CN115115291A - Session information quality inspection method, device and computer readable storage medium - Google Patents
Session information quality inspection method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN115115291A CN115115291A CN202211044623.XA CN202211044623A CN115115291A CN 115115291 A CN115115291 A CN 115115291A CN 202211044623 A CN202211044623 A CN 202211044623A CN 115115291 A CN115115291 A CN 115115291A
- Authority
- CN
- China
- Prior art keywords
- quality inspection
- index
- session information
- indexes
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- 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
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a session information quality inspection method, a session information quality inspection device and a computer readable storage medium, wherein the method comprises the steps of obtaining session information between a client and a customer service end, and extracting quality inspection indexes according to the session information, wherein the quality inspection indexes comprise session indexes, emotion indexes and feedback indexes; determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree; and determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index. The problem of session information need artifical quality control to lead to quality control inefficiency is solved.
Description
Technical Field
The present invention relates to the field of quality inspection technologies, and in particular, to a session information quality inspection method, device, and computer-readable storage medium.
Background
When the business personnel and the client communicate online through the internet, the conversation is realized based on the conversation page for displaying the conversation message, namely the conversation message is sent and displayed in the conversation page of the own party, so that the corresponding conversation message is received and displayed in the conversation page of the opposite party, and the conversation between the business personnel and the client is further realized. At present, in order to ensure that business personnel do not relate to illegal behaviors when conducting business expansion, quality control personnel manually review session messages formed when the business personnel and clients are in conversation so as to judge whether the session messages contain illegal key contents.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
The invention mainly aims to provide a session information quality inspection method, a session information quality inspection device and a computer readable storage medium, and aims to solve the problem of low efficiency of manual session information quality inspection.
In order to achieve the purpose, the invention provides a method for acquiring session information between a client and a customer service end, and extracting quality inspection indexes according to the session information, wherein the quality inspection indexes comprise session indexes, conversational indexes, emotional indexes and feedback indexes;
determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree;
and determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index.
Optionally, the conversation index includes speech rate information, volume information, conversation duration, mute duration, and mute duty ratio, the conversation index includes sensitive word information, reply matching degree, reply standard degree, and flow matching degree, the emotion index includes client emotion change parameter and customer service emotion change parameter, and the feedback index includes evaluation result.
Optionally, the step of determining the matching degree between the quality inspection index and a preset quality inspection index, and determining the score corresponding to each quality inspection index according to the matching degree includes:
acquiring index data corresponding to each preset quality inspection index, wherein the index data comprises an index range and an index score;
comparing each quality inspection index with an index range corresponding to a preset quality inspection index to obtain a matching degree;
and determining the score of each quality inspection index according to the matching degree and the index score.
Optionally, the step of determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index includes:
determining the weight corresponding to each quality inspection index according to the index data of the preset quality inspection indexes, wherein the index data comprises the weight;
performing weighted summation operation according to the weight and the fraction to generate a weighted sum value;
and determining the quality inspection result according to the weighted sum value.
Optionally, the step of determining the quality inspection result according to the weighted sum includes:
comparing the weighted sum value with a preset quality inspection result interval to determine a target interval corresponding to the weighted sum value;
and generating the quality inspection result according to the quality inspection grade corresponding to the target interval.
Optionally, the method further comprises:
determining a quality inspection rule according to the type of the session information;
and according to the quality inspection rule, executing the steps of determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree.
Optionally, when the session information is voice session information, the step of extracting a quality inspection index according to the session information includes:
performing speech rate detection operation, silence detection operation, volume detection operation and conversation duration detection operation according to the conversation information to extract a conversation index corresponding to the conversation information;
determining text information corresponding to the session information according to a preset voice-to-text technology;
performing role separation operation on the text information to acquire first text information and second text information;
and performing preset processing operation based on the first text information and the second text information to obtain a conversational index, an emotional index and a feedback index corresponding to the conversational information, wherein the preset processing operation comprises sensitive word detection operation, knowledge base matching operation, regular expression rule matching operation and keyword fuzzy query operation.
Optionