CN118132686A - Service robot problem matching method and system for multi-service scene - Google Patents
Service robot problem matching method and system for multi-service scene Download PDFInfo
- Publication number
- CN118132686A CN118132686A CN202410557671.1A CN202410557671A CN118132686A CN 118132686 A CN118132686 A CN 118132686A CN 202410557671 A CN202410557671 A CN 202410557671A CN 118132686 A CN118132686 A CN 118132686A
- Authority
- CN
- China
- Prior art keywords
- text
- question
- manual conversion
- service
- qualified
- 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
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000006243 chemical reaction Methods 0.000 claims description 83
- 238000004458 analytical method Methods 0.000 claims description 30
- 238000007405 data analysis Methods 0.000 claims description 20
- 238000012790 confirmation Methods 0.000 claims description 18
- 238000013139 quantization Methods 0.000 claims description 5
- 230000001172 regenerating effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
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
- 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
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- Human Computer Interaction (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the field of service robots, in particular to a service robot problem matching method and a service robot problem matching system for a multi-service scene, which are used for solving the problems that the existing service robot problem matching method for the multi-service scene cannot judge the similarity between a questioned problem and a preset problem in multiple aspects, the similarity judgment is not accurate enough, and a problem library cannot be continuously and intelligently expanded; the method realizes high-accuracy similarity judgment and screening through multiple factors, realizes high-performance and high-accuracy problem matching, improves the accuracy and efficiency of answer return, can adapt to the requirements of different service scenes, provides powerful support for the wide application of service robots, realizes intelligent matching and answering of the problems of multiple service scenes, and continuously improves the problem identification and matching capability along with the continuous expansion of a problem search library, thereby having flexibility and expandability and improving the problem processing capability of the service robots.
Description
Technical Field
The invention relates to the field of service robots, in particular to a service robot problem matching method and system for a multi-service scene.
Background
With the development of artificial intelligence technology, service robots are widely used in various fields such as catering, retail, medical and educational fields. However, due to the diversity of user demands in different business scenarios, service robots remain challenging in terms of understanding user problems and providing accurate services. The patent with the application number of CN202011319407.2 discloses a service robot problem matching method facing to a multi-service scene, which comprises the following steps: s110: the service robot obtains the user inquiry and inputs the user inquiry and the service scene number into the problem matching system; s120: the problem matching system finds a corresponding problem library and an index according to the service scene number; s130: the question library corresponding to the service scene is initially screened by means of a search engine, and N most similar questions are obtained; s140: reordering the N problems by using a reordering model shared by all service scenes, and outputting the QID of the problem ranked first to the service robot; s150: and the service robot finds out a proper answer from the answer library according to the QID and returns the answer to the user. The primary screening and reordering 'two-step' strategy provided by the invention has the advantages of high performance and high accuracy, but still has the following defects: the similarity between the questioned questions and the preset questions cannot be judged through multiple aspects, the similarity judgment is not accurate enough, and the question library cannot be continuously and intelligently expanded, so that when the multi-service scene questions are processed, the service robot often faces challenges of inaccurate question matching and untimely answer return. Therefore, how to improve the capability of the service robot to handle the multi-service scenario problem is a current urgent problem to be solved.
Disclosure of Invention
In order to overcome the technical problems, the invention aims to provide a service robot problem matching method and a service robot problem matching system for a multi-service scene, which solve the problems that the existing service robot problem matching method for the multi-service scene cannot judge the similarity of the asked problem and a preset problem in multiple aspects, the similarity judgment is not accurate enough, and a problem library cannot be continuously and intelligently expanded, so that when the problem of the multi-service scene is processed, the service robot always faces challenges of inaccurate problem matching and untimely answer return.
