CN110308947A - Processing method, system, computer equipment and the storage medium of question and answer information - Google Patents
Processing method, system, computer equipment and the storage medium of question and answer information Download PDFInfo
- Publication number
- CN110308947A CN110308947A CN201910440492.9A CN201910440492A CN110308947A CN 110308947 A CN110308947 A CN 110308947A CN 201910440492 A CN201910440492 A CN 201910440492A CN 110308947 A CN110308947 A CN 110308947A
- Authority
- CN
- China
- Prior art keywords
- information
- page
- mark
- question
- answer
- 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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
- G06F9/44521—Dynamic linking or loading; Link editing at or after load time, e.g. Java class loading
- G06F9/44526—Plug-ins; Add-ons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The present invention relates to semantic analytic technique field, the present invention provides processing method, system, computer equipment and the storage medium of a kind of question and answer information, the setting flag input control on the page of product introduction;When marking input control to be triggered, the label that user is inputted on the page, and the trace of recording mark and position are obtained;The problem of identifying user according to the page of the trace of label and position and product introduction information;Problem information is inputted into Question-Answering Model, obtain and exports target answer in the page of product introduction.The above method can be convenient and simply input on the page of current product introduction the label for being used for asked questions, the problem of proposed information is identified according to the content information of label and the page, the easy process for proposing problem information, improves the efficiency of question answering process;By obtaining the target answer in close relations with the content information of the page, improving target answer accuracy to having the Question-Answering Model for reading the intelligence for understanding function to input problem information.
Description
Technical field
The present invention relates to semantic analytic technique fields, specifically, the present invention relates to a kind of processing method of question and answer information,
Processing system, computer equipment and the storage medium of question and answer information.
Background technique
With the raising of people's quality of life and the progress of society's level, everybody is also more next to the attention rate of product quality
More pay attention to, the introduction and displaying content to product are conscientiously meticulously read.Sometimes the content information of product is highly professional
Reading content, often inevitably can propose problem because of indigestion when browsing product.Especially increasingly by the public's
The insurance products of welcome, insurance act in insurance products it is professional very strong, insurance act is easy to appear more in reading
Problem.
And generally when reading the page for showing product, it is contemplated that need in the page of the customer service dialog box jumped after query
Input problem, but usually excessive because operating, and have forgotten how the problem of to be putd question to states, cause in order to put question to and back and forth more
A page is turned back, very the consuming time, is caused the process for proposing problem information complicated, is led to the low efficiency of question answering process.
Summary of the invention
The purpose of the present invention is intended at least can solve above-mentioned one of technological deficiency, the especially low skill of problem process efficiency
Art defect.
The present invention provides a kind of processing method of question and answer information, includes the following steps:
The setting flag input control on the page of product introduction, wherein the label input control is for recording and showing
Show the trace of the label generated on the page;
When the label input control is triggered, the label that user is inputted on the page is obtained, and record institute
State trace and the position of label;
The problem of identifying the user according to the page of the trace of the label and position and product introduction information;
By described problem information input Question-Answering Model, obtains and exports target answer in the page of the product introduction,
In, the model for the problem of Question-Answering Model is for answering the page shown product.
In one embodiment, the label includes touch track or mouse track.
In one embodiment, described to be identified according to the page of the trace of the label and position and the product introduction
The step of the problem of user information, comprising:
The mark information is divided into several marks according to the trace of the label and position;Identify each mark
Token types;According to the token types of the page of the product introduction and each mark, each note is identified one by one respectively
Number corresponding text information;Described problem information is generated according to the text information.
In one embodiment, described according to the page of the product introduction and the token types of each mark, respectively
The step of identifying the corresponding text information of each mark one by one, comprising:
When the token types of the mark are text track, the mark is converted into handwriting image, identifies the pen
Text information in mark image, using the text information in the handwriting image as the corresponding text information of the mark;
The mark type be underscore or circles mark when, according to the mark the product introduction the page
In position, the corresponding text information in the position is extracted from the content of text of the product introduction, the position is corresponding
Text information as the corresponding text information of the mark.
In one embodiment, described the step of described problem information is generated according to the text information, comprising:
Identify the text and punctuation mark of the text information;Obtain the text, the punctuation mark and the mark
Between mark position;According to the text, the punctuation mark and the mark position, described problem letter is generated in conjunction with semanteme
Breath.
In one embodiment, the step of page in the product introduction exports target answer, comprising:
Obtain preset answer output mode;When the answer output mode is voice output, the Xiang Suoshu page is returned
The corresponding voice signal of the target answer;When the answer output mode is text output, institute is shown in the page
State the corresponding text information of target answer.
In one embodiment, the page of the product introduction is for showing insurance clause, and the Question-Answering Model is for solving
Answer answer of the described problem information under the requirement of the insurance clause.
The present invention also provides a kind of processing systems of question and answer information, comprising:
Setup module, for the setting flag input control on the page of product introduction, wherein the label input control
For recording and showing the trace of the label generated on the page;
Module is obtained, for when the label input control is triggered, obtaining what user was inputted on the page
Label, and record trace and the position of the label;
Identification module, for identifying the use according to the page of the trace of the label and position and the product introduction
The problem of family information;
Output module, for obtaining described problem information input Question-Answering Model and defeated in the page of the product introduction
Target answer out, wherein the model for the problem of Question-Answering Model is for answering the page shown product.
The present invention also provides a kind of computer equipment, including memory and processor, calculating is stored in the memory
Machine readable instruction, when the computer-readable instruction is executed by the processor, so that the processor executes such as any implementation
The step of processing method of the example question and answer information.
The present invention also provides a kind of storage mediums for being stored with computer-readable instruction, and the computer-readable instruction is by one
When a or multiple processors execute, so that one or more processors execute the processing method of question and answer information described in any embodiment
The step of.
Processing method, system, computer equipment and the storage medium of above-mentioned question and answer information can be convenient and simply exist
Input is used for the label of asked questions on the page of current product introduction, according to the content information of label and the page to identify
The problem of proposition information, the efficiency of question answering process can be improved in the easy process for proposing problem information.And by having
The Question-Answering Model input problem information for reading the intelligence for understanding function, can obtain the mesh in close relations with the content information of the page
Answer is marked, the accuracy of target answer is improved.
The additional aspect of the present invention and advantage will be set forth in part in the description, these will become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is the implementation environment figure of the processing method of the question and answer information provided in one embodiment;
Fig. 2 is the flow chart of the processing method of question and answer information in one embodiment;
Fig. 3 is the flow chart that problem information is identified in one embodiment;
Fig. 4 is the page figure of the product introduction of tape label in one embodiment;
Fig. 5 is the flow chart of the processing method of question and answer information in another embodiment;
Fig. 6 is the structural schematic diagram of the processing system of question and answer information in one embodiment;
Fig. 7 is the schematic diagram of internal structure of computer equipment in one embodiment.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art
Language and scientific term), there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, it should be understood that have in the context of the prior art
The consistent meaning of meaning, and unless idealization or meaning too formal otherwise will not be used by specific definitions as here
To explain.
As shown in FIG. 1, FIG. 1 is the implementation environment figures of the processing method of the question and answer information provided in one embodiment, at this
In implementation environment, including server-side 110 and client 120.
Server-side 110 can be run in computer equipment or server apparatus, and client 120 may be mounted at terminal
On, Question-Answering Model can be configured in server-side 110 or client 120, which can have the language that text is read in identification
Justice will read text and be contacted with institute input problem and export the ability of corresponding answer.In the terminal mounted of client 120
Having the function of data input, such as terminal can connect mouse and record mouse track, terminal can also be equipped with touch screen,
The touch screen can recorde the touch track of user.Terminal can be smart phone, tablet computer, laptop, desk-top calculating
Machine etc., however, it is not limited to this.It can be communicated by network between server-side 110 and client 120.
In one embodiment, as shown in Fig. 2, Fig. 2 is the flow chart of the processing method of question and answer information in one embodiment,
A kind of processing method of question and answer information is proposed in the present embodiment, the processing method of the question and answer information can be applied to above-mentioned visitor
In family end 120, it can specifically include following steps:
Step S210: the setting flag input control on the page of product introduction, wherein label input control is for recording
And it is shown in the trace of the label generated on the page.
In this step, client can be to show product on the page, the product introduction of client is arranged in label input control
The page on, during browsing the page of current product introduction, user can directly trigger the label in current page
Input control can carry out the input of information by label input control in the page, search without jumping to other pages
The label input control.
Step S220: when marking input control to be triggered, the label that user is inputted on the page is obtained, and record mark
The trace of note and position.
When user browses the page of product introduction, the mobile trace of mobile trace or mouse is touched on the touch screen of terminal
Mark can generate the trace of label on the page.In this step, after label input control triggering, client can recorde user
The mobile trace of mobile trace or mouse is touched in the page of browsing product introduction, the trace of these movements is defeated as institute
The label entered.
The input marking on the page of product introduction, user can need to avoid when proposing asked questions in product introduction
It turns back and checks in the page and the page of customer service dialog box, reduce the round-trip process of the page;And user can also reduce tissue and
The time of the text information of input problem improves the efficiency of enquirement.
User can be as made of several discontinuous ichnocoenosis in certain label inputted, can recorde label
In the position of each trace and each trace.Part is more concentrated in position or track has staggered part trace,
This part trace can be combined, and record the integral position of the part trace.In addition client can also record the mark
Remember the position in the shared region in displayed page, the position can be associated with the layout in displayed page at this time, subsequent visitor
Family end can improve the accuracy of the problem of identified information according to the association come information the problem of identification.
Further, label may include touch track and/or mouse track.When user's browsing client product introduction
When page terminal mounted is the display equipment with touch screen function, during marking input control to enable, client
The touch in touch display screen can be incuded, while recording the trace of the touch, is i.e. record touch track, by the trace of the touch
As label, and identify the position for the page that the label is shown relative to product.In addition, working as the product exhibition of user's browsing client
When the page shown, if the client terminal mounted of browsing is in such a way that mouse is clicked come triggered mark input control
With input marking, then client can be marked by the trace of record mouse click, i.e., by the page of product introduction
In the label that is inputted as user of mouse track, while also identifying the position for the page that the label is shown relative to product.
