CN106776832A - Processing method, apparatus and system for question and answer interactive log - Google Patents
Processing method, apparatus and system for question and answer interactive log Download PDFInfo
- Publication number
- CN106776832A CN106776832A CN201611061635.8A CN201611061635A CN106776832A CN 106776832 A CN106776832 A CN 106776832A CN 201611061635 A CN201611061635 A CN 201611061635A CN 106776832 A CN106776832 A CN 106776832A
- Authority
- CN
- China
- Prior art keywords
- answer
- question
- interactive log
- banquet
- attending
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/242—Dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- 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
- G06Q30/01—Customer relationship services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Electrically Operated Instructional Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a kind of processing method for question and answer interactive log, the question and answer interactive log includes user's question sentence and answer of attending a banquet accordingly, and the processing method includes：Words art analysis is performed to the question and answer interactive log, to determine the answer of attending a banquet with the presence or absence of taboo language and/or negative emotion；Business diagnosis is performed to the question and answer interactive log using CRM database, to judge whether user's question sentence matches with the business that user handled and whether the answer of attending a banquet is consistent with the CRM data for extracting；Question and answer are performed using question and answer knowledge base to the question and answer interactive log to analyze, it is whether correct to determine the answer of attending a banquet；And to by words art analysis, business diagnosis and in question and answer analysis, the question and answer interactive log of at least one is not marked for subsequent artefacts' quality inspection.
Description
Technical field
The present invention relates to the information processing technology, the more particularly, to processing method of question and answer interactive log, apparatus and system.
Background technology
Customer service system is more popularized with the development of ecommerce.Client can inquire phase interested by customer service system
Pass information, transacting business etc..For example, user can understand the information related to commodity, consultation service by customer service system.
Customer service system can store user's question sentence of user's query and answer of attending a banquet accordingly in use, in these
Hold and preserved so as to subsequent examination as quality detecting data, to ensure CSAT.Prior art is substantially by artificial sampling observation
Mode operated.
Quality inspection personnel is surveyed sample by the interaction data attended a banquet to phone or online text, according to quality inspection personnel to industry
The understanding of business, judges whether the problem that contact staff answers is correct, and whether answer has inappropriate tone.If being related to user
The problem of accounts information or basic service is, it is necessary to check to judge that whether it is correct that client answers in systems.For some
More complicated question and answer are, it is necessary to check example document to judge whether correctly.
Above-mentioned artificial sampling observation pattern is time-consuming, laborious, and quality inspection efficiency and sampling observation coverage rate are low, and are difficult to find that short slab adds rapidly
To improve.
The content of the invention
The brief overview of one or more aspects given below is providing to the basic comprehension in terms of these.This general introduction is not
The extensive overview of all aspects for contemplating, and it is also non-to be both not intended to identify the key or decisive key element of all aspects
Attempt to define the scope in terms of any or all.Its unique purpose is to provide the one of one or more aspects in simplified form
A little concepts think the sequence of more detailed description given later.
According to an aspect of the present invention, there is provided a kind of processing method for question and answer interactive log, question and answer interaction day
Will includes user's question sentence and answer of attending a banquet accordingly, and the processing method includes：
Words art analysis is performed to the question and answer interactive log, to determine the answer of attending a banquet with the presence or absence of taboo language and/or negative feelings
Sense；
Using CRM database to the question and answer interactive log perform business diagnosis, with judge user's question sentence whether with user
The business handled is matched and whether the answer of attending a banquet is consistent with the CRM data for extracting；
Question and answer are performed using question and answer knowledge base to the question and answer interactive log to analyze, it is whether correct to determine the answer of attending a banquet；
And
The question and answer interactive log of at least one is marked in not analyzed by words art analysis, business diagnosis and question and answer
For subsequent artefacts' quality inspection.
According to another aspect of the present invention, there is provided a kind of processing unit for question and answer interactive log, question and answer interaction
Daily record includes user's question sentence and answer of attending a banquet accordingly, and the processing unit includes：
Words art analytic unit, for performing words art analysis to the question and answer interactive log, to determine whether the answer of attending a banquet deposits
Prohibiting language and/or negative emotion；
Business diagnosis unit, for performing business diagnosis to the question and answer interactive log using CRM database, to judge the use
Whether family question sentence matches with the business that user handled and whether the answer of attending a banquet is consistent with the CRM data for extracting；
Question and answer analytic unit, for performing question and answer analysis to the question and answer interactive log using question and answer knowledge base, to determine to be somebody's turn to do
Whether answer of attending a banquet is correct；And
Indexing unit, for handing over not passing through the question and answer of at least one during words art analysis, business diagnosis and question and answer are analyzed
Mutual daily record is marked for subsequent artefacts' quality inspection.