The aim of the invention can be achieved by the following technical scheme:
A multi-business scenario oriented service robot problem matching system, comprising:
The user consultation module is used for inputting the consultation questions in the dialog box by the user and sending the consultation questions to the question analysis module;
The problem analysis module is used for acquiring problem similar parameters according to the problem text and the reference text i and sending the problem similar parameters to the data analysis module; the problem similarity parameters comprise word number information ZS, homonym information TF, keyword information GJ and number information BH;
the data analysis module is used for obtaining a problem similarity coefficient XSi according to the problem similarity parameters and sending the problem similarity coefficient XSi to the problem matching platform;
The specific process of obtaining the problem similarity coefficient XSi by the data analysis module is as follows:
Quantizing the word number information ZS, the co-symbol information TF, the keyword information GJ and the number information BH, extracting numerical values of the word number information ZS, the co-symbol information TF, the keyword information GJ and the number information BH, substituting the numerical values into a formula for calculation, and calculating according to the formula Obtaining a problem similarity coefficient XSi, wherein delta is a preset error adjustment factor, delta=0.925 is taken as a mathematical constant, s1, s2, s3 and s4 are respectively preset weight factors corresponding to set word number information ZS, homonym information TF, keyword information GJ and number information BH, s1, s2, s3 and s4 meet the condition that s4 & gts 3 & gts 2 & gts 1 & gt0.867, s1=1.18, s2=1.83, s3=2.55 and s4=3.26;
the problem similarity coefficient XSi is sent to a problem matching platform;
The problem matching platform is used for classifying the reference text i into an unqualified reference text and a qualified reference text according to the problem similarity coefficient XSi, obtaining a qualified quantity value, generating a manual conversion instruction according to the qualified quantity value, and sending the manual conversion instruction to the manual conversion module, or obtaining a reference problem list according to the qualified quantity value, and sending the reference problem list to the user consultation module;
the user consultation module is used for automatically replying to the problems consulted by the user in the dialog box according to the reference problem list, or generating a manual conversion instruction and sending the manual conversion instruction to the manual conversion module;
And the manual conversion module is used for converting the problem that the staff replies the user consultation after receiving the manual conversion instruction.
As a further scheme of the invention: the specific process of the problem analysis module obtaining word number information ZS is as follows:
Acquiring the text of the consulted questions, marking the text as a question text, sequentially marking all preset query question texts stored in a question retrieval library as reference texts i, wherein i=1, … …, n and n are positive integers, n is the total number of the reference texts, and i is the number of any one of the reference texts;
Acquiring the total number of characters in the question text and the reference text i, acquiring the difference between the two, and marking the difference as word number information ZS; wherein, the characters comprise Chinese characters, numbers, english letters and punctuation marks.
As a further scheme of the invention: the specific process of the problem analysis module obtaining the same symbol information TF is as follows:
The total number of identical characters in the question text and the reference text i is obtained and marked as co-character information TF.
As a further scheme of the invention: the specific process of the problem analysis module obtaining the keyword information GJ is as follows:
Matching the question text with a preset keyword list, obtaining the number of keywords in the question text, marking the number as a word value CS, obtaining the number of keywords in the question text and the number of keywords in a reference text i, obtaining the difference between the number of keywords and the number of keywords as a same word value TC, performing quantization processing on the word value CS and the same word value TC, extracting the values of the word value CS and the same word value TC, substituting the values into a formula, calculating according to the formula, and obtaining the number of keywords in the question text and the number of keywords in the reference text i And obtaining keyword information GJ, wherein g1 and g2 are respectively set word values CS and preset proportional coefficients corresponding to the same word values TC, g1 and g2 meet g1+g2=1, 0 < g1 < g2 < 1, g1=0.26 and g2=0.74.
As a further scheme of the invention: the specific process of the problem analysis module obtaining the number information BH is as follows:
And numbering the characters in the question text and the reference text i from front to back, taking the number of the first character of the keyword as the keyword number, acquiring the number difference value between the keyword numbers of the same keyword in the question text and the reference text i, marking the number difference value as a number difference value HC, acquiring the average value of the number difference values HC of all keywords, and marking the average value as number information BH.
As a further scheme of the invention: the specific process of the problem matching platform obtaining the unqualified reference text or the qualified reference text is as follows:
All the problem similarity coefficients XSi are sequentially compared with a preset problem similarity threshold XSy:
if the problem similarity coefficient XSi is smaller than the problem similarity threshold XSy, marking the reference text i corresponding to the problem similarity coefficient XSi as a disqualified reference text;
if the problem similarity coefficient XSi is greater than or equal to the problem similarity threshold XSy, marking the reference text i corresponding to the problem similarity coefficient XSi as a qualified reference text.
As a further scheme of the invention: the specific process of generating the manual conversion instruction by the problem matching platform is as follows:
And acquiring the number of the qualified reference texts, marking the number of the qualified reference texts as a qualified number value, generating a manual conversion instruction if the qualified number value=0, and sending the manual conversion instruction to the manual conversion module.
As a further scheme of the invention: the specific process of the problem matching platform obtaining the reference problem list is as follows:
And if the qualified quantity value is more than 0, sequencing all the qualified reference texts according to the sequence of the problem similarity coefficients XSi from large to small to form a reference problem list, and sending the reference problem list to the user consultation module.
As a further scheme of the invention: the specific process of the user consultation module for automatically replying or generating the manual conversion instruction is as follows:
Marking a qualified reference text positioned at the first place in a reference question list as a selected reference text, and generating a confirmation option according to the selected reference text, wherein the confirmation option comprises a confirmation question and a selection button, and the specific content of the confirmation question is 'is the question asking you to consult' is the selected reference text? "the specific contents of the selection keys are" yes "key and" no ";
If the user clicks the 'yes' key, a preset question reference answer text corresponding to the selected reference text is obtained, and the preset question reference answer text is sent to a dialog box;
If the user clicks the 'no' key, marking the qualified reference text positioned at the second position in the reference question list as the selected reference text, regenerating the confirmation options, if the total times of generating the confirmation options reach the preset times, generating a manual conversion instruction if the user still does not click the 'yes' key, and sending the manual conversion instruction to the manual conversion module.
As a further scheme of the invention: a service robot problem matching method facing to a multi-service scene comprises the following steps:
step one: the user inputs the consulted questions in the dialog box by using the user consultation module, and sends the consulted questions to the question analysis module;
Step two: the problem analysis module acquires problem similar parameters according to the problem text and the reference text i, wherein the problem similar parameters comprise word number information ZS, homonym information TF, keyword information GJ and number information BH, and the problem similar parameters are sent to the data analysis module;
Step three: the data analysis module obtains a problem similarity coefficient XSi according to the problem similarity parameters, and sends the problem similarity coefficient XSi to the problem matching platform;
Step four: the problem matching platform classifies the reference text i into an unqualified reference text and a qualified reference text according to the problem similarity coefficient XSi, obtains a qualified quantity value, generates a manual conversion instruction according to the qualified quantity value, and sends the manual conversion instruction to the manual conversion module, or obtains a reference problem list according to the qualified quantity value, and sends the reference problem list to the user consultation module;
step five: the user consultation module automatically replies the problems consulted by the user in the dialog box according to the reference problem list, or generates a manual conversion instruction and sends the manual conversion instruction to the manual conversion module;
step six: and after receiving the manual conversion instruction, the manual conversion module converts the problem that the staff replies the user consultation.
The invention has the beneficial effects that:
according to the service robot problem matching method and system for the multi-service scene, firstly, the problem text and the reference text are compared to obtain the problem similarity parameters, the similarity degree of the problem text and the reference text can be comprehensively measured according to the problem similarity coefficients obtained by the problem similarity parameters, the greater the problem similarity coefficients are, the higher the similarity degree is, finally, the problem is efficiently replied by multiple times of matching, so that the problem is reasonably matched, if the problem cannot be reasonably and pre-set to inquire, the problem is replied manually, the replied problem and answer are supplemented to a problem retrieval library, and the storage capacity of the problem retrieval library is enlarged;
the method realizes high-accuracy similarity judgment and screening through multiple factors, realizes high-performance and high-accuracy problem matching, improves the accuracy and efficiency of answer return, can adapt to the requirements of different service scenes, provides powerful support for the wide application of service robots, realizes intelligent matching and answering of the problems of multiple service scenes, and continuously improves the problem identification and matching capability along with the continuous expansion of a problem search library, thereby having flexibility and expandability, improving the problem processing capability of the service robots, improving the service quality and having better universality.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a schematic block diagram of a service robot problem matching system for a multi-business scenario in the present invention;
Fig. 2 is a service robot problem matching method facing to a multi-service scene in the present invention.
Detailed Description
The technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1: referring to fig. 1, the present embodiment is a service robot problem matching system for a multi-service scenario, including the following modules: the system comprises a user consultation module, a problem analysis module, a data analysis module, a problem matching platform and a manual conversion module;
The user consultation module is used for inputting the consultation questions in the dialog box by the user and sending the consultation questions to the question analysis module; the system is also used for automatically replying to the problems consulted by the user in the dialog box according to the reference problem list, or generating a manual conversion instruction and sending the manual conversion instruction to the manual conversion module;
The problem analysis module is used for acquiring similar parameters of the problem according to the problem text and the reference text i and sending the similar parameters of the problem to the data analysis module; the problem similarity parameters comprise word number information ZS, homonym information TF, keyword information GJ and number information BH;
the data analysis module is used for obtaining a problem similarity coefficient XSi according to the problem similarity parameters and sending the problem similarity coefficient XSi to the problem matching platform;
The problem matching platform is used for classifying the reference text i into an unqualified reference text and a qualified reference text according to the problem similarity coefficient XSi, obtaining a qualified quantity value, generating a manual conversion instruction according to the qualified quantity value, and sending the manual conversion instruction to the manual conversion module, or obtaining a reference problem list according to the qualified quantity value, and sending the reference problem list to the user consultation module;
the manual conversion module is used for receiving manual conversion instructions and then converting the problems that staff replies to user consultation.
Example 2: referring to fig. 2, the present embodiment is a service robot problem matching method for a multi-service scenario, including the following steps:
step one: the user inputs the consulted questions in the dialog box by using the user consultation module, and sends the consulted questions to the question analysis module;
Step two: the problem analysis module acquires problem similar parameters according to the problem text and the reference text i, wherein the problem similar parameters comprise word number information ZS, homonym information TF, keyword information GJ and number information BH, and the problem similar parameters are sent to the data analysis module;
Step three: the data analysis module obtains a problem similarity coefficient XSi according to the problem similarity parameters, and sends the problem similarity coefficient XSi to the problem matching platform;
Step four: the problem matching platform classifies the reference text i into an unqualified reference text and a qualified reference text according to the problem similarity coefficient XSi, obtains a qualified quantity value, generates a manual conversion instruction according to the qualified quantity value, and sends the manual conversion instruction to the manual conversion module, or obtains a reference problem list according to the qualified quantity value, and sends the reference problem list to the user consultation module;
step five: the user consultation module automatically replies the problems consulted by the user in the dialog box according to the reference problem list, or generates a manual conversion instruction and sends the manual conversion instruction to the manual conversion module;
step six: and after receiving the manual conversion instruction, the manual conversion module converts the problem that the staff replies the user consultation.
Example 3: based on any one of the above embodiments, embodiment 3 of the present invention is a user consultation module, which has two functions;
One function is to input the problem of consultation, and the specific process is as follows:
the user inputs the consulted questions in the dialog box by using the user consultation module, and sends the consulted questions to the question analysis module;
the second function is to automatically reply the problem of the user consultation or generate a manual conversion instruction, and the specific process is as follows:
The user consultation module marks the qualified reference text positioned at the first position in the reference question list as a selected reference text, and generates a confirmation option according to the selected reference text, wherein the confirmation option comprises a confirmation question and a selection button, and the specific content of the confirmation question is 'is the question asking you to consult' is the selected reference text? "the specific contents of the selection keys are" yes "key and" no ";
If the user clicks the 'yes' key, a preset question reference answer text corresponding to the selected reference text is obtained, and the preset question reference answer text is sent to a dialog box;
If the user clicks the 'no' key, marking the qualified reference text positioned at the second position in the reference question list as the selected reference text, regenerating the confirmation options, if the total times of generating the confirmation options reach the preset times, generating a manual conversion instruction if the user still does not click the 'yes' key, and sending the manual conversion instruction to the manual conversion module.
Example 4: based on any of the above embodiments, embodiment 4 of the present invention is a problem analysis module, which is used for obtaining similar parameters of a problem, where the similar parameters of the problem include word number information ZS, homonym information TF, keyword information GJ, and number information BH, and the specific process is as follows:
The method comprises the steps that a question analysis module obtains a text of a consulted question, marks the text as a question text, marks all preset question text stored in a question retrieval library as reference texts i, i=1, … …, n and n in sequence, wherein the value of n is a positive integer, n is the total number of the reference texts, and i is the number of any one of the reference texts;
The problem analysis module acquires the total number of characters in the problem text and the reference text i, acquires the difference between the two, and marks the difference as word number information ZS; wherein, the characters comprise Chinese characters, numbers, english letters and punctuation marks;
the problem analysis module acquires the total number of the same characters in the problem text and the reference text i, and marks the same characters as co-symbol information TF;
The problem analysis module matches the problem text with a preset keyword list, acquires the number of keywords in the problem text, marks the number of keywords as a word value CS, acquires the number of keywords in the problem text and the number of keywords in a reference text i, acquires the difference between the number of keywords and the number of keywords as a same word value TC, carries out quantization processing on the word value CS and the same word value TC, extracts the values of the word value CS and the same word value TC, substitutes the values into a formula to calculate, and calculates according to the formula Obtaining keyword information GJ, wherein g1 and g2 are respectively set word values CS and preset proportional coefficients corresponding to the same word values TC, g1 and g2 meet g1+g2=1, 0 < g1 < g2 < 1, g1=0.26 and g2=0.74;
The method comprises the steps that a problem analysis module carries out numerical numbering on characters in a problem text and a reference text i from front to back, takes the number of a first character of a keyword as a keyword number, obtains a numerical difference value between keyword numbers of the same keyword in the problem text and the reference text i, marks the numerical difference value as a numerical difference value HC, obtains an average value of the numerical difference values HC of all keywords, and marks the average value as numerical information BH;
The problem analysis module sends word number information ZS, same symbol information TF, keyword information GJ and number information BH to the data analysis module.
Example 5: based on any of the above embodiments, embodiment 5 of the present invention is a data analysis module, which is used for obtaining a problem similarity coefficient XSi, and specifically includes the following steps:
the data analysis module carries out quantization processing on word number information ZS, co-symbol information TF, keyword information GJ and number information BH, extracts numerical values of the word number information ZS, the co-symbol information TF, the keyword information GJ and the number information BH, substitutes the numerical values into a formula to calculate, and then calculates according to the formula Obtaining a problem similarity coefficient XSi, wherein delta is a preset error adjustment factor, delta=0.925 is taken as a mathematical constant, s1, s2, s3 and s4 are respectively preset weight factors corresponding to set word number information ZS, homonym information TF, keyword information GJ and number information BH, s1, s2, s3 and s4 meet the condition that s4 & gts 3 & gts 2 & gts 1 & gt0.867, s1=1.18, s2=1.83, s3=2.55 and s4=3.26;
the data analysis module sends the problem similarity coefficient XSi to the problem matching platform.
Example 6: based on any of the above embodiments, embodiment 6 of the present invention is a problem matching platform, where the problem matching platform is used to generate a manual conversion instruction or obtain a reference problem list according to a qualified quantity value, and the specific process is as follows:
The problem matching platform sequentially compares all the problem similarity coefficients XSi with a preset problem similarity threshold XSy:
if the problem similarity coefficient XSi is smaller than the problem similarity threshold XSy, marking the reference text i corresponding to the problem similarity coefficient XSi as a disqualified reference text;
If the problem similarity coefficient XSi is more than or equal to the problem similarity threshold XSy, marking the reference text i corresponding to the problem similarity coefficient XSi as a qualified reference text;
The method comprises the steps that a problem matching platform obtains the number of qualified reference texts and marks the number of the qualified reference texts as a qualified number value, if the qualified number value=0, a manual conversion instruction is generated, the manual conversion instruction is sent to a manual conversion module, if the qualified number value is more than 0, all the qualified reference texts are ordered according to the sequence of problem similarity coefficients XSi from large to small, a reference problem list is formed, and the reference problem list is sent to a user consultation module.
Example 7: based on any one of the above embodiments, embodiment 7 of the present invention is an artificial conversion module, where the function of the artificial conversion module is to recover the problem of user consultation by the conversion staff, and the specific process is as follows:
the manual conversion module receives a manual conversion instruction, and then converts a problem consulted by a user, and matches the consulted problem with a replied problem answer, and stores the problem in a problem retrieval library as a preset query problem text and a preset problem reference answer text.
Based on examples 1-7, the working principle of the invention is as follows:
According to the service robot problem matching method and system for the multi-service scene, a user inputs the problem of consultation in a dialog box by using a user consultation module, a problem analysis module acquires problem similar parameters according to a problem text and a reference text, wherein the problem similar parameters comprise word number information, homonym information, keyword information and serial number information, a data analysis module acquires a problem similar coefficient according to the problem similar parameters, a problem matching platform classifies the reference text into an unqualified reference text and a qualified reference text according to the problem similar coefficient, a qualified quantity value is acquired, a manual conversion instruction is generated according to the qualified quantity value, or a reference problem list is acquired according to the qualified quantity value, the user consultation problem is automatically replied in the dialog box by using the user consultation module according to the reference problem list, or a manual conversion instruction is generated, and a conversion staff replies the problem of the user consultation after receiving the manual conversion instruction by using the manual conversion module; according to the method, firstly, a question text and a reference text are compared to obtain a question similarity parameter, the similarity degree of the question text and the reference text can be comprehensively measured according to the question similarity coefficient obtained by the question similarity parameter, the larger the question similarity coefficient is, the higher the similarity degree is, and finally, the questions are efficiently replied to users through multiple times of matching, so that the questions are reasonably matched, if the reasonable preset inquiry questions cannot be matched, the questions are manually replied, the replied questions and answers are supplemented to a question retrieval library, and the storage capacity of the question retrieval library is enlarged; the method realizes high-accuracy similarity judgment and screening through multiple factors, realizes high-performance and high-accuracy problem matching, improves the accuracy and efficiency of answer return, can adapt to the requirements of different service scenes, provides powerful support for the wide application of service robots, realizes intelligent matching and answering of the problems of multiple service scenes, and continuously improves the problem identification and matching capability along with the continuous expansion of a problem search library, thereby having flexibility and expandability, improving the problem processing capability of the service robots, improving the service quality and having better universality.
It should be further described that, the above formulas are all the dimensionality removing and numerical calculation, the formulas are formulas for obtaining the latest real situation by software simulation by collecting a large amount of data, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
In the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The foregoing is merely illustrative and explanatory of the invention, as various modifications and additions may be made to the particular embodiments described, or in a similar manner, by those skilled in the art, without departing from the scope of the invention or exceeding the scope of the invention as defined in the claims.
Claims (10)
1. The service robot problem matching system for the multi-service scene is characterized by comprising the following components:
The user consultation module is used for inputting the consultation questions in the dialog box by the user and sending the consultation questions to the question analysis module;
The problem analysis module is used for acquiring problem similar parameters according to the problem text and the reference text i and sending the problem similar parameters to the data analysis module; the problem similarity parameters comprise word number information ZS, homonym information TF, keyword information GJ and number information BH;
the data analysis module is used for obtaining a problem similarity coefficient XSi according to the problem similarity parameters and sending the problem similarity coefficient XSi to the problem matching platform;
The specific process of obtaining the problem similarity coefficient XSi by the data analysis module is as follows:
Quantizing word number information ZS, homonymous information TF, keyword information GJ and number information BH, and according to a formula Obtaining a problem similarity coefficient XSi, wherein delta is a preset error adjustment factor, e is a mathematical constant, and s1, s2, s3 and s4 are preset weight factors corresponding to set word number information ZS, homonym information TF, keyword information GJ and number information BH respectively;
the problem similarity coefficient XSi is sent to a problem matching platform;
The problem matching platform is used for classifying the reference text i into an unqualified reference text and a qualified reference text according to the problem similarity coefficient XSi, obtaining a qualified quantity value, generating a manual conversion instruction according to the qualified quantity value, and sending the manual conversion instruction to the manual conversion module, or obtaining a reference problem list according to the qualified quantity value, and sending the reference problem list to the user consultation module;
the user consultation module is used for automatically replying to the problems consulted by the user in the dialog box according to the reference problem list, or generating a manual conversion instruction and sending the manual conversion instruction to the manual conversion module;
And the manual conversion module is used for converting the problem that the staff replies the user consultation after receiving the manual conversion instruction.
2. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of obtaining word number information ZS by the problem analysis module is as follows:
Acquiring the text of the consulted questions, marking the text as a question text, and sequentially marking all preset query question texts stored in a question retrieval library as a reference text i;
Acquiring the total number of characters in the question text and the reference text i, acquiring the difference between the two, and marking the difference as word number information ZS; wherein, the characters comprise Chinese characters, numbers, english letters and punctuation marks.
3. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of the problem analysis module obtaining the co-symbol information TF is as follows:
The total number of identical characters in the question text and the reference text i is obtained and marked as co-character information TF.
4. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of obtaining the keyword information GJ by the problem analysis module is as follows:
matching the question text with a preset keyword list, obtaining the number of keywords in the question text, marking the number of keywords as a word value CS, obtaining the number of keywords in the question text and the number of keywords in a reference text i, obtaining the difference between the two, marking the difference as a same word value TC, carrying out quantization processing on the word value CS and the same word value TC, and carrying out quantization processing on the same word value TC according to a formula And obtaining keyword information GJ, wherein g1 and g2 are respectively set word values CS and preset proportional coefficients corresponding to the same word value TC.
5. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of the problem analysis module obtaining the number information BH is as follows:
And numbering the characters in the question text and the reference text i from front to back, taking the number of the first character of the keyword as the keyword number, acquiring the number difference value between the keyword numbers of the same keyword in the question text and the reference text i, marking the number difference value as a number difference value HC, acquiring the average value of the number difference values HC of all keywords, and marking the average value as number information BH.
6. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of obtaining the unqualified reference text or the qualified reference text by the problem matching platform is as follows:
All the problem similarity coefficients XSi are sequentially compared with a preset problem similarity threshold XSy:
if the problem similarity coefficient XSi is smaller than the problem similarity threshold XSy, marking the reference text i corresponding to the problem similarity coefficient XSi as a disqualified reference text;
if the problem similarity coefficient XSi is greater than or equal to the problem similarity threshold XSy, marking the reference text i corresponding to the problem similarity coefficient XSi as a qualified reference text.
7. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of generating the manual conversion instruction by the problem matching platform is as follows:
And acquiring the number of the qualified reference texts, marking the number of the qualified reference texts as a qualified number value, generating a manual conversion instruction if the qualified number value=0, and sending the manual conversion instruction to the manual conversion module.
8. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of the problem matching platform obtaining the reference problem list is as follows:
And if the qualified quantity value is more than 0, sequencing all the qualified reference texts according to the sequence of the problem similarity coefficients XSi from large to small to form a reference problem list, and sending the reference problem list to the user consultation module.
9. The service robot problem matching system for multi-service scenarios according to claim 1, wherein the specific process of the user consultation module for automatically replying to or generating the manual conversion instruction is as follows:
Marking a qualified reference text positioned at the first place in a reference question list as a selected reference text, and generating a confirmation option according to the selected reference text, wherein the confirmation option comprises a confirmation question and a selection button, and the specific content of the confirmation question is 'is the question asking you to consult' is the selected reference text? "the specific contents of the selection keys are" yes "key and" no ";
If the user clicks the 'yes' key, a preset question reference answer text corresponding to the selected reference text is obtained, and the preset question reference answer text is sent to a dialog box;
If the user clicks the 'no' key, marking the qualified reference text positioned at the second position in the reference question list as the selected reference text, regenerating the confirmation options, if the total times of generating the confirmation options reach the preset times, generating a manual conversion instruction if the user still does not click the 'yes' key, and sending the manual conversion instruction to the manual conversion module.
10. The service robot problem matching method for the multi-service scene is characterized by comprising the following steps of:
step one: the user inputs the consulted questions in the dialog box by using the user consultation module, and sends the consulted questions to the question analysis module;
Step two: the problem analysis module acquires problem similar parameters according to the problem text and the reference text i, wherein the problem similar parameters comprise word number information ZS, homonym information TF, keyword information GJ and number information BH, and the problem similar parameters are sent to the data analysis module;
Step three: the data analysis module obtains a problem similarity coefficient XSi according to the problem similarity parameters, and sends the problem similarity coefficient XSi to the problem matching platform;
Step four: the problem matching platform classifies the reference text i into an unqualified reference text and a qualified reference text according to the problem similarity coefficient XSi, obtains a qualified quantity value, generates a manual conversion instruction according to the qualified quantity value, and sends the manual conversion instruction to the manual conversion module, or obtains a reference problem list according to the qualified quantity value, and sends the reference problem list to the user consultation module;
step five: the user consultation module automatically replies the problems consulted by the user in the dialog box according to the reference problem list, or generates a manual conversion instruction and sends the manual conversion instruction to the manual conversion module;
step six: and after receiving the manual conversion instruction, the manual conversion module converts the problem that the staff replies the user consultation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410557671.1A CN118132686B (en) | 2024-05-08 | 2024-05-08 | Service robot problem matching method and system for multi-service scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410557671.1A CN118132686B (en) | 2024-05-08 | 2024-05-08 | Service robot problem matching method and system for multi-service scene |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118132686A true CN118132686A (en) | 2024-06-04 |
| CN118132686B CN118132686B (en) | 2024-06-28 |
Family
ID=91240596
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410557671.1A Active CN118132686B (en) | 2024-05-08 | 2024-05-08 | Service robot problem matching method and system for multi-service scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118132686B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315766A (en) * | 2017-05-16 | 2017-11-03 | 广东电网有限责任公司江门供电局 | A kind of voice response method and its device for gathering intelligence and artificial question and answer |
| JP2021144348A (en) * | 2020-03-10 | 2021-09-24 | 富士通株式会社 | Information processing device and information processing method |
-
2024
- 2024-05-08 CN CN202410557671.1A patent/CN118132686B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107315766A (en) * | 2017-05-16 | 2017-11-03 | 广东电网有限责任公司江门供电局 | A kind of voice response method and its device for gathering intelligence and artificial question and answer |
| JP2021144348A (en) * | 2020-03-10 | 2021-09-24 | 富士通株式会社 | Information processing device and information processing method |
Non-Patent Citations (2)
| Title |
|---|
| DIMAS WIBISONO PRAKOS 等: "Short text similarity measurement methods: a review", 《SOFT COMPUTING》, 3 January 2021 (2021-01-03), pages 4699 * |
| 付鹏斌 等: "结合学科情感分析与依存关系的相似度评分", 《计算机技术与发展》, 10 February 2022 (2022-02-10), pages 32 - 38 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118132686B (en) | 2024-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111125328B (en) | Text processing method and related equipment | |
| JP3587120B2 (en) | Questionnaire response analysis system | |
| CN112784011B (en) | Emotion problem processing method, device and medium based on CNN and LSTM | |
| CN109902087B (en) | Data processing method and device for questions and answers and server | |
| CN117235238B (en) | Question answering method, question answering device, storage medium and computer equipment | |
| CN109492085A (en) | Method, apparatus, terminal and storage medium are determined based on the answer of data processing | |
| CN112965991A (en) | Pre-calculation result generation method and device, electronic equipment and storage medium | |
| CN116303558A (en) | Query statement generation method, data query method and generation model training method | |
| CN118132686B (en) | Service robot problem matching method and system for multi-service scene | |
| CN110633363B (en) | Text entity recommendation method based on NLP and fuzzy multi-criterion decision | |
| CN117112902A (en) | Off-line reinforcement learning commodity recommendation system based on multi-mode contrast learning | |
| CN117033579A (en) | Novel knowledge base system and method based on LLMS | |
| CN113722452B (en) | Semantic-based rapid knowledge hit method and device in question-answering system | |
| CN116881409A (en) | Commodity information automatic question-answering method based on e-commerce knowledge graph | |
| CN111104951A (en) | Active learning method and device and terminal equipment | |
| CN114417010A (en) | Knowledge graph construction method and device for real-time workflow and storage medium | |
| CN112214683B (en) | Mixed recommendation model processing method, system and medium based on heterogeneous information network | |
| CN114528446A (en) | Hierarchical query method and device for industrial internet network identification data | |
| CN111897937A (en) | Question answering method, system, computing device and storage medium combining RPA and AI | |
| CN116578682B (en) | Intelligent consultation method and system for legal service | |
| JP2020027549A (en) | Argument analyzer and program | |
| CN117194991B (en) | High-dimensional data recommendation system and method based on GPU cluster | |
| CN118410152B (en) | Information processing method, question-answering method and question-answering system | |
| CN113568929B (en) | Data storage and query method and device and electronic equipment | |
| CN117633540B (en) | Sample data construction method and device |
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 |