Some terminals for having both external mouse input and touch screen function are mounted in for client, client will can touch rail simultaneously
Mark and mouse track are recorded, the label inputted on the page as user.It, can be with according to client equipment mounted
Adaptively select touch track and/or mouse track label as input.
At this point, the processing method of question and answer information can use touch track or mouse rail according to the capacity of equipment of client
For mark as the label inputted, the interactivity of this label input mode is strong, can be convenient by touching contact or mouse action
And label relevant to question information is inputted with heart, for being unfamiliar with the user of the product introduction page, user also can be quick
Grasp the label how inputted for enquirement.
Step S230: the problem of user is identified according to the page of the trace of label and position and product introduction information.
In this step, client extracts the trace of label and identifies position of the trace in the product introduction page, institute
The label of acquisition can be used for identifying problem information expressed by user, and the trace of institute's recording mark and position can be with intermediate descriptions
Text information involved in problem information.
For example, the trace of label may include text handwriting, symbol person's handwriting and marker graphic, thus show that user is stated
Text information;Position in addition combined with label and on the page of product introduction after content information, client can also identify accordingly
Mark referred to text information;Or the language of text information that client can also be identified according to the position analysis of label
Justice.
Step S240: inputting Question-Answering Model for problem information, obtains and exports target answer in the page of product introduction,
In, the model for the problem of Question-Answering Model is for answering the page shown product.
In this step, the problem of client will be identified information input Question-Answering Model, Question-Answering Model has the reading of intelligence
Understand function.Question-Answering Model can interpret the content information that the page is shown, the content information pair shown in conjunction with the page
Problem information is associated and analyzes, and obtains the target answer for answering the problem information, i.e. acquisition target answer.Question and answer mould
If type configures in client, client calls directly Question-Answering Model, generates the problem of identified information by the Question-Answering Model
Corresponding target answer;If Question-Answering Model configures in server-side, client sends the problem of institute identifies letter to server-side
Breath, and receive the corresponding target answer of the problem of institute that server issues identifies information.
Target answer can also be exported in this step in the page of the product introduction of client, target pages can be
Regional area is exported in the page, and the content information of target answer and product introduction can compare in the same page to be read
It reads, user can receive the target answer while reading the page of product introduction, and user with Fast Reading and can understand mesh
The content information of answer and product introduction is marked, reading efficiency is improved.
The processing method of above-mentioned question and answer information can be convenient and simply input on the page of current product introduction and uses
In the label of asked questions, the problem of proposed information is identified according to the content information of label and the page, easy proposition is asked
The process for inscribing information, can be improved the efficiency of question answering process.And pass through the question and answer mould to the intelligence for understanding function with reading
Type inputs problem information, can obtain the target answer in close relations with the content information of the page, improves the accurate of target answer
Degree.
Further, before identifying problem information, perhaps before exporting target answer or in output target answer
When, identified problem information can be shown in regional area, in the page convenient for understanding target answer according to problem information.
Further, the output form of target answer can be selected according to user or preset answer output model come
It determines.In one embodiment, may include: the step of the page of product introduction exports target answer in step S240
(1) preset answer output mode is obtained.
In this step, preset answer output mode is first searched, determines answer output mould that user selects or preset
The corresponding information format of formula exports the target answer under the information format.It below will be according to the corresponding information of answer output mode
Format converts target answer.
(2) when answer output mode is voice output, the corresponding voice signal of target answer is returned to the page.
It can be with clear when listening by the voice broadcast target answer, user in this step, under the mode of voice output
The mode of the page of looking at understands target answer, keeps the page for resting on the product introduction currently browsed, reduces jumping for the page.It can
The corresponding voice signal of target answer is played for triggering to center in some of the page setup of product introduction
Voice control.
Further, which can be also used for the voice signal of acquisition user's input.In the language for obtaining the input
After sound signal, the problem of including in the voice signal of input information is obtained by speech recognition, and execute problem information is defeated
Enter Question-Answering Model, obtain and the step of the page of product introduction exports target answer, completes mode by voice input, into
Row problem is putd question to and answer.
(3) when answer output mode is text output, the corresponding text information of displaying target answer in the page.
In this step, the target answer can be shown under the mode of text output by text information, in product introduction
Text display window is set on local location in the page, and text display window is displayed for the target answer, shows simultaneously
Show the content information of product introduction, can in the same page displaying target answer and product introduction content information, reduce
The page jumps.
The processing method of above-mentioned question and answer information can provide a variety of answer output modes, export according to preset answer
Mode exports the target answer under corresponding informance format, adapts to the reading habit of user;It can also be when exporting target answer
Jumping for the page is reduced or avoided, the content information shown convenient for combination product understands target answer.
In one embodiment, as shown in figure 3, Fig. 3 is the flow chart for identifying problem information in one embodiment, step
It the step of the problem of user is identified according to the page of the trace of label and position and product introduction in S230 information, can wrap
It includes:
Step S231: mark information is divided by several marks according to the trace of label and position.
Can wrap in the label of certain output containing several discontinuous traces dragged or touch, but these traces it
Between there are incidence relations, such as the shape of text or symbol can be shown after the ichnocoenosis of regional area, some traces
It can also show the figure of written notes.One group of trace with incidence relation can be divided into a note in the markers
Number.
In this step, the incidence relation or feature between trace can be judged by the trace and position of label.Inhomogeneity
The mark of type has specific feature, by taking the mark of text as an example, may be constructed meeting relative close in position between the trace of text,
And overlapping is usually had between trace;For crossing below, the length of trace can be relatively long;By taking circles mark as an example, trace
The curvature and curvature of mark can be larger.Then the trace in label is divided by several notes according to the incidence relation between trace
Number.
Step S232: the token types of each mark are identified.
In this step, the mark of mark can be judged according to the incidence relation between the feature or trace of trace in mark
Type identifies the token types of each mark.
Step S233: according to the token types of the page of product introduction and each mark, identify that each mark is corresponding one by one respectively
Text information.
It can include referred to text information in each mark, the token types of each mark further identify that it is referred to
Text.Since mark is generated in the page of product introduction, some marks need that the content information in the page is combined to look into
The page looked for referred to text information, therefore combination product is also needed to show carries out the knowledge of the corresponding text information of each mark
Not.
Specifically, according to the token types of the page of product introduction and each mark, identify that each mark is corresponding one by one respectively
The step of text information may include:
A1: when the token types of mark are text track, mark is converted into handwriting image, is identified in handwriting image
Text information, using the text information in handwriting image as the corresponding text information of mark.
In this step, when mark belongs to text track, the trace of mark, which is superimposed, can be visually observed that the note
Number corresponding text or symbol, recycle image procossing to identify the text or symbol, regard the text identified or symbol as this
The corresponding text information of mark.
A2: when the type of mark is underscore or circles mark, according to position of the mark in the page of product introduction,
The corresponding text information in position is extracted from the content of text of product introduction, it is corresponding using the corresponding text information in position as mark
Text information.
In this step, graphic sign may include underscore or circle, be shown in figure note according to graphic sign lookup
Number position on text information.
In a kind of wherein embodiment, the row where text information can be determined according to the position of underscore, under
The head and the tail position of scribing line determines the initial position of text information from the row, in the content information in the page from product introduction
Extract the text information;In addition the text information that circle is covered can also be determined according to the position of circle, this is capped
Text information is extracted from the content information in the page of product introduction.
In a kind of wherein embodiment, graphic sign position can also be intercepted in the page of product introduction
Product introduction image identifies text and symbol in the product introduction image of the interception by image procossing, by the text of identification
With symbol as the corresponding text information of mark.
The embodiment of the corresponding text information of above-mentioned each mark of identification, efficiently and accurately can be converted to text for label
Word information;The text information of acquisition can be identified by computer and Question-Answering Model, in order to subsequent calls Question-Answering Model, be improved
The efficiency and accuracy of question answering.
How the page-describing for specifically having markd product introduction with one is identified into the corresponding text of each mark below
Information, as shown in figure 4, Fig. 4 is the page figure of the product introduction of tape label in one embodiment.Wherein, mark 1, mark 2, note
Numbers 3 and label of the mark 4 as the product introduction page.For mark 1, mark 2 and mark 4, mark is converted into figure respectively
Picture, and carry out image procossing, identify text " more ", text " long " and symbol "? ".For mark 3, due to the length of mark 3
Longer, it is lower that mark 3 deviates horizontal angle, it is possible to determine that the type of mark 3 is underscore;Mark 3 is obtained relative to production
The position of product displayed page and the head and the tail position of mark 3, from being extracted in the content information of the page " to this main danger after waiting period
It is that expiration of contract occurs a few days ago and extend to after this main dangerous expiration of contract in 30 days " text information.
Step S234: problem information is generated according to text information.
In this step, text information is combined and is handled according to the syntax rule of natural language, generation meets nature
The problem of language semantic information.
Specifically, according to text information generation problem information the step of, may include:
B1: the text and punctuation mark of text information are identified.
B2: the mark position between text, punctuation mark and mark is obtained.
B3: according to text, punctuation mark and mark position, problem information is generated in conjunction with semanteme.
It is illustrated by taking the page of the product introduction of the tape label of Fig. 4 as an example: due to the position of text information " more " and " long "
Solution is set, text information group can be combined into " how long ", while being also able to satisfy the semanteme of natural language.Will "? " it is accorded with as punctuate
Number, according to the syntactic rule of natural language, "? " it can be in a manner of at the end of problem information." to this main dangerous contract after waiting period
It is occurring before Expiration Date and extend to after this main dangerous expiration of contract in 30 days " in comprising the word with time meaning, can be with
Judge that this section of text information is related with the time, and " how long " also related with the time, " how long " be located at " after waiting period ... "
Left side can combine to obtain " that this main dangerous expiration of contract after waiting period occurs a few days ago and after extending to this main dangerous expiration of contract
It how long is in 30 days? " or it is " how long this main dangerous expiration of contract after waiting period being occurred a few days ago and extend to this main dangerous contract and arrive
After phase in 30 days ", it is handled according to the syntax rule of natural language, coherent table can be used for the text addition after combination
It expresses one's ideas the word of think of, the sentence with maximum likelihood is calculated, using the sentence of the maximum likelihood as problem information.
In the embodiment of above-mentioned generation problem information, it can with artificial intelligence identify the problem of user expresses information, improve
The efficiency and intelligence degree of problem information input.
In one embodiment, the page of product introduction is for showing insurance clause, and Question-Answering Model is for letter of answering a question
Cease the answer under the requirement of insurance clause.
It being illustrated by taking the page of the product introduction of the tape label of Fig. 4 as an example, the page of product introduction shows insurance clause,
But insurance clause is professional very strong, reading gets up the case where indigestion usually occur, to indicate clearly to want enquirement
Problem be also it is highly difficult, such as Fig. 4 product introduction the page in " this main dangerous expiration of contract after waiting period is occurred a few days ago
And extend to after this main dangerous expiration of contract in 30 days " in comprising multiple terms, user needs when needing to put question to the clause
The clause, round-trip Terms page and customer service dialog box are repeated, this has certain description for not understanding when the user of clause comes
The difficulty of problem, the process that also significantly complicated user puts question to.
In the present embodiment, when being marked at the clause, so that it may automatically according to the marker recognition clause and generation
The problem of client is about the clause, easy user propose the process of problem.And Question-Answering Model can have natural language
Reading comprehension, in conjunction with the clause of the insurance products practical semanteme analyzed after, problem information is answered, institute it is defeated
Target answer out and the relevance of the insurance clause are strong, and the accuracy of target answer is high.
The accuracy of answer insurance act problem can be improved in the processing method of above-mentioned question and answer information, improves insurance products
Customer service quality.
In another embodiment, as shown in figure 5, Fig. 5 is the stream of the processing method of question and answer information in another embodiment
Cheng Tu.A kind of processing method of question and answer information is proposed in the present embodiment, the processing method of the question and answer information can be applied to
In the client 120 stated, client 120 is used to show the insurance act of insurance products by the page, can specifically include following
Step:
Step S510: the setting flag input control on the insurance act page of insurance products.Wherein, input control is marked
For recording and being shown in the touch track generated on the insurance act page and/or mouse track.
Step S520: when marking input control to be triggered, the touch that user is inputted on the insurance act page is obtained
Track and/or mouse track, and record trace and the position of touch track and/or mouse track.
Step S530: according to the trace and position and the insurance act page of touch track and/or mouse track, identification is used
The insurance products problem at family.
Specifically, several marks are divided into according to the trace and position of touch track and/or mouse track.Identification
The token types of each mark.
According to the token types of the insurance act of the insurance act page and each mark, identify that each mark is corresponding one by one respectively
Text information.When the token types of mark are text track, mark is converted into handwriting image, identifies the text in handwriting image
Word information, using the text information in handwriting image as the corresponding text information of mark;It is underscore or circle in the type of mark
When circle label, according to position of the mark in the page of product introduction, it is corresponding that position is extracted from the content of text of insurance act
Text information, using the corresponding text information in position as the corresponding text information of mark.
Insurance products problem is generated according to text information.Identify the text and punctuation mark of text information;Obtain text, mark
Mark position between point symbol and mark;According to text, punctuation mark and mark position, insurance products are generated in conjunction with semanteme and are asked
Topic.
Step S540: Question-Answering Model is used to combine the answer of the insurance act answer insurance products problem currently browsed, will
Insurance products problem inputs Question-Answering Model, obtains and exports target answer in the insurance act page.
In the step of insurance act page exports target answer, it can specifically include: obtaining preset answer output mode;
When answer output mode is voice output, the corresponding voice signal of target answer is returned to the insurance act page;It is defeated in answer
When mode is text output out, the corresponding text information of displaying target answer in the insurance act page.
The processing method of above-mentioned question and answer information, can be convenient and simply input is protected for reflecting in the insurance act page
The label of dangerous product problem identifies proposed insurance products problem according to label and the content information of insurance act, easy
The process for proposing insurance products problem, can be improved the efficiency of the question answering process of insurance products problem.And Question-Answering Model has
There is reading and understanding insurance act function, inputs insurance products problem to Question-Answering Model, the content with insurance act can be obtained
The close target answer of information relationship improves the accuracy of target answer.
In one embodiment, as shown in fig. 6, the structure that Fig. 6 is the processing system of question and answer information in one embodiment is shown
It is intended to, proposes a kind of processing system of question and answer information in the present embodiment, the processing system of the question and answer information can specifically include
Setup module 610 obtains module 620, identification module 630 and output module 640, in which:
Setup module 610, for the setting flag input control on the page of product introduction, wherein label input control
For recording and being shown in the trace of the label generated on the page.
610 client of setup module can be to show product on the page, the product exhibition of client is arranged in label input control
On the page shown, during browsing the page of current product introduction, user can directly trigger the mark in current page
Remember input control, the input of information can be carried out by label input control in the page, looked into without jumping to other pages
Look for the label input control.
Module 620 is obtained, for when marking input control to be triggered, obtaining the label that user is inputted on the page,
And trace and the position of recording mark.
When user browses the page of product introduction, the mobile trace of mobile trace or mouse is touched on the touch screen of terminal
Mark can generate the trace of label on the page.Module 620 is obtained after label input control triggering, client can recorde
User touches mobile trace or the mobile trace of mouse in the page of browsing product introduction, using the trace of these movements as
The label inputted.
Identification module 630, for according to the page of the trace of label and position and product introduction identify user the problem of
Information.
Identification module 630 extracts the trace of label and identifies position of the trace in the product introduction page, acquired
Label can be used for identifying problem information expressed by user, the trace of institute's recording mark and position can be with intermediate description problems
Text information involved in information.
Output module 640, for problem information to be inputted Question-Answering Model, obtain and exports target in the page of product introduction
Answer, wherein the model for the problem of Question-Answering Model is for answering the page shown product.
There is the reading of intelligence to understand for the problem of output module 640 will be identified information input Question-Answering Model, Question-Answering Model
Function.Question-Answering Model can interpret the content information that the page is shown, the content information shown in conjunction with the page is to problem
Information is associated and analyzes, and obtains the target answer for answering the problem information, i.e. acquisition target answer.If Question-Answering Model
Configuration is in client, then client calls directly Question-Answering Model, generates the problem of identified information by the Question-Answering Model and corresponds to
Target answer;If Question-Answering Model configures in server-side, client sends the problem of institute identifies information to server-side, and
Receive the problem of institute that server issues the identifies corresponding target answer of information.
The processing system of above-mentioned question and answer information can be convenient and simply input on the page of current product introduction and uses
In the label of asked questions, the problem of proposed information is identified according to the content information of label and the page, easy proposition is asked
The process for inscribing information, can be improved the efficiency of question answering process.And pass through the question and answer mould to the intelligence for understanding function with reading
Type inputs problem information, can obtain the target answer in close relations with the content information of the page, improves the accurate of target answer
Degree.
The specific of processing system about question and answer information limits the processing method that may refer to above for question and answer information
Restriction, details are not described herein.Modules in the processing system of above-mentioned question and answer information can be fully or partially through software, hard
Part and combinations thereof is realized.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment,
It can also be stored in a software form in the memory in computer equipment, execute the above modules in order to which processor calls
Corresponding operation.
As shown in fig. 7, Fig. 7 is the schematic diagram of internal structure of computer equipment in one embodiment.The computer equipment packet
Include processor, non-volatile memory medium, memory and the network interface connected by system bus.Wherein, which sets
Standby non-volatile memory medium is stored with operating system and computer-readable instruction, which is held by processor
When row, processor may make to realize a kind of processing method of question and answer information.The processor of the computer equipment is for providing calculating
And control ability, support the operation of entire computer equipment.It can be stored in the memory of the computer equipment computer-readable
Instruction, when which is executed by processor, may make processor to execute a kind of processing method of question and answer information.It should
The network interface of computer equipment is used for and terminal connection communication.It will be understood by those skilled in the art that being tied shown in Fig. 7
Structure, the only block diagram of part-structure relevant to application scheme are not constituted and are applied thereon to application scheme
The restriction of computer equipment, specific computer equipment may include than more or fewer components as shown in the figure, or combination
Certain components, or with different component layouts.
In one embodiment it is proposed that a kind of computer equipment, including memory and processor, it is stored in memory
Computer-readable instruction, when computer-readable instruction is executed by processor, so that processor is executed such as above-mentioned any embodiment
The step of processing method of question and answer information.
In one embodiment it is proposed that a kind of storage medium for being stored with computer-readable instruction, this is computer-readable
When instruction is executed by one or more processors, so that one or more processors execute the question and answer information of any of the above-described embodiment
Processing method the step of.
It should be understood that although each step in the flow chart of attached drawing is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, can execute in the other order.Moreover, at least one in the flow chart of attached drawing
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, execution sequence, which is also not necessarily, successively to be carried out, but can be with other
At least part of the sub-step or stage of step or other steps executes in turn or alternately.
The above is only some embodiments of the invention, it is noted that for the ordinary skill people of the art
For member, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications are also answered
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of processing method of question and answer information, which comprises the steps of:
The setting flag input control on the page of product introduction, wherein the label input control is for recording and being shown in
The trace of the label generated on the page;
When the label input control is triggered, the label that user is inputted on the page is obtained, and record the mark
The trace of note and position;
The problem of identifying the user according to the page of the trace of the label and position and product introduction information;
By described problem information input Question-Answering Model, obtains and export target answer in the page of the product introduction, wherein institute
State the model for the problem of Question-Answering Model is for answering the page shown product.
2. the processing method of question and answer information according to claim 1, which is characterized in that it is described label include touch track or
Mouse track.
3. the processing method of question and answer information according to claim 1, which is characterized in that the trace according to the label
The step of the problem of identifying the user with the page of position and product introduction information, comprising:
The mark information is divided into several marks according to the trace of the label and position;
Identify the token types of each mark;
According to the token types of the page of the product introduction and each mark, identify that each mark is corresponding one by one respectively
Text information;
Described problem information is generated according to the text information.
4. the processing method of question and answer information according to claim 3, which is characterized in that described according to the product introduction
The token types of the page and each mark, the step of identifying the corresponding text information of each mark one by one respectively, comprising:
When the token types of the mark are text track, the mark is converted into handwriting image, identifies the person's handwriting figure
Text information as in, using the text information in the handwriting image as the corresponding text information of the mark;
When the type of the mark is underscore or circles mark, according to the mark in the page of the product introduction
The corresponding text information in the position is extracted in position from the content of text of the product introduction, by the corresponding text in the position
This information is as the corresponding text information of the mark.
5. the processing method of question and answer information according to claim 3, which is characterized in that described raw according to the text information
The step of at described problem information, comprising:
Identify the text and punctuation mark of the text information;
Obtain the mark position between the text, the punctuation mark and the mark;
According to the text, the punctuation mark and the mark position, described problem information is generated in conjunction with semanteme.
6. the processing method of question and answer information according to claim 1, which is characterized in that the page in the product introduction
Face exports the step of target answer, comprising:
Obtain preset answer output mode;
When the answer output mode is voice output, the Xiang Suoshu page returns to the corresponding voice signal of the target answer;
When the answer output mode is text output, the corresponding text envelope of the target answer is shown in the page
Breath.
7. the processing method of question and answer information according to claim 1, which is characterized in that the page of the product introduction is used for
Show insurance clause, the Question-Answering Model is for answering answer of the described problem information under the requirement of the insurance clause.
8. a kind of processing system of question and answer information characterized by comprising
Setup module, for the setting flag input control on the page of product introduction, wherein the label input control is used for
Record and show the trace of the label generated on the page;
Module is obtained, for when the label input control is triggered, obtaining the label that user is inputted on the page,
And record trace and the position of the label;
Identification module, for identifying the user's according to the page of the trace of the label and position and the product introduction
Problem information;
Output module, for by described problem information input Question-Answering Model, obtaining and exporting mesh in the page of the product introduction
Mark answer, wherein the model for the problem of Question-Answering Model is for answering the page shown product.
9. a kind of computer equipment, including memory and processor, it is stored with computer-readable instruction in the memory, it is special
Sign is, when the computer-readable instruction is executed by the processor, so that the processor executes such as claim 1 to 7
Any one of described in question and answer information processing method the step of.
10. a kind of storage medium for being stored with computer-readable instruction, which is characterized in that the computer-readable instruction is by one
Or multiple processors are when executing, so that one or more processors execute the question and answer information as described in any one of claims 1 to 7
Processing method the step of.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910440492.9A CN110308947A (en) | 2019-05-24 | 2019-05-24 | Processing method, system, computer equipment and the storage medium of question and answer information |
PCT/CN2020/086128 WO2020238498A1 (en) | 2019-05-24 | 2020-04-22 | Question and answer information processing method and system, computer device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910440492.9A CN110308947A (en) | 2019-05-24 | 2019-05-24 | Processing method, system, computer equipment and the storage medium of question and answer information |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110308947A true CN110308947A (en) | 2019-10-08 |
Family
ID=68074842
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910440492.9A Pending CN110308947A (en) | 2019-05-24 | 2019-05-24 | Processing method, system, computer equipment and the storage medium of question and answer information |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN110308947A (en) |
WO (1) | WO2020238498A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111951133A (en) * | 2020-07-31 | 2020-11-17 | 北京伴我学教育科技有限公司 | Method, device and storage medium for assisting problem solving |
WO2020238498A1 (en) * | 2019-05-24 | 2020-12-03 | 深圳壹账通智能科技有限公司 | Question and answer information processing method and system, computer device and storage medium |
CN114238787A (en) * | 2020-08-31 | 2022-03-25 | 腾讯科技(深圳)有限公司 | Answer processing method and device |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104331441A (en) * | 2014-10-24 | 2015-02-04 | 北京奇虎科技有限公司 | Method and device for providing answers to questions based on search engine |
CN105139698A (en) * | 2015-07-31 | 2015-12-09 | 广东小天才科技有限公司 | Information input method and device for finger reader |
CN108763561A (en) * | 2018-06-04 | 2018-11-06 | 深圳市零度智控科技有限公司 | E-book method for pushing, device and computer readable storage medium |
CN109285030A (en) * | 2018-08-29 | 2019-01-29 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, terminal and computer readable storage medium |
CN109670088A (en) * | 2018-12-13 | 2019-04-23 | 平安科技(深圳)有限公司 | Voice response exchange method, device, computer equipment and storage medium |
CN109697228A (en) * | 2018-12-13 | 2019-04-30 | 平安科技(深圳)有限公司 | Intelligent answer method, apparatus, computer equipment and storage medium |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103885966A (en) * | 2012-12-20 | 2014-06-25 | 北京京东尚科信息技术有限公司 | Question and answer interaction method and system of electronic commerce transaction platform |
US10210383B2 (en) * | 2015-09-03 | 2019-02-19 | Microsoft Technology Licensing, Llc | Interacting with an assistant component based on captured stroke information |
CN108345692B (en) * | 2018-03-16 | 2020-07-31 | 北京京东尚科信息技术有限公司 | Automatic question answering method and system |
CN110308947A (en) * | 2019-05-24 | 2019-10-08 | 深圳壹账通智能科技有限公司 | Processing method, system, computer equipment and the storage medium of question and answer information |
-
2019
- 2019-05-24 CN CN201910440492.9A patent/CN110308947A/en active Pending
-
2020
- 2020-04-22 WO PCT/CN2020/086128 patent/WO2020238498A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104331441A (en) * | 2014-10-24 | 2015-02-04 | 北京奇虎科技有限公司 | Method and device for providing answers to questions based on search engine |
CN105139698A (en) * | 2015-07-31 | 2015-12-09 | 广东小天才科技有限公司 | Information input method and device for finger reader |
CN108763561A (en) * | 2018-06-04 | 2018-11-06 | 深圳市零度智控科技有限公司 | E-book method for pushing, device and computer readable storage medium |
CN109285030A (en) * | 2018-08-29 | 2019-01-29 | 深圳壹账通智能科技有限公司 | Products Show method, apparatus, terminal and computer readable storage medium |
CN109670088A (en) * | 2018-12-13 | 2019-04-23 | 平安科技(深圳)有限公司 | Voice response exchange method, device, computer equipment and storage medium |
CN109697228A (en) * | 2018-12-13 | 2019-04-30 | 平安科技(深圳)有限公司 | Intelligent answer method, apparatus, computer equipment and storage medium |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020238498A1 (en) * | 2019-05-24 | 2020-12-03 | 深圳壹账通智能科技有限公司 | Question and answer information processing method and system, computer device and storage medium |
CN111951133A (en) * | 2020-07-31 | 2020-11-17 | 北京伴我学教育科技有限公司 | Method, device and storage medium for assisting problem solving |
CN111951133B (en) * | 2020-07-31 | 2024-03-22 | 北京伴我学教育科技有限公司 | Method, device and storage medium for assisting in solving questions |
CN114238787A (en) * | 2020-08-31 | 2022-03-25 | 腾讯科技(深圳)有限公司 | Answer processing method and device |
CN114238787B (en) * | 2020-08-31 | 2024-03-29 | 腾讯科技(深圳)有限公司 | Answer processing method and device |
Also Published As
Publication number | Publication date |
---|---|
WO2020238498A1 (en) | 2020-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
RU2683174C2 (en) | Ink to text representation conversion | |
US10114809B2 (en) | Method and apparatus for phonetically annotating text | |
CN109348275B (en) | Video processing method and device | |
US8887044B1 (en) | Visually distinguishing portions of content | |
US10192544B2 (en) | Method and system for constructing a language model | |
CN110308947A (en) | Processing method, system, computer equipment and the storage medium of question and answer information | |
CN111459977B (en) | Conversion of natural language queries | |
CN101561725B (en) | Method and system of fast handwriting input | |
CN109783796A (en) | Predict that the pattern in content of text destroys | |
US20220188514A1 (en) | System for analyzing and prescribing content changes to achieve target readability level | |
US20150154295A1 (en) | Searching method, system and storage medium | |
CN103294665A (en) | Text translation method for electronic reader and electronic reader | |
CN102063620A (en) | Handwriting identification method, system and terminal | |
CN102073884A (en) | Handwriting recognition method, system and handwriting recognition terminal | |
US20230214579A1 (en) | Intelligent character correction and search in documents | |
CN112084342A (en) | Test question generation method and device, computer equipment and storage medium | |
US20210350090A1 (en) | Text to visualization | |
US20160103799A1 (en) | Methods and systems for automated detection of pagination | |
US20050276480A1 (en) | Handwritten input for Asian languages | |
KR100707725B1 (en) | System and method for providing automatically completed query and computer readable recording medium recording program for implementing the method | |
US10133920B2 (en) | OCR through voice recognition | |
CN103632236A (en) | Statement voice-reading device | |
US20160253556A1 (en) | Ink stroke grouping based on stroke attributes | |
US20140095464A1 (en) | Information providing system, information providing device, information providing method and program | |
JP2006040263A (en) | Handwritten input for asian language |
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: 20191008 |
|
RJ01 | Rejection of invention patent application after publication |