In accordance with a further aspect of the present invention, there is provided a kind of processing system for question and answer interactive log, question and answer interaction
Daily record includes user's question sentence and answer of attending a banquet accordingly, and the processing system includes：
CRM database, for storing the CRM data being associated with client；
Question and answer knowledge base, asks and asks the model answer being associated with each standard for storing standard；And
Foregoing processing unit.
Undesirable question and answer can be handed over automatically by the processing method for question and answer interactive log of the invention
Mutual daily record is marked, and plays a part of preliminary examination, for artificial further quality inspection.In this way, one is to substantially increase quality inspection
Efficiency, greatly alleviates the workload of the artificial quality inspection from magnanimity interactive log of quality inspection personnel.On the other hand, due to quality inspection personnel
Quality inspection can only be carried out by the way of sampling observation, cause quality inspection result can not completely react truth.Scheme of the invention is right
All of question and answer interactive log all processed, it is to avoid the situation of missing inspection.
Brief description of the drawings
After the detailed description for reading embodiment of the disclosure in conjunction with the following drawings, better understood when of the invention
Features described above and advantage.In the accompanying drawings, each component is not necessarily drawn to scale, and with similar correlation properties or feature
Component may have same or like reference.
Fig. 1 shows the flow chart of the processing method for question and answer interactive log according to an aspect of the present invention；
Fig. 2 shows the block diagram of the processing unit for question and answer interactive log according to an aspect of the present invention；
Fig. 3 shows the block diagram of the business diagnosis unit according to an embodiment；
Fig. 4 shows the block diagram of the question and answer analytic unit according to an embodiment；And
Fig. 5 shows the block diagram of the processing system for question and answer interactive log according to an aspect of the present invention.
Specific embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.Note, it is below in conjunction with accompanying drawing and specifically real
It is only exemplary to apply the aspects of example description, and is understood not to carry out any limitation to protection scope of the present invention.
Customer service system can produce substantial amounts of question and answer interactive log in use, describe user's question sentence and corresponding
Attend a banquet answer.According to an aspect of the present invention, there is provided a kind of processing method for question and answer interactive log, can be automatically right
Undesirable question and answer interactive log is marked, and plays a part of preliminary examination, for artificial further quality inspection.In this way,
One is to substantially increase quality inspection efficiency, greatly alleviates the workload of the artificial quality inspection from magnanimity interactive log of quality inspection personnel.Separately
On the one hand, because quality inspection personnel can only carry out quality inspection by the way of sampling observation, quality inspection result is caused can not completely to react truth.
Scheme of the invention, all processed all of question and answer interactive log, it is to avoid the situation of missing inspection.
Fig. 1 shows the flow chart of the processing method 100 for question and answer interactive log according to an aspect of the present invention.
As shown in figure 1, processing method 100 may include following steps：
Step 101：Pretreatment is performed to question and answer interactive log.
Interactive log can be phonetic matrix.For example, customer service system is probably phone customer service system, resulting question and answer
Interactive log is the form of voice.
In the case of phonetic matrix, the question and answer interactive log of speech form can be converted to text formatting by pretreatment first
Question and answer interactive log.As example, can be by ASR (Automatic Speech Recognition, automatic speech recognition)
Perform voice to the conversion of word.
More preferably, pretreatment may also include and insignificant question and answer data in question and answer interactive log are filtered.
Step 102：Art analysis is talked about to the execution of question and answer interactive log, to determine to attend a banquet answer with the presence or absence of taboo language and/or bear
Face emotion.
The judgement for prohibiting language can be realized by prohibiting words and phrases storehouse.In one example, there is provided have taboo words and phrases storehouse, wherein having included quilt
Labeled as the word for prohibiting language, including for example uncivil term, do not meet any incorrect etc. the word of state's laws regulation and answer
Disabled word.Specifically, can be marked based on whether there is in the answer of attending a banquet for prohibiting words and phrases library searching question and answer interactive log
It is designated as prohibiting the word of language, if in the presence of not by talking about art analysis.
In addition to prohibiting language and judging, words art analysis can also judge to attend a banquet that answer, with the presence or absence of negative emotion, is judged with this
The attitude of customer service.Sentiment analysis are general to be weighed in terms of two, Sentiment orientation direction and Sentiment orientation degree.
Sentiment orientation direction is also referred to as feeling polarities, it can be understood as user expresses certain object the state that itself viewpoint is held
Degree be support, oppose, neutral, i.e., generally signified positive emotion, negative emotion, neutral emotion.Sentiment orientation degree refers to main body
Degree of strength when positive emotion or negative emotion is expressed object, and different emotion degree are often by different emotion words
Or the emotion tone etc. embodies.
In one example, the judgement of negative emotion can be realized by sentiment dictionary.For example, sentiment dictionary can be provided with,
Wherein include and be marked as the word with negative emotion.Specifically, question and answer interactive log can be retrieved based on sentiment dictionary
Answer of attending a banquet in the presence or absence of being marked as the text with negative emotion, if in the presence of not analyzed by talking about art.
In another example, the judgement of negative emotion can be realized by emotion classifiers.In this embodiment, first may be used
Method based on machine learning, emotion classifiers are trained by the use of Large Scale Corpus as training set.The training side of grader
Method is known technology, be will not be repeated here.
Above-mentioned emotional semantic classification and taboo language judges to be performed as unit with word, sentence or paragraph.Paragraph
Chapter level sentiment analysis carry out tendentiousness judgement primarily directed to certain theme or event, generally require the emotion for building correspondence event
Dictionary.The sentiment analysis of Sentence-level are obtained by calculating the average value of all emotion words included in sentence.
The analysis of words art can individually include that prohibiting language judges or Judgment by emotion, it is also possible to while including both.For example, can be first
Attended a banquet described in words and phrases library searching question and answer interactive log with the presence or absence of the word for being marked as prohibiting language in answer based on prohibiting, if in the presence of
Do not analyzed by talking about art then, otherwise whether there is in answer of attending a banquet of the continuation based on sentiment dictionary retrieval question and answer interactive log and marked
The text with negative emotion is designated as, if not analyzed by talking about art in the presence of if, is otherwise analyzed by the words art.It is this kind of first to prohibit language
Judge again the mode of Judgment by emotion, both ensure that the high efficiency of analysis, also ensure that the high-accuracy of analysis.
Step 103：Using CRM database to question and answer interactive log perform business diagnosis, with judge user's question sentence whether with
The business that user handled is matched and whether answer of attending a banquet is consistent with the CRM data for extracting.
Be stored with CRM numbers in CRM (Customer relationship management, customer relation management) database
According to CRM data have recorded the business datum related to client.User is in inquiry and personal itself related individual business problem
When, for example oneself have subscribed what set meal, inquiry into balance etc., customer service needs inquiry CRM database to answer.Business diagnosis master
The individual business problem related to user itself that be directed to user's proposition is analyzed.
In one example, participle can be carried out to the user's question sentence in question and answer interactive log first.Can use in the present invention
Any suitable segmentation methods carry out participle to user's question sentence.Conventional segmentation methods may include character match method, understanding method, system
Meter method etc..
After participle, the CRM data related to word segmentation result can be extracted from CRM database.In one example, can be by inciting somebody to action
Word segmentation result is matched with CRM data in CRM database, to obtain the CRM data related to word segmentation result.Then, from CRM
The CRM data related to word segmentation result is extracted in database, and judges whether the CRM data for extracting is consistent with answer of attending a banquet,
It is such as consistent, then by business diagnosis.
For example user's question sentence can be：Inquiry into balance；
Corresponding answer of attending a banquet can be：48 yuan.
According to examples detailed above, the user balance this CRM field is transferred from CRM database based on inquiry into balance corresponding
Data：45 yuan.
Now, because 48 yuan of answer of attending a banquet inconsistent with 45 yuan of CRM data, then business diagnosis does not pass through.
When CRM data is extracted, it is also possible to extract failure, such as user's question sentence is not one related to user identity
Traffic issues, at this point it is possible to Null is returned, to represent that business diagnosis is invalid.
Step 104：Question and answer are performed using question and answer knowledge base to question and answer interactive log to analyze, to determine whether just to attend a banquet answer
Really.
Substantial amounts of professional knowledge point is stored in question and answer knowledge base in the form of problem-answer knowledge point." standard is asked " is
For representing the word of certain knowledge point, main target is clear expression, is easy to safeguard.For example, " rate of CRBT " are exactly table
Description is asked up to clearly standard.Here " asking " should not be narrowly interpreted as " inquiry ", and should broadly understand that one is " defeated
Enter ", being somebody's turn to do " input " has corresponding " output ".For example, for the semantics recognition for control system, a finger of user
Order, for example " turn on radio " should also be understood to be one " asking ", and now corresponding " answering " can be performed for phase
The control program that should control is called.
In one example, it may be determined that whether there is in question and answer knowledge base and match with the user's question sentence in question and answer interactive log
Standard ask.This judgement can be performed by Semantic Similarity Measurement.For example, user's question sentence can be asked with the standard in knowledge base
Semantic Similarity Measurement is performed, the standard with highest semantic similarity is asked and is asked as matching standard.
Arithmetic of Semantic Similarity refers to that the similarity degree between two different words and expressions is calculated by certain method.
The semantic similarity degree between sentence would generally be weighed with a percentage.
Common similarity of character string algorithm include editing distance algorithm (EditDistance), n-gram algorithms,
JaroWinkler algorithms and Soundex algorithms etc..
Two similarity problems of sentence can be attributed to and change into one of sentence character string by editing distance algorithm
Another cost to be paid of sentence character string.The cost of conversion is higher, illustrates that the similarity of two character strings is lower.Generally
The transform mode that can be selected includes insertion, replaces and deletes.
N-Gram algorithms are based on such a hypothesis：N-th appearance of word and above n-1 i.e. in character string
Individual word is related, and all uncorrelated to other any words, and the probability that whole character string occurs is exactly multiplying for the probability that each word occurs
Product.Length is the substring of n during N-gram also represents target string in itself, citing, and " ARM " is a 3- in " ARMY "
gram.When in two character strings, when identical n-gram is more, two word strings will be considered as even more like.
Jaro Winkler are then that n-gram algorithms more go a step further.Unmatched part in n-gram is entered simultaneously
Gone transposition consideration so that more accurately similarity degree can be obtained.JaroWinkler is comparing two feelings compared with short character strings
Under condition, good result can be obtained.
For example, user's question sentence can be：I wants to ask lower current 88 yuan of 4G set meals include how many flow.
The standard of matching is asked can be：4G set meal flows.
If the standard in the presence of matching is asked, it is determined that the corresponding model answer that the standard for being matched is asked and question and answer interactive log
In corresponding answer of attending a banquet it is whether consistent, analyzed by question and answer if matching, do not analyzed by question and answer otherwise.
For example, asking that the function that " 4G set meals flow " corresponding model answer is call parameters " 88 yuan " is returned with standard
Return value, such as 2 flows of G.Now can be right compared with answer of attending a banquet by it, you can know answer correctness of attending a banquet.
Question and answer analysis be directed to it is mostly be general traffic issues, for this class of inquiry into balance and the close phase of user identity
When the traffic issues of pass carry out question and answer analysis, the general knowledge point for not having this personalization in question and answer knowledge base, therefore question and answer
The result of analysis is null, represents that question and answer analysis is invalid.
In certain embodiments, question and answer analysis may also be combined with long text parser, can be by by long text analysis method
Determining user's question sentence includes M problem, and is divided into the individual sub- answers of M by the answer that will attend a banquet of long text analysis method, in this base
The enterprising hand-manipulating of needle of plinth carries out question and answer analysis respectively to each problem and its corresponding sub- answer.Only when all problems and its corresponding
When sub- answer is all analyzed by question and answer, just can determine that answer of attending a banquet is correct.
Step 105：To not analyzed by talking about art, during business diagnosis and question and answer are analyzed, the question and answer interactive log of at least one enters
Line flag is for subsequent artefacts' quality inspection.
Rower is entered by the question and answer interactive log of at least one to not analyzed by talking about art, during business diagnosis and question and answer are analyzed
Note, can cause manually only quality inspection these question and answer interactive logs for being marked with, so as to save manpower.
In one example, customer service information can be extracted from labeled question and answer interactive log, the customer service information can be wrapped
Include：Temporal information, personal information, labeled type are (i.e.：Do not analyzed by talking about art, do not pass through business diagnosis and/or by asking
Answer analysis), labeled number of times etc., so that subsequently corresponding contact staff is targetedly trained or is punished.
Art analysis, business diagnosis and question and answer analysis can select therein any one according to the selection of user if above-mentioned
Plant, performed for two or three.Or, can be by pre-setting the relation that user's question sentence is analyzed with quality inspection, so as to subsequently receive
Real-time quality inspection scheme is generated during to specific user's question sentence.For example, being related to user identity individual when receive user's question sentence
During property traffic issues, business diagnosis can be performed, and be the general traffic issues unrelated with user identity in user's question sentence
When, question and answer analysis can be performed.
Although for make explanation simplify the above method is illustrated and is described as a series of actions, it should be understood that and understand,
The order that these methods are not acted is limited, because according to one or more embodiments, some actions can occur in different order
And/or with from it is depicted and described herein or herein it is not shown and describe but it will be appreciated by those skilled in the art that other
Action concomitantly occurs.
Fig. 2 shows the block diagram of the processing unit 200 for question and answer interactive log according to an aspect of the present invention.Place
Reason device 200 may include pretreatment unit 201, words art analytic unit 202, business diagnosis unit 203, question and answer analytic unit 204,
Indexing unit 205.
The question and answer interactive log that the question and answer interactive log of phonetic matrix can be converted to text formatting by pretreatment unit 201 is simultaneously
Filter insignificant question and answer data.
Words art analytic unit 202 can be used to perform question and answer interactive log words art analysis, and to determine to attend a banquet, whether answer deposits
Prohibiting language and/or negative emotion.In one example, words art analytic unit 202 can be based on taboo words and phrases library searching question and answer interactive log
Answer of attending a banquet in the presence or absence of be marked as prohibit language word, if in the presence of, not by talk about art analyze.In another example,
Words art analytic unit 202 can be based in the answer of attending a banquet of sentiment dictionary retrieval question and answer interactive log with the presence or absence of being marked as having
The text of negative emotion, if in the presence of not by talking about art analysis.In another example, words art analytic unit 202 can be based on language
Expect that the answer of attending a banquet as the built-up emotion classifiers of training set to question and answer interactive log carries out emotional semantic classification, when classification is tied
When fruit is for negative emotion, then do not analyzed by talking about art.
Above-mentioned words art analytic unit 202 can perform words art analysis as unit with word, sentence or paragraph.
Business diagnosis unit 203 performs business diagnosis using CRM database to question and answer interactive log, to judge user's question sentence
It is whether whether consistent with the CRM data for extracting with the business matching and answer of attending a banquet that user handled.
In one example, as shown in figure 3, business diagnosis unit 203 may include that participle unit 2031, CRM data obtains list
Unit 2032 and comparing unit 2033.Participle unit 2031 can carry out participle, CRM to the user's question sentence in question and answer interactive log
Data capture unit 2032 can extract the CRM data related to word segmentation result from CRM database, and comparing unit 2033 can be sentenced
Whether the disconnected CRM data for extracting is consistent with answer of attending a banquet, such as consistent, then by business diagnosis.
Question and answer analytic unit 204 can perform question and answer analysis to question and answer interactive log using question and answer knowledge base, to determine to attend a banquet
Whether answer is correct.
Specifically, question and answer analytic unit 204 can determine that whether there is in question and answer knowledge base with question and answer interactive log
The standard that user's question sentence matches asks, if in the presence of, it is determined that the corresponding model answer that the standard for being matched is asked and the question and answer
Whether the answer of attending a banquet in interactive log is consistent, is analyzed by question and answer if matching.
In one example, as shown in figure 4, question and answer analytic unit 204 may include Semantic Similarity Measurement unit 2041 with logical
Semantic Similarity Measurement is crossed to determine the matching of user's question sentence.In another example, question and answer analytic unit 204 may include long text
Analytic unit 2042 is divided into M problem with by user's question sentence, and the answer that will attend a banquet is divided into M sub- answer.
Indexing unit 205 question and answer of at least one can be interacted to not analyzed by talking about art, during business diagnosis and question and answer are analyzed
Daily record is marked for subsequent artefacts' quality inspection.
In one embodiment, the processing unit 200 may also include extraction unit and be carried with from labeled question and answer interactive log
Take customer service information.
According to another aspect of the present invention, a kind of processing system for question and answer interactive log is additionally provided, such as Fig. 5 institutes
Show.Processing system 500 may include CRM database 510, question and answer knowledge base 520 and processing unit 530.Here processing unit
530 can have similar structure with previously described processing unit 200.
Processing unit 530 in Fig. 5 illustrate only business diagnosis unit 531 and question and answer analytic unit 532.Business diagnosis list
Unit 531 can perform business diagnosis using CRM database to question and answer interactive log, to judge whether user's question sentence was handled with user
Business matching and attend a banquet answer whether with extract CRM data it is consistent.Question and answer analytic unit 532 can utilize question and answer knowledge base pair
Whether question and answer interactive log performs question and answer analysis, correct to determine answer of attending a banquet.
It is for so that any person skilled in the art can all make or use this public affairs to provide of this disclosure being previously described
Open.Various modifications of this disclosure all will be apparent for a person skilled in the art, and as defined herein general
Suitable principle can be applied to spirit or scope of other variants without departing from the disclosure.Thus, the disclosure is not intended to be limited
Due to example described herein and design, but should be awarded and principle disclosed herein and novel features phase one
The widest scope of cause.
Claims (25)
1. a kind of processing method for question and answer interactive log, the question and answer interactive log includes user's question sentence and corresponding sits
Seat answer, the treating method comprises：
Words art analysis is performed to the question and answer interactive log, language and/or negative feelings are prohibited to determine that the answer of attending a banquet whether there is
Sense；
Using CRM database to the question and answer interactive log perform business diagnosis, with judge user's question sentence whether with user
The business handled is matched and whether the answer of attending a banquet is consistent with the CRM data for extracting；
Question and answer are performed using question and answer knowledge base to the question and answer interactive log to analyze, it is whether correct to determine the answer of attending a banquet；
And
To not analyzed by the words art, business diagnosis and question and answer analyze in the question and answer interactive log of at least one be marked with
For subsequent artefacts' quality inspection.
