generation of the slot value type of the e-commerce words art
The slot value type of table;
Wherein, the reference information association of every kind of intention type represents one or more description informations of the intention type, every kind of slot
The reference information association of Value Types represents one or more slot values of the slot value type.
12. service unit as claimed in claim 10, wherein the e-commerce words art includes the reference letter of slot value type
Breath;
The slot value type includes at least one in classification type, brand styles or the modifier type of product;
Wherein, the classification that the classification type of product includes is determined according to the classification in e-commerce purchases platform;
Or, the brand that the brand styles of product include is determined according to the brand in e-commerce purchases platform;
Or, the modifier that the modifier type of product includes is true according to the description information of the product in e-commerce purchases platform
It is fixed.
13. service unit as claimed in claim 12, wherein the determination method of modifier includes:
The product that the search key of record user's input and user click from search result, described search is the result is that utilize institute
It states search key and searches for from product description information and obtain;
If the usage frequency of described search keyword is more than preset value, described search keyword is determined as to the product being clicked
Modifier.
14. service unit as claimed in claim 10, wherein craft services module is used for:
The service request is compared with each e-commerce words art, to select and the service request similarity
Art is talked about in highest e-commerce;
The markup information of intention of the highest e-commerce words art of the similarity and the markup information of slot value are identified as institute
State the intention and slot value of service request.
15. service unit as claimed in claim 14, wherein it is described by the service request and the e-commerce talk about art into
Row relatively includes:
Word segmentation processing is carried out respectively to the service request and each e-commerce words art;
Each participle of each participle of the service request and the e-commerce words art is compared, to select and institute
State the highest e-commerce words art of service request similarity;
Wherein, the method for word segmentation processing is:The sentence for treating participle carries out first word segmentation processing, if after first word segmentation processing
To continuous multiple participles appear in preset e-commerce dictionary as a phrase, will continuously multiple participle cuttings
It is segmented for one.
16. service unit as claimed in claim 10, wherein business logic modules phase in executing the service logic
When the processing logic answered, including:
If the service request is intended that shopping, according to the classification of the product of the slot value of service request instruction, brand
At least one of or in modifier, execute the product search logic in the service logic;
If the service request is intended that inquiry order, the query logic of user's order in the service logic is executed.
17. service unit as claimed in claim 10, wherein the craft services module includes:
Unit is transferred, for transferring the performance information and service logic of the E-business applications;
Semantic resolution unit determines the service request for the performance information according to the E-business applications transferred
Intention and slot value;
Dialog engine issues corresponding inquiry to the user for the intention and slot value according to the determining service request
Response obtains the user and is directed to the response that the query-response is made;Wherein, the slot value type of the query-response with it is described
The slot value type of service request is configured as difference;
Also, when handling logic accordingly in the business logic modules execution service logic, for according to the business
The intention and slot value and the response of request, execute in the service logic and handle logic accordingly.
18. it is a kind of using open platform, including:
Front end interface is configured to supply the performance information of E-business applications and the input function of service logic;
Registration interface is configured to supply the performance information of E-business applications and the registering functional of service logic;
And
The described in any item service units of claim 10-17;
Wherein, the performance information of the E-business applications includes the mark of preset e-commerce words art and its intention and slot value
Information.
19. a kind of service unit, including:
Memory;
And
It is coupled to the processor of the memory, the processor is configured to the instruction based on storage in the memory,
Perform claim requires method of servicing described in any one of 1-9.
20. a kind of computer readable storage medium, is stored thereon with computer program, power is realized when which is executed by processor
Benefit requires method of servicing described in any one of 1-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810574020.8A CN108876527A (en) | 2018-06-06 | 2018-06-06 | Method of servicing and service unit, using open platform and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810574020.8A CN108876527A (en) | 2018-06-06 | 2018-06-06 | Method of servicing and service unit, using open platform and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108876527A true CN108876527A (en) | 2018-11-23 |
Family
ID=64337115
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810574020.8A Pending CN108876527A (en) | 2018-06-06 | 2018-06-06 | Method of servicing and service unit, using open platform and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108876527A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020140487A1 (en) * | 2019-01-02 | 2020-07-09 | 苏宁云计算有限公司 | Speech recognition method for human-machine interaction of smart apparatus, and system |
CN111782776A (en) * | 2019-09-26 | 2020-10-16 | 北京沃东天骏信息技术有限公司 | Method and device for realizing intention identification through slot filling |
CN112100391A (en) * | 2019-05-31 | 2020-12-18 | 阿里巴巴集团控股有限公司 | User intention identification method, device, server, client and terminal equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150179168A1 (en) * | 2013-12-20 | 2015-06-25 | Microsoft Corporation | Multi-user, Multi-domain Dialog System |
CN104750795A (en) * | 2015-03-12 | 2015-07-01 | 北京云知声信息技术有限公司 | Intelligent semantic searching system and method |
CN107315737A (en) * | 2017-07-04 | 2017-11-03 | 北京奇艺世纪科技有限公司 | A kind of semantic logic processing method and system |
-
2018
- 2018-06-06 CN CN201810574020.8A patent/CN108876527A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150179168A1 (en) * | 2013-12-20 | 2015-06-25 | Microsoft Corporation | Multi-user, Multi-domain Dialog System |
CN104750795A (en) * | 2015-03-12 | 2015-07-01 | 北京云知声信息技术有限公司 | Intelligent semantic searching system and method |
CN107315737A (en) * | 2017-07-04 | 2017-11-03 | 北京奇艺世纪科技有限公司 | A kind of semantic logic processing method and system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020140487A1 (en) * | 2019-01-02 | 2020-07-09 | 苏宁云计算有限公司 | Speech recognition method for human-machine interaction of smart apparatus, and system |
CN112100391A (en) * | 2019-05-31 | 2020-12-18 | 阿里巴巴集团控股有限公司 | User intention identification method, device, server, client and terminal equipment |
CN112100391B (en) * | 2019-05-31 | 2023-06-13 | 阿里巴巴集团控股有限公司 | User intention recognition method, device, service end, client and terminal equipment |
CN111782776A (en) * | 2019-09-26 | 2020-10-16 | 北京沃东天骏信息技术有限公司 | Method and device for realizing intention identification through slot filling |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104639420B (en) | The information processing method and system of instant messaging | |
US6556991B1 (en) | Item name normalization | |
WO2019099310A1 (en) | Resource-efficient generation of a knowledge graph | |
CN109858007A (en) | Semantic analysis answering method and device, computer equipment and storage medium | |
CN107122431A (en) | A kind of real-time calculating platform and the method for computing data based on real-time calculating platform | |
CN108876527A (en) | Method of servicing and service unit, using open platform and storage medium | |
WO2014008139A2 (en) | Generating search results | |
CN105812394A (en) | Novel application of cloud computing to cross-border electronic commerce | |
WO2017005119A1 (en) | Method and device for implementing individualized guidance | |
CN106326243B (en) | Data processing method and device | |
CN109783589A (en) | The method, apparatus and storage medium of electronic map parsing address | |
US11062224B2 (en) | Prediction using fusion of heterogeneous unstructured data | |
CN108491408A (en) | A kind of processing method of action message, device, electronic equipment and storage medium | |
CN107748748A (en) | Water conservancy and hydropower technical standard text retrieval system | |
CN111522840B (en) | Label configuration method, device, equipment and computer readable storage medium | |
CN110517079B (en) | Data processing method and device, electronic equipment and storage medium | |
CN112130846A (en) | Three-micro one-screen publishing engine system and publishing method | |
WO2021088589A1 (en) | Task query method and device | |
CN109086426A (en) | Data query method, apparatus, computer equipment and readable storage medium storing program for executing | |
WO2022127005A1 (en) | Application intention response method and apparatus, device, storage medium and program product | |
US20220342919A1 (en) | Method and system for generating a response to an unstructured natural language (nl) query | |
CN113298397A (en) | Information processing method and device | |
US20120072227A1 (en) | Automatically generating high quality soa design from business process maps based on specified quality goals | |
CN112860811A (en) | Method and device for determining data blood relationship, electronic equipment and storage medium | |
CN109033070A (en) | A kind of data processing method, server and computer-readable medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |