CN108153879A - The method and device of recommendation information is provided a user by human-computer interaction - Google Patents
The method and device of recommendation information is provided a user by human-computer interaction Download PDFInfo
- Publication number
- CN108153879A CN108153879A CN201711431022.3A CN201711431022A CN108153879A CN 108153879 A CN108153879 A CN 108153879A CN 201711431022 A CN201711431022 A CN 201711431022A CN 108153879 A CN108153879 A CN 108153879A
- Authority
- CN
- China
- Prior art keywords
- user
- information
- recommended
- human
- interaction
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9032—Query formulation
- G06F16/90332—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0257—User requested
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Accounting & Taxation (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The disclosure provides a kind of method and device that recommendation information is provided a user by human-computer interaction.Included according to disclosed method (200)：Human-computer interaction (S210) is carried out with user；Information (S220) is extracted according to interaction content in interactive process；According to the information of extraction, continuous updating information recommended to the user (S230)；In interactive process, newer information is constantly recommended into user (S240).Recommending module and interactive engine are mainly included according to the device of the disclosure, are respectively intended to perform aforesaid operations.By disclosed method and device, recommending module constantly adjusts recommendation results according to user feedback, until user finds the product oneself liked or service.
Description
Technical field
The present invention relates to human-computer interactions and information recommendation, are more particularly to provide a user recommendation by human-computer interaction
The method and device of breath.
Background technology
The system that man-machine interactive system has been widely used in graphical interaction mode is user such as mobile phone app and PC websites
Provide the experience better than search.But the man-machine interactive system under graphical interaction mode because be limited to graphical interfaces size and
Abundant interaction can not be carried out with user, user is very limited to the feedback form of recommendation results, common only " liking ",
The actions such as " not liking ", " skipping ".The success rate for oneself liking product is successfully positioned this greatly reduces user.
Accordingly, it is desirable to provide a kind of method and device that recommendation information is provided a user by human-computer interaction, Neng Gougen
According to the various feedback of user, abundant coping style is provided, so that user eventually finds the product oneself liked or service.
Invention content
It as described above, can be in the interaction with product or service man-machine interactive system present invention aims to obtain user
Middle offer various feedback, and man-machine interactive system can provide abundant coping style according to the feedback of user, so as to
Family eventually finds the product oneself liked.
According to the first aspect of the invention, a kind of method that recommendation information is provided a user by human-computer interaction is provided.
This method may include steps of：Human-computer interaction is carried out with user；Information is extracted according to interaction content in interactive process；
According to the information of extraction, continuous updating information recommended to the user；In interactive process, newer information is constantly recommended into use
Family.
Preferably, the human-computer interaction includes the interaction carried out by word, voice, expression or link.
Preferably, it is described to include actively inquiring user with user's progress human-computer interaction.
Preferably, the step of carrying out human-computer interaction with user may further include：It is carried out according to user's portrait with user
Human-computer interaction.The step of extracting information according to interaction content in interactive process may further include：In interactive process
User's portrait is further updated according to interaction content.According to the information of extraction, the step of continuous updating information recommended to the user
Suddenly it may further include：It is drawn a portrait according to user, continuous updating information recommended to the user.
User's portrait refers to according to user's registration information or the individual subscriber obtained according to the interaction with user
Information and preference information.
Preferably, information recommended to the user includes following at least one：
Product or the tabulation of service；
Specific product or service list；
Product or service related information；
Product or service related information link；
Reply to customer problem；
For the option of user's selection.
Preferably, according to a first aspect of the present invention by human-computer interaction and provide a user the method for recommendation information can be with
Further comprise：If user's expression is satisfied with recommendation information, recommend more similar information to user；If user expresses
Recommendation information is unsatisfied with, then to user again recommendation information.
Preferably, the step of extracting information according to interaction content in interactive process may further include：Extraction is used
Family is to the structural data of product or service request.According to the information of extraction, the step of continuous updating information recommended to the user
It may further include：According to user to the structural data adjustment algorithm of product or service request, update recommended to the user
Product or service list.
Preferably, the human-computer interaction carries out under specific industry background.According to the information of extraction, continuous updating to
The step of information that user recommends, further comprises：According to the information and industry background extracted, continuous updating is recommended to the user
Product or service list.
According to the second aspect of the invention, a kind of device that recommendation information is provided a user by human-computer interaction is provided.
The device can include：Recommending module is configured for generating information recommended to the user；Interactive engine is configured for：With
User carries out human-computer interaction；Information is extracted according to interaction content in interactive process；By the recommending module generate to
The information that family is recommended is presented to the user.Wherein, the recommending module can be further configured and be used for：According to the interactive engine
The information of extraction, continuous updating information recommended to the user；The interactive engine can be further configured interacting
Newer information is constantly presented to the user by Cheng Zhong.
According to the third aspect of the invention we, a kind of computer-readable medium is provided, for recording what can be performed by processor
Instruction, described instruction is when being executed by processor so that processor is performed provides a user recommendation information by human-computer interaction
Method, including operating as follows：Human-computer interaction is carried out with user；Information is extracted according to interaction content in interactive process；Root
According to the information of extraction, continuous updating information recommended to the user；In interactive process, newer information is constantly recommended into use
Family.
Man-machine interactive system can with user by word, voice, expression, the interaction that progress take turns more such as link.If with
The feedback of family input is natural language, and system can utilize natural language processing and machine learning techniques to understand the feedback of user,
Man-machine interactive system accordingly adjusts recommendation results later.This process constantly repeats to find the production that oneself is enabled to be satisfied with until user
Product or service.
Description of the drawings
Below with reference to the accompanying drawings it is described in conjunction with the embodiments the present invention.In the accompanying drawings：
Fig. 1 is a specific embodiment according to the present invention, the flow chart of the method for recommended products in human-computer interaction.
Fig. 2 is the flow chart according to the present invention that the method for recommendation information is provided a user by human-computer interaction.
Fig. 3 is the schematic block diagram according to the present invention that the device of recommendation information is provided a user by human-computer interaction.
Specific embodiment
Attached drawing is given for example only explanation, it is impossible to be interpreted as limitation of the present invention.With reference to the accompanying drawings and examples to this
The technical solution of invention is described further.
Embodiment 1
Fig. 1 is a specific embodiment according to the present invention, the flow chart of the method for recommended products in human-computer interaction.
In the method 100 of Fig. 1, first, in step S110, according to user's portrait and newest demand, recommend for user
Product list.Specifically, man-machine interactive system can be user's recommended products list according to following two data：User draws a portrait
Product requirement that (new user is without this data) and user provide in current interaction (such as " I to take Me Home child insure ").Institute
State user's portrait refer to according to user's registration information or according to the userspersonal information obtained with being interacted before user and
Preference information.For example, before Products Show is started, it may be interacted excessively with user or user was expressed to certain
The concern of item or certain class product.
In addition, it will be appreciated by those skilled in the art that the present invention may be applied in specific industry background.Namely
It says, the interaction with user carries out under specific industry background.For example, user is handed in the online customer service with insurance industry
Mutually, thus produce recommend insurance products demand.Therefore, the step of in interaction for user's recommended products list, includes foundation
User draw a portrait and industry background and be user's recommended products list.
Man-machine interactive system can carry out user's portrait using natural language processing (NLP), utilize machine learning, extensive chemical
It practises and the technologies such as proposed algorithm makes targetedly Products Show.Then, by interaction, product list is presented to use
Family.
In the method for Fig. 1, user is after recommended products list is had viewed, and in step S120, user may provide and answer
It is multiple.Specifically, the present invention, after recommended products list is provided, user can make product list different feedback actions.With
The feedback action at family can be following several situations respectively.
(1) user (such as may illustrate the requirement of product also not complete due to dissatisfied recommendation results or other reasons
Into) be continuing with the modes such as voice or word and express greater demand to product.
(2) user may click the product list of recommendation, continue to check product details.
(3) user is also possible to make other feedbacks, such as：It inquires the background knowledge of recommended products, terminates interaction etc..
In order to be directed to different user feedbacks, rational reply is provided, in the method for the invention, in step S130, sheet
Invention needs to judge the more requirements whether user is expressed to product.It should be appreciated by those skilled in the art that judgement here
The feedback content of extraction user is needed, and for feedback content specifically understand and analyze.That is, in step S130, needle is needed
To the feedback action of user, judge user whether in more requirements of the expression to product.
If the feedback action of user is in more requirements of the expression to product, i.e. the judging result of step S130 is
"Yes" then enters step S140；Otherwise, i.e., the judging result of step S130 is "No", then enters step S150.
In step S140, man-machine interactive system judges that the feedback action of user is more to product in expression
It asks, then needs to extract the product requirement in user's interaction.In user feedback is extracted for the requirement data of product after, carrying
Structural data after taking is supplied to recommending module.Then, adjustment algorithm is required according to newest user by man-machine interactive system,
And updated Products Show list is returned to user.In other words, the handling result of step S140 is to return to step
S110 is user's recommended products list according to user's portrait and newest demand.In fact, due in step S110 before
In, there has been provided recommended products list, so after step S140, it can be understood as, the present invention draws a portrait and uses according to user
The last word requirement at family is updated to the product list of user's recommendation in interaction.
Above and hereafter all refer to " structural data ".Typically structural data such as can be in form
It is such：{"insurance_type":" travel accident insurance ", " user_age ":"35"}.Structural data can be stored in database
In, it is realized with bivariate table structure come logical expression.
User's natural language word is exactly unstructured data, for example " which type of travel accident insurance can be bought within 35 years old
" just belong to unstructured data.Unstructured data includes document, picture, report, video, audio of various forms etc.,
It is exactly the various information obtained in interactive process, but these information programmes can not be used directly.
So we need to extract structural data, it is exactly in interactive process, acquires raw information and needed according to algorithm
It asks and is handled, enable information that can become the data that computer identifies.According to above example, exactly say from user that
Which type of travel accident insurance word " can buy in 35 years old" inner extract " user_age " be 35, " insurance_type " is trip
Trip danger.
Non-structured message structure is allowed the mode that their content is read from the mankind to be converted by this extraction process
The mode that program can parse.
In step S150, that is, the feedback action for judging user is not in more requirements of the expression to product, then specifically sentences
Disconnected user view, and provide corresponding answer.Specifically, man-machine interactive system can be sentenced according to the result of natural language understanding
What disconnected user feedback is intended that, then provides the corresponding answer of system according to user view.
Finally, in step S160, determine whether that this terminates to interact.The situation for terminating interaction is likely to be：It is providing pair
User does not have new reply or user to express END instruction after certain period of time after the answer answered, such as user says and " thanks
Thank ", " goodbye ", " clear " etc..Alternatively, can also after corresponding answer is provided after certain period of time user do not have it is new
During reply, actively ask the user whether also what to be said to confirm that user whether want terminate interaction.If user's
It is intended that end interaction, then whole process terminates.Otherwise, system returns to step S120, and user is waited for provide corresponding action
Feedback.
According to above flow, the method for recommended products can be summarized as following step in human-computer interaction of the invention：
It is user's recommended products list (corresponding to the step S110 in Fig. 1) in interaction；(corresponded in Fig. 1 according to the answer of user
Step S120), judge whether user expresses more requirements (corresponding to the step S130 in Fig. 1) to product；If with
Family expresses more requirements to product, then extracts the product requirement (corresponding to the step S140 in Fig. 1) in user's interaction；With
And required according to the last word of extraction, the product list that user's recommendation is updated in interaction (corresponds to the step in Fig. 1
S110)。
On the other hand, if user does not have to express more requirements to product, judge user view, and provide corresponding
It replies and (corresponds to the step S150 in Fig. 1).
And in a case where, then it can terminate interaction (the step S160 for corresponding to Fig. 1)：Providing corresponding answer
User does not have new reply or user to express END instruction after certain period of time afterwards.
Preferably, be user's recommended products list in the interaction the step of step S110 of Fig. 1 (correspond to) can
To further comprise：It is user's recommended products list according to user's portrait.User's portrait refers to according to user's registration information
Or according to interacted before user and the userspersonal information obtained and preference information.
And after the step S140 corresponding to Fig. 1, it is required according to the last word of extraction, user is updated in interaction
The step of product list of recommendation, (the step S110 for still corresponding to Fig. 1) then may further include：It draws a portrait and uses according to user
The last word requirement at family is updated to the product list of user's recommendation in interaction.
In the step S130 corresponding to Fig. 1, if the answer of user includes situations below, judge that user does not express
To more requirements (the "No" branch of step S130) of product：User clicks the product list recommended to check product details, uses
The background knowledge of family inquiry recommended products or user express END instruction.
In the step S140 of Fig. 1, extraction user is to the structural data of product requirement.Correspondingly, step S140 it
In step S110 afterwards, adjustment recommendation method is required, and updated Products Show list in interaction according to newest user
In return to user.
Mentioned above, human-computer interaction carries out under specific industry background.Therefore, in the step S110 of Fig. 1, according to
It is user's recommended products list according to user's portrait and industry background.
Embodiment 2
It is assumed that user is a man for often carrying out transnational travel for commercial purpose, and man-machine interactive system is specifically being protected
Dangerous field carries out the man-machine interactive system of Products Show.
Under this scene, when user accesses man-machine interactive system and is pushed away it is desirable that obtaining some products from man-machine interactive system
When recommending, man-machine interactive system can actively start interaction and recommend, after can also waiting for the specific query of user's proposition or requiring
Starting interaction and recommending.
It is drawn a portrait according to user, man-machine interactive system can actively initiate significant interaction.Preferably, man-machine interactive system
It can actively inquire user.For example, man-machine interactive system can actively refer to that user often carries out transnational commercial trip in interaction
Capable topic, for example propose that " Mr. X, you often go on business to userWhich country often gone to" this topic is provided
More information, such as user often go to Europe, then can recommend Shen root travel insurance or visa danger etc. to user.
On the other hand, user may initiatively propose that the demand of oneself, such as user may propose that " I will go a moral
State, the insurance kind what has suitable" at this moment, man-machine interactive system can recommend the travelling of Shen root according to this specific requirements of user
Insurance or visa danger etc..Specifically, man-machine interactive system can extract the information such as " Europe ", " insurance kind ", according to algorithm
To generate corresponding recommendation.
In further interaction, man-machine interactive system can actively inquire whether user " has had visa" or
User clearly informs after the recommendation information in relation to visa danger is seen：" having had visa, do not need to visa danger ".In this feelings
Under condition, in the information of recommendation can not include visa danger, and more focused on travel insurance in itself in compensation the amount of money difference.Example
Such as, specific different insurance products can be supplied to user in the form of a link by man-machine interactive system, and user can pass through a little
It hits link and further appreciates that corresponding insurance products；In the case where having to specific product understanding, man-machine interactive system can
To give the user specific product option, so as to user after choosing an option, man-machine interactive system, which provides, further to be had
Close the explanation of rate, compensation, clause, buying pattern, effective date etc..
Embodiment 3
In this embodiment, man-machine interactive system is still that the man-machine interactive system of Products Show is carried out in insurance field.
User may be a new user, so human-computer interaction is at the beginning, man-machine interactive system may inquire user some
About his or her personal information, such as date of birth (age), home background, income situation, health condition, these problems can be with
It is proposed in the form of option, can also be exactly common word dialog.
After having preliminary understanding, man-machine interactive system can be attempted to recommend several different life insurance sides to user
Case.Meanwhile before recommendation, man-machine interactive system can further inquire user which is more paid close attention in terms of guarantee, than
Such as share out bonus, support parents, major disease, hospitalization benefit, order is further accurately provided so as to the answer provided according to user
Customer satisfaction system insurance products.
Certainly, user is possible to because of changes in demand or is managed since statement before is unclear by man-machine interactive system mistake
Solution.At this moment, user may by dialogue inform man-machine interactive system, oneself come be in order to select a vehicle insurance product (rather than
Dividend type life insurance, sickness insurance or endowment insurance).In the case, man-machine interactive system is extracted " vehicle insurance " according to this variation
Such information is regenerated using algorithm or updates the information to be recommended.
Man-machine interactive system can continue to ask some problems, for example requries the users vehicle, age, license plate number etc. and ask
Topic.After obtaining user and replying, recommendation results are provided；Or propose further subdivision problem, for example " three nearly selects：A、5
Ten thousand；B, 100,000；C, 200,000；D, option 500,000 ".User may be found that oneself think choosing 1,000,000 not in option, so not
It can select option and man-machine interactive system " 1,000,000 " can be informed with language form.Man-machine interactive system then can be according to from interaction
The information extracted in content further updates the information of its recommendation.Finally, customer satisfaction system product list or specific product are pushed away
It recommends to user.
In addition, after vehicle insurance has been recommended, if user is satisfied with, man-machine interactive system can be with it is further recommended that other protect
Dangerous project.For example, by interaction, man-machine interactive system is continuously updated user's portrait, for such user, generates corresponding
Life insurance or other type insurance products recommendation, then man-machine interactive system such insurance products can be taken advantage of this interaction machine
Meeting and recommend user.
Furthermore it is also possible to provide the other information alerts related with the insurance products.Man-machine interactive system can provide
Link, option etc. cause user further to obtain new knowledge.Man-machine interactive system can continue to ask the user whether it is to be understood that should
The meaning of insurance products, specific buying pattern, points for attention etc..Obtain user it should be understood that answer after, further to
User's recommendation information.In addition, man-machine interactive system can not also pass through inquiry just initiatively recommends these information to user.
According to above specific embodiment, those skilled in the art should have the present invention one intuitive understanding.Under
Further explanation and extension are made in face of some concepts mentioned in the present invention and specific steps.These are explained and extend some
It is related in the above embodiment, some do not have then.For those unmentioned variations in the above-described embodiments, sheet
Field technology personnel can should as the case may be visualize corresponding scene.
In the present invention, " interaction " is the statement of broad sense, not only word dialog, and the form of human-computer interaction includes text
There are many word, link, voice, expression, picture, video, music etc..
As the mode of human-computer interaction, can there are button, option etc..For example, man-machine interactive system gives user some options,
Including option 1, option 2 and option 3, user is allowed to select input, user inputs " 1 ", man-machine interactive system can provide option 1 pair
The answer answered.
In human-computer interaction, the mode of communication may include these following situations and change：1st, user actively says, human-computer interaction
System is extracted, such as the situation in embodiment 1；2nd, man-machine interactive system may actively inquire user in some cases, than
If user's statement is unclear, man-machine interactive system can be putd question to actively so as to clear and definite user demand；3rd, demand is sent out when interaction is intermediate
Given birth to variation, the thing that user originally wanted now again should not, man-machine interactive system can find this variation of user, again
Start to extract information, recommend new product；4th, according to the portrait of user, the problem of man-machine interactive system is actively inquired can be adjusted.
It, can be according to requiring direct recommended products list during specific recommend；Alternatively, according to demand, first recommend
Product classification after obtaining user's classification hobby, then recommends specific product list.
The information of recommendation is not exclusively to product, it is also possible to be related to servicing.
The processing of recommendation results is then included：If user's expression is satisfied with product list, more multiphase can be obtained
Like recommendation；And if user is dissatisfied to recommending, it needs to recommend again.
It, can be directly to answer if the reply of man-machine interactive system has diversified forms, such as FAQs (FAQ)；
Alternatively, user can be allowed to select to option, user is allowed to inquire details to a link.
In addition, be not only a product or service list according to the information recommended of the present invention, can also provide and the production
Product service related other information alerts.It, can be according to human-computer interaction for example, if user wants to continue to understand other information
Link, option that system provides etc. further obtains new knowledge.Imagine such a scene：In insurance field, pushed away to user
After having recommended serious illness insurance, it also will continue to ask the user whether it is to be understood that the meaning of serious illness insurance, specific buying pattern, points for attention
Deng.By our man-machine interactive system, what user obtained will be a set of complete knowledge for meeting his or she demand.
The information provided in interaction according to user is referred in an embodiment of the present invention come adjustment algorithm, and update is recommended
Information.It is mentioned here that the algorithm of the prior art can both have been included to provide the algorithm of recommendation according to information, such as cooperateed with
Filter algorithm, Item-based/User-based kNN, Matrix Factorization (MF), FM (Factorization
Machines), RBM, GBDT, Random Forest, deep learning (DNN/CNN/RNN), depth enhancing study (DRL) etc.
(for the algorithm of the prior art, those skilled in the art can understand from various open channels and obtain about such algorithm
Introduction, and such algorithm can be implemented in the construction of specific man-machine interactive system, therefore omitted herein to it
It is discussed in detail one by one), it can also include realizing and constantly update the information of recommendation (master by extracting information in interaction
If product or service) any other algorithm and method.
Explanation and extension according to above specific embodiment and to related notion, can be more clearly total by the present invention
Tie and be summarized as a kind of method and apparatus that recommendation information is provided a user by human-computer interaction.It will be directed to such side below
Method and device are described, so that those skilled in the art further appreciate that scope of the present invention and spirit.
Fig. 2 is the flow chart according to the present invention that the method for recommendation information is provided a user by human-computer interaction.
As shown in Figure 2, it is according to the present invention to be started by human-computer interaction to provide a user the method 200 of recommendation information
In step S210, in this step, human-computer interaction is carried out with user.Human-computer interaction can be included through word, voice, expression or chain
Tap into capable interaction.
The human-computer interaction of step S210 is happened in the flow of entire method 200.Preferably, can according to user draw a portrait with
User carries out human-computer interaction.As described above, user's portrait refers to according to user's registration information or according to the friendship with user
The mutually userspersonal information obtained and preference information.More specifically, user's portrait includes：Demographic information, Yong Huli
History behavior, the user geographical location authorized, phone number, address list, list of friends, information on social networks etc..
The beginning of the human-computer interaction of step S210 can be simple greeting or directly recommend without preamble
Information.Preferably, it is described to include actively inquiring user with user's progress human-computer interaction.
In step S220, information is extracted according to interaction content in interactive process.
Preferably, user's portrait can be further updated according to interaction content in interactive process in step S220.
Furthermore, it is possible to extract structural data of the user to product or service request.
In step S230, according to the information of extraction, continuous updating information recommended to the user.
Preferably, it can be drawn a portrait in step S230 according to user, continuous updating information recommended to the user.
According to above specific embodiment, information recommended to the user can include following at least one：Product or service
Tabulation；Specific product or service list；Product or service related information；Product or service related information link；To with
The reply of family problem；For the option of user's selection.
Product recommended to the user can be updated according to user to the structural data adjustment algorithm of product or service request
Or service list.
It will be appreciated by those skilled in the art that the human-computer interaction carries out under specific industry background.Therefore, step
The information according to extraction described in S230, continuous updating information recommended to the user may further include：Letter according to extraction
Breath and industry background and continuous updating product recommended to the user or service list.It should be appreciated by those skilled in the art that although
In the embodiment of specification, the industry background that human-computer interaction is occurred is exemplified as insurance industry, but the industry back of the body in practice
A variety of situations such as scape can also be electric appliances service, fitting-up, seek medical advice and medicine, entertainment recommendations.On the other hand, it is of the invention various
Method is also suitable for these different industry background.
In step S240, in interactive process, newer information is constantly recommended into user.
During to user's recommendation information, if user's expression is satisfied with recommendation information, recommend to user more
Similar information.On the other hand, if user's expression is unsatisfied with recommendation information, adjustment algorithm is needed, is pushed away again to user
Recommend information.
It should be noted that in the method flow diagram of Fig. 2, although each step is sequentially numbered, and according to general
Flow in meaning can first carry out step S210, then step S220, S230, until step S240；But this field skill
Art personnel should be appreciated that in particular embodiments the sequence between each step is it is possible that being interchangeable or adjusting.Example
It such as, can be in the interactive information for starting to be provided with to recommend, then during continuous interaction, continuous updating information.Therefore,
The sequence of step in Fig. 2 can not limit the scope of the invention.
If user is finally satisfied with recommendation results, man-machine interactive system does not need the information recommended more, then man-machine
Interaction can terminate.I.e. method 200 terminates.
Fig. 3 is the schematic block diagram according to the present invention that the device of recommendation information is provided a user by human-computer interaction.
Method and step in Fig. 2 can be completed by specific device.Specifically, according to the present invention pass through man-machine friendship
Mutually and the device 300 of recommended products includes：Recommending module 301, for generating information recommended to the user；Interactive engine 302 is used
In carrying out human-computer interaction (the step S210 for corresponding to Fig. 2) with user, information is extracted according to interaction content in interactive process
(the step S220 for corresponding to Fig. 2), the information recommended to the user that the recommending module generates is presented to the user.
Recommending module 301 is further used for the information extracted according to the interactive engine 302, and continuous updating is recommended to user
The information step S230 of Fig. 2 (correspond to)；Interactive engine 302 is then further used in interactive process, by newer information
It is constantly presented to the user (the step S240 for corresponding to Fig. 2).
It will be appreciated by one of ordinary skill in the art that the method for the present invention can be implemented as computer program.As above it ties
The method that described in Fig. 1, Fig. 2 and Fig. 3, above-described embodiment is performed by one or more programs is closed, to calculate including instructing
Machine or processor perform the method with reference to described in attached drawing.These programs can use various types of non-instantaneous computer-readable Jie
Matter is stored and provided to computer or processor.Non-transitory computer-readable medium includes various types of tangible storage mediums.
The example of non-transitory computer-readable medium includes magnetic recording medium (such as floppy disk, tape and hard disk drive), magneto-optic is remembered
Recording medium (such as magneto-optic disk), CD-ROM (compact disk read-only memory), CD-R, CD-R/W and semiconductor memory are (such as
ROM, PROM (programming ROM), EPROM (erasable PROM), flash rom and RAM (random access memory)).Further, this
A little programs can be supplied to computer by using various types of instantaneous computer-readable mediums.Instantaneous computer-readable Jie
The example of matter includes electric signal, optical signal and electromagnetic wave.Instantaneous computer-readable medium can be used for through such as electric wire and light
Fine wired communication path or wireless communications path provide program to computer.
Therefore, according to the invention, it is further possible to propose a kind of computer program or a kind of computer-readable medium, for recording
Can by processor perform instruction, described instruction is when being executed by processor so that processor perform by human-computer interaction and to
The method that user provides recommendation information, including operating as follows：Human-computer interaction is carried out with user；According in interaction in interactive process
Hold and extract information；According to the information of extraction, continuous updating information recommended to the user；In interactive process, by newer letter
Breath constantly recommends user.
Various embodiments of the present invention and implementation situation are described above.But the spirit and scope of the present invention are not
It is limited to this.Those skilled in the art will it is according to the present invention introduction and make more applications, and these application all this
Within the scope of invention.
That is, the above embodiment of the present invention be only to clearly illustrate examples of the invention rather than to this
The restriction of invention embodiment.For those of ordinary skill in the art, it can also do on the basis of the above description
Go out other various forms of variations or variation.There is no necessity and possibility to exhaust all the enbodiments.It is all in the present invention
Spirit and principle within any modification, replacement or the improvement made etc., should be included in the protection model of the claims in the present invention
Within enclosing.
Claims (10)
1. a kind of method that recommendation information is provided a user by human-computer interaction, including：
Human-computer interaction is carried out with user；
Information is extracted according to interaction content in interactive process；
According to the information of extraction, continuous updating information recommended to the user；
In interactive process, newer information is constantly recommended into user.
2. according to the method described in claim 1, wherein, the human-computer interaction includes tapping by word, voice, expression or chain
Capable interaction.
It is described to carry out human-computer interaction with user and include actively inquiring user 3. according to the method described in claim 1, wherein.
4. according to the method described in claim 1, wherein,
Described further comprises the step of carrying out human-computer interaction with user：Human-computer interaction is carried out with user according to user's portrait,
Described further comprises the step of extracting information according to interaction content in interactive process：The basis in interactive process
Interaction content and further update user portrait,
The step of information according to extraction, continuous updating information recommended to the user, further comprises：It is drawn according to user
Picture, continuous updating information recommended to the user,
Wherein, user's portrait refers to according to user's registration information or the individual subscriber obtained according to the interaction with user
Information and preference information.
5. according to the method described in claim 1, wherein, information recommended to the user includes following at least one：
Product or the tabulation of service；
Specific product or service list；
Product or service related information；
Product or service related information link；
Reply to customer problem；
For the option of user's selection.
6. according to the method described in claim 1, further comprise：
If user's expression is satisfied with recommendation information, recommend more similar information to user；
If user's expression is unsatisfied with recommendation information, to user again recommendation information.
7. according to the method described in claim 1, wherein,
Described further comprises the step of extracting information according to interaction content in interactive process：Extract user to product or
The structural data of service request；
The step of information according to extraction, continuous updating information recommended to the user, further comprises：According to user couple
The structural data of product or service request adjusts recommendation method, updates product or service list recommended to the user.
8. according to the method described in claim 1, wherein, the human-computer interaction carries out under specific industry background, and
The step of information according to extraction, continuous updating information recommended to the user, further comprises：According to extraction
Information and industry background and continuous updating product recommended to the user or service list.
9. a kind of device that recommendation information is provided a user by human-computer interaction, including：
Recommending module is configured for generating information recommended to the user；
Interactive engine is configured for：Human-computer interaction is carried out with user；Letter is extracted according to interaction content in interactive process
Breath；The information recommended to the user that the recommending module generates is presented to the user,
Wherein, the recommending module, which is further configured, is used for：According to the interactive engine extract information, continuous updating to
The information that family is recommended；The interactive engine is further configured that in interactive process, newer information is constantly presented to
User.
10. a kind of computer-readable medium, for recording the instruction that can be performed by processor, described instruction is being executed by processor
When so that the method that processor execution provides a user recommendation information by human-computer interaction, including operating as follows：
Human-computer interaction is carried out with user；
Information is extracted according to interaction content in interactive process；
According to the information of extraction, continuous updating information recommended to the user；
In interactive process, newer information is constantly recommended into user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711431022.3A CN108153879A (en) | 2017-12-26 | 2017-12-26 | The method and device of recommendation information is provided a user by human-computer interaction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711431022.3A CN108153879A (en) | 2017-12-26 | 2017-12-26 | The method and device of recommendation information is provided a user by human-computer interaction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108153879A true CN108153879A (en) | 2018-06-12 |
Family
ID=62462956
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711431022.3A Pending CN108153879A (en) | 2017-12-26 | 2017-12-26 | The method and device of recommendation information is provided a user by human-computer interaction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108153879A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846766A (en) * | 2018-06-25 | 2018-11-20 | 江苏汉德天坤数字技术有限公司 | The self-service Claims Resolution success rate prediction technique of vehicle insurance based on deep learning |
| CN109102855A (en) * | 2018-07-03 | 2018-12-28 | 北京康夫子科技有限公司 | Drug recommended method |
| CN109242514A (en) * | 2018-08-28 | 2019-01-18 | 腾讯科技（深圳）有限公司 | Client's label recommendation method, device and system |
| CN109684466A (en) * | 2019-01-04 | 2019-04-26 | 钛氧(上海)教育科技有限公司 | A kind of intellectual education advisor system |
| CN110458742A (en) * | 2019-08-06 | 2019-11-15 | 田祚玉 | A kind of inhabitation circle, enterprise, government information intercommunion platform |
| CN110489623A (en) * | 2019-07-10 | 2019-11-22 | 本识科技(深圳)有限公司 | A kind of intelligent assistant's system and intelligent assistant robot based on user information interaction |
| CN111159558A (en) * | 2019-12-31 | 2020-05-15 | 支付宝(杭州)信息技术有限公司 | Recommendation list generation method and device and electronic equipment |
| CN111177413A (en) * | 2019-12-30 | 2020-05-19 | 北京谦仁科技有限公司 | Learning resource recommendation method and device and electronic equipment |
| CN111475013A (en) * | 2019-01-24 | 2020-07-31 | 阿里巴巴集团控股有限公司 | Commodity object information processing method and device and electronic equipment |
| CN111667361A (en) * | 2020-06-19 | 2020-09-15 | 上海印闪网络科技有限公司 | Loan product recommendation method based on user behaviors |
| CN111752932A (en) * | 2020-08-31 | 2020-10-09 | 江苏智谋科技有限公司 | Client data processing system and method based on knowledge management |
| CN112380334A (en) * | 2020-12-04 | 2021-02-19 | 三星电子（中国）研发中心 | Intelligent interaction method and device and intelligent equipment |
| CN113706244A (en) * | 2021-08-26 | 2021-11-26 | 哈尔滨工业大学（威海） | Implementation method of intelligent clothes management collocation recommendation system based on end cloud fusion |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208086A (en) * | 2010-03-31 | 2011-10-05 | 北京邮电大学 | Field-oriented personalized intelligent recommendation system and implementation method |
| CN103544219A (en) * | 2013-09-24 | 2014-01-29 | 北京光年无限科技有限公司 | Question-answering system with intelligent recommendation |
| CN106448670A (en) * | 2016-10-21 | 2017-02-22 | 竹间智能科技（上海）有限公司 | Dialogue automatic reply system based on deep learning and reinforcement learning |
| CN106559321A (en) * | 2016-12-01 | 2017-04-05 | 竹间智能科技（上海）有限公司 | The method and system of dynamic adjustment dialog strategy |
| US20170132230A1 (en) * | 2015-11-09 | 2017-05-11 | WP Company LLC d/b/a The Washington Post | Systems and methods for recommending temporally relevant news content using implicit feedback data |
-
2017
- 2017-12-26 CN CN201711431022.3A patent/CN108153879A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208086A (en) * | 2010-03-31 | 2011-10-05 | 北京邮电大学 | Field-oriented personalized intelligent recommendation system and implementation method |
| CN103544219A (en) * | 2013-09-24 | 2014-01-29 | 北京光年无限科技有限公司 | Question-answering system with intelligent recommendation |
| US20170132230A1 (en) * | 2015-11-09 | 2017-05-11 | WP Company LLC d/b/a The Washington Post | Systems and methods for recommending temporally relevant news content using implicit feedback data |
| CN106448670A (en) * | 2016-10-21 | 2017-02-22 | 竹间智能科技（上海）有限公司 | Dialogue automatic reply system based on deep learning and reinforcement learning |
| CN106559321A (en) * | 2016-12-01 | 2017-04-05 | 竹间智能科技（上海）有限公司 | The method and system of dynamic adjustment dialog strategy |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846766A (en) * | 2018-06-25 | 2018-11-20 | 江苏汉德天坤数字技术有限公司 | The self-service Claims Resolution success rate prediction technique of vehicle insurance based on deep learning |
| CN109102855A (en) * | 2018-07-03 | 2018-12-28 | 北京康夫子科技有限公司 | Drug recommended method |
| CN109242514A (en) * | 2018-08-28 | 2019-01-18 | 腾讯科技（深圳）有限公司 | Client's label recommendation method, device and system |
| CN109684466A (en) * | 2019-01-04 | 2019-04-26 | 钛氧(上海)教育科技有限公司 | A kind of intellectual education advisor system |
| CN109684466B (en) * | 2019-01-04 | 2023-10-13 | 钛氧(上海)教育科技有限公司 | Intelligent education advisor system |
| CN111475013A (en) * | 2019-01-24 | 2020-07-31 | 阿里巴巴集团控股有限公司 | Commodity object information processing method and device and electronic equipment |
| CN110489623A (en) * | 2019-07-10 | 2019-11-22 | 本识科技(深圳)有限公司 | A kind of intelligent assistant's system and intelligent assistant robot based on user information interaction |
| CN110458742A (en) * | 2019-08-06 | 2019-11-15 | 田祚玉 | A kind of inhabitation circle, enterprise, government information intercommunion platform |
| CN111177413A (en) * | 2019-12-30 | 2020-05-19 | 北京谦仁科技有限公司 | Learning resource recommendation method and device and electronic equipment |
| CN111159558B (en) * | 2019-12-31 | 2023-07-18 | 支付宝(杭州)信息技术有限公司 | Recommendation list generation method and device and electronic equipment |
| CN111159558A (en) * | 2019-12-31 | 2020-05-15 | 支付宝(杭州)信息技术有限公司 | Recommendation list generation method and device and electronic equipment |
| CN111667361A (en) * | 2020-06-19 | 2020-09-15 | 上海印闪网络科技有限公司 | Loan product recommendation method based on user behaviors |
| CN111752932A (en) * | 2020-08-31 | 2020-10-09 | 江苏智谋科技有限公司 | Client data processing system and method based on knowledge management |
| CN111752932B (en) * | 2020-08-31 | 2021-06-29 | 江苏智谋科技有限公司 | Client data processing system and method based on knowledge management |
| CN112380334A (en) * | 2020-12-04 | 2021-02-19 | 三星电子（中国）研发中心 | Intelligent interaction method and device and intelligent equipment |
| CN112380334B (en) * | 2020-12-04 | 2023-03-24 | 三星电子（中国）研发中心 | Intelligent interaction method and device and intelligent equipment |
| CN113706244A (en) * | 2021-08-26 | 2021-11-26 | 哈尔滨工业大学（威海） | Implementation method of intelligent clothes management collocation recommendation system based on end cloud fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108153879A (en) | The method and device of recommendation information is provided a user by human-computer interaction | |
| US10726427B2 (en) | Method and apparatus for building a user profile, for personalization using interaction data, and for generating, identifying, and capturing user data across interactions using unique user identification | |
| Munn | Angry by design: toxic communication and technical architectures | |
| US10878198B2 (en) | Intent arbitration for a virtual assistant | |
| US10460024B2 (en) | Interactive electronic form workflow assistant that guides interactions with electronic forms in a conversational manner | |
| US9661067B2 (en) | Systems and methods for facilitating dialogue mining | |
| US10885529B2 (en) | Automated upsells in customer conversations | |
| CN115398437A (en) | Improved out-of-domain (OOD) detection techniques | |
| EP3799687B1 (en) | Recording evidence of communication in human-machine interactions | |
| CN111625632A (en) | Question-answer pair recommendation method, device, equipment and storage medium | |
| EP2944077B1 (en) | Method and apparatus for analyzing leakage from chat to voice | |
| CN104268185B (en) | A kind of method and apparatus in application distribution platform search application | |
| KR20150060534A (en) | Method and system for creating an intelligent digital self representation | |
| WO2024066253A1 (en) | Interactive fiction-based product recommendation method and related apparatus | |
| CN105518661A (en) | Browsing images via mined hyperlinked text snippets | |
| US20220114187A1 (en) | Techniques for searching using target applications | |
| US11120455B2 (en) | Intelligent channel steering based on dynamically determined intents during channel interactions | |
| CN111144906A (en) | Data processing method and device and electronic equipment | |
| CN114341922A (en) | Information processing system, information processing method, and program | |
| Liu | Hanging out with my pandemic pal: contextualizing motivations of anthropomorphizing voice assistants during COVID-19 | |
| US20220343186A1 (en) | Information processing device and information processing method | |
| CN117131179A (en) | Dialogue processing method and device, storage medium and electronic equipment | |
| CN117350799A (en) | Method for providing commodity comparison information and electronic equipment | |
| KR20180042763A (en) | Chatting type financial robot and method for providing financial service using the same | |
| JP2005339368A (en) | Emotion grasping system and emotion grasping method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180612 |
|
| RJ01 | Rejection of invention patent application after publication |