2. processing method as claimed in claim 1, it is characterised in that the execution words art analysis includes：
Based on prohibiting in answer of being attended a banquet described in question and answer interactive log described in words and phrases library searching with the presence or absence of the word for being marked as prohibiting language
Language, if in the presence of not by the words art analysis.
3. processing method as claimed in claim 1, it is characterised in that the execution words art analysis includes：
It is marked as with negative based on whether there is in answer of being attended a banquet described in the sentiment dictionary retrieval question and answer interactive log
The text of emotion, if in the presence of not by the words art analysis.
4. processing method as claimed in claim 1, it is characterised in that the execution words art analysis includes：
Based on using language material as the built-up emotion classifiers of training set to answer of being attended a banquet described in the question and answer interactive log
Emotional semantic classification is carried out, when classification results are negative emotion, is not then analyzed by the words art.
5. the processing method as described in claim 3 or 4, it is characterised in that the words art analysis is with word, sentence or section
Fall for unit to performing.
6. processing method as claimed in claim 1, it is characterised in that the execution words art analysis includes：
Based on prohibiting in answer of being attended a banquet described in question and answer interactive log described in words and phrases library searching with the presence or absence of the word for being marked as prohibiting language
Language, if not analyzed by the words art in the presence of if, otherwise continues to retrieve the described of the question and answer interactive log based on sentiment dictionary
Attend a banquet in answer with the presence or absence of being marked as the text with negative emotion, if in the presence of if not by the words art analysis, otherwise
Analyzed by the words art.
7. processing method as claimed in claim 1, it is characterised in that the execution business diagnosis includes：
User's question sentence in the question and answer interactive log carries out participle；
The CRM data related to word segmentation result is extracted from CRM database, and
Judge whether the CRM data for extracting is consistent with the answer of attending a banquet, it is such as consistent, then by the business diagnosis.
8. processing method as claimed in claim 1, it is characterised in that the execution question and answer analysis includes：
Determine to whether there is the mark matched with the user's question sentence in the question and answer interactive log in the question and answer knowledge base
Standard is asked；
If in the presence of attending a banquet described in, it is determined that the corresponding model answer that the standard for being matched is asked and the question and answer interactive log
Whether answer is consistent, is analyzed by the question and answer if matching.
9. processing method as claimed in claim 8, it is characterised in that the determination of user's question matching is by semantic similar
Degree calculates to perform.
10. processing method as claimed in claim 9, it is characterised in that also include：Determined by long text analysis method described
User's question sentence includes M problem, and the answer of attending a banquet is divided into M sub- answer by long text analysis method.
11. processing methods as claimed in claim 1, it is characterised in that also include：
The question and answer interactive log of phonetic matrix is converted into the question and answer interactive log of text formatting and insignificant question and answer number is filtered
According to.
12. processing methods as claimed in claim 1, it is characterised in that also include：
Customer service information is extracted from labeled question and answer interactive log.
A kind of 13. processing units for question and answer interactive log, the question and answer interactive log is including user's question sentence and accordingly
Attend a banquet answer, the processing unit includes：
Words art analytic unit, for performing words art analysis to the question and answer interactive log, to determine whether the answer of attending a banquet deposits
Prohibiting language and/or negative emotion；
Business diagnosis unit, for performing business diagnosis to the question and answer interactive log using CRM database, to judge the use
Whether family question sentence matches with the business that user handled and whether the answer of attending a banquet is consistent with the CRM data for extracting；
Question and answer analytic unit, it is described to determine for performing question and answer analysis to the question and answer interactive log using question and answer knowledge base
Whether answer of attending a banquet is correct；And
Indexing unit, for the question and answer of at least one are not interacted by words art analysis, business diagnosis and in question and answer analysis
Daily record is marked for subsequent artefacts' quality inspection.
14. processing units as claimed in claim 13, it is characterised in that the words art analytic unit is based on prohibiting words and phrases library searching
With the presence or absence of the word for being marked as prohibiting language in the answer of attending a banquet of the question and answer interactive log, if in the presence of not by institute
State words art analysis.
15. processing units as claimed in claim 13, it is characterised in that the words art analytic unit is retrieved based on sentiment dictionary
Whether there is in the answer of attending a banquet of the question and answer interactive log and be marked as the text with negative emotion, if in the presence of,
Do not analyzed by the words art.
16. processing units as claimed in claim 13, it is characterised in that the words art analytic unit is based on using language material as instruction
Practice the built-up emotion classifiers of collection carries out emotional semantic classification to answer of being attended a banquet described in the question and answer interactive log, when classification is tied
When fruit is for negative emotion, then do not analyzed by the words art.
17. processing unit as described in claim 15 or 16, it is characterised in that the words art analytic unit is with word, sentence
Art is analyzed if son or paragraph are performed for unit.
18. processing units as claimed in claim 13, it is characterised in that the words art analytic unit is based on prohibiting words and phrases library searching
With the presence or absence of the word for being marked as prohibiting language in the answer of attending a banquet of the question and answer interactive log, described in not passing through in the presence of if
Words art is analyzed, and otherwise be whether there is in the answer of attending a banquet of the continuation based on the sentiment dictionary retrieval question and answer interactive log and is marked
The text with negative emotion is designated as, if not analyzed by the words art in the presence of if, is otherwise analyzed by the words art.
19. processing units as claimed in claim 13, it is characterised in that the business diagnosis unit includes：
Participle unit, to the question and answer interactive log in user's question sentence carry out participle；
CRM data acquiring unit, extracts the CRM data related to word segmentation result from CRM database, and
Comparing unit, judges whether the CRM data for extracting is consistent with the answer of attending a banquet, such as consistent, then by the business point
Analysis.
20. processing units as claimed in claim 13, it is characterised in that the question and answer analytic unit determines the question and answer knowledge
Whether there is the standard matched with the user's question sentence in the question and answer interactive log in storehouse to ask；If in the presence of, it is determined that institute
The corresponding model answer that the standard of matching is asked with the question and answer interactive log described in answer of attending a banquet it is whether consistent, if matching
Then analyzed by the question and answer.
21. processing units as claimed in claim 20, it is characterised in that the question and answer analytic unit includes semantic similarity meter
Unit is calculated to determine the matching of user's question sentence by Semantic Similarity Measurement.
22. processing units as claimed in claim 21, it is characterised in that the question and answer analytic unit is also analyzed including long text
Unit is divided into M problem with by user's question sentence, and the answer of attending a banquet is divided into M sub- answer.
23. processing units as claimed in claim 13, it is characterised in that also include：
Pretreatment unit, for the question and answer interactive log of phonetic matrix to be converted to the question and answer interactive log of text formatting and is filtered
Insignificant question and answer data.
24. processing units as claimed in claim 13, it is characterised in that also include：
Extraction unit, for extracting customer service information from labeled question and answer interactive log.
A kind of 25. processing systems for question and answer interactive log, the question and answer interactive log is including user's question sentence and accordingly
Attend a banquet answer, the processing system includes：
CRM database, for storing the CRM data being associated with client；
Question and answer knowledge base, asks and asks the model answer being associated with each standard for storing standard；And
Processing unit as any one of claim 13-24.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611061635.8A CN106776832B (en) | 2016-11-25 | 2016-11-25 | Processing method, apparatus and system for question and answer interactive log |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611061635.8A CN106776832B (en) | 2016-11-25 | 2016-11-25 | Processing method, apparatus and system for question and answer interactive log |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106776832A true CN106776832A (en) | 2017-05-31 |
| CN106776832B CN106776832B (en) | 2019-07-19 |
Family
ID=58911129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611061635.8A Active CN106776832B (en) | 2016-11-25 | 2016-11-25 | Processing method, apparatus and system for question and answer interactive log |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106776832B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766560A (en) * | 2017-11-03 | 2018-03-06 | 广州杰赛科技股份有限公司 | The evaluation method and system of customer service flow |
| CN108153732A (en) * | 2017-12-25 | 2018-06-12 | 科大讯飞股份有限公司 | The checking method and device of a kind of hearing record |
| CN108712581A (en) * | 2018-06-07 | 2018-10-26 | 上海航动科技有限公司 | A kind of traffic method of skill training, device, equipment and system |
| WO2019179010A1 (en) * | 2018-03-22 | 2019-09-26 | 平安科技（深圳）有限公司 | Data set acquisition method, classification method and device, apparatus, and storage medium |
| CN111488439A (en) * | 2020-03-18 | 2020-08-04 | 杭州迪普科技股份有限公司 | System and method for saving and analyzing log data |
| CN112086092A (en) * | 2019-06-14 | 2020-12-15 | 广东技术师范大学 | Intelligent extraction method of dialect based on emotion analysis |
| WO2021159670A1 (en) * | 2020-02-11 | 2021-08-19 | 深圳壹账通智能科技有限公司 | Method and apparatus for processing unknown question in intelligent questions and answers, computer device, and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101076184A (en) * | 2006-07-31 | 2007-11-21 | 腾讯科技（深圳）有限公司 | Method and system for realizing automatic reply |
| US20080195378A1 (en) * | 2005-02-08 | 2008-08-14 | Nec Corporation | Question and Answer Data Editing Device, Question and Answer Data Editing Method and Question Answer Data Editing Program |
| CN102291617A (en) * | 2011-09-03 | 2011-12-21 | 四川公用信息产业有限责任公司 | End-to-end fault diagnosing and positioning platform of IPTV (Internet Protocol Television) business |
| CN104679910A (en) * | 2015-03-25 | 2015-06-03 | 北京智齿博创科技有限公司 | Intelligent answering method and system |
-
2016
- 2016-11-25 CN CN201611061635.8A patent/CN106776832B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080195378A1 (en) * | 2005-02-08 | 2008-08-14 | Nec Corporation | Question and Answer Data Editing Device, Question and Answer Data Editing Method and Question Answer Data Editing Program |
| CN101076184A (en) * | 2006-07-31 | 2007-11-21 | 腾讯科技（深圳）有限公司 | Method and system for realizing automatic reply |
| CN102291617A (en) * | 2011-09-03 | 2011-12-21 | 四川公用信息产业有限责任公司 | End-to-end fault diagnosing and positioning platform of IPTV (Internet Protocol Television) business |
| CN104679910A (en) * | 2015-03-25 | 2015-06-03 | 北京智齿博创科技有限公司 | Intelligent answering method and system |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766560A (en) * | 2017-11-03 | 2018-03-06 | 广州杰赛科技股份有限公司 | The evaluation method and system of customer service flow |
| CN108153732A (en) * | 2017-12-25 | 2018-06-12 | 科大讯飞股份有限公司 | The checking method and device of a kind of hearing record |
| CN108153732B (en) * | 2017-12-25 | 2021-08-03 | 浙江讯飞智能科技有限公司 | Examination method and device for interrogation notes |
| WO2019179010A1 (en) * | 2018-03-22 | 2019-09-26 | 平安科技（深圳）有限公司 | Data set acquisition method, classification method and device, apparatus, and storage medium |
| CN108712581A (en) * | 2018-06-07 | 2018-10-26 | 上海航动科技有限公司 | A kind of traffic method of skill training, device, equipment and system |
| CN112086092A (en) * | 2019-06-14 | 2020-12-15 | 广东技术师范大学 | Intelligent extraction method of dialect based on emotion analysis |
| WO2021159670A1 (en) * | 2020-02-11 | 2021-08-19 | 深圳壹账通智能科技有限公司 | Method and apparatus for processing unknown question in intelligent questions and answers, computer device, and medium |
| CN111488439A (en) * | 2020-03-18 | 2020-08-04 | 杭州迪普科技股份有限公司 | System and method for saving and analyzing log data |
| CN111488439B (en) * | 2020-03-18 | 2023-04-18 | 杭州迪普科技股份有限公司 | System and method for saving and analyzing log data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106776832B (en) | 2019-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106776832B (en) | Processing method, apparatus and system for question and answer interactive log | |
| CN106295807B (en) | A kind of method and device of information processing | |
| US7725318B2 (en) | System and method for improving the accuracy of audio searching | |
| WO2020228173A1 (en) | Illegal speech detection method, apparatus and device and computer-readable storage medium | |
| CN109543020B (en) | Query processing method and system | |
| CN105930452A (en) | Smart answering method capable of identifying natural language | |
| CN113505586A (en) | Seat-assisted question-answering method and system integrating semantic classification and knowledge graph | |
| Macary et al. | AlloSat: A new call center french corpus for satisfaction and frustration analysis | |
| Prasad et al. | Intelligent chatbot for lab security and automation | |
| CN112765974B (en) | Service assistance method, electronic equipment and readable storage medium | |
| Kopparapu | Non-linguistic analysis of call center conversations | |
| CN109872714A (en) | A kind of method, electronic equipment and storage medium improving accuracy of speech recognition | |
| CN110136726A (en) | A kind of estimation method, device, system and the storage medium of voice gender | |
| CN108877812A (en) | A kind of method for recognizing sound-groove, device and storage medium | |
| CN106205635A (en) | Method of speech processing and system | |
| CN110347812A (en) | A kind of search ordering method and system towards judicial style | |
| Koudounas et al. | Italic: An italian intent classification dataset | |
| CN111916110A (en) | Voice quality inspection method and device | |
| CN115063155B (en) | Data labeling method, device, computer equipment and storage medium | |
| CN108830318A (en) | Quality detecting method, device, equipment and storage medium | |
| Durrani et al. | Transfer learning based speech affect recognition in Urdu | |
| CN109284364B (en) | Interactive vocabulary updating method and device for voice microphone-connecting interaction | |
| CN115438158A (en) | Intelligent dialogue method, device, equipment and storage medium | |
| CN112527973A (en) | Method and system for searching disease information | |
| CN113345419B (en) | Dialect accent based speech translation method, system and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Processing method, device and system for Q&A interactive log Effective date of registration: 20230223 Granted publication date: 20190719 Pledgee: China Construction Bank Corporation Shanghai No.5 Sub-branch Pledgor: SHANGHAI XIAOI ROBOT TECHNOLOGY Co.,Ltd. Registration number: Y2023980033272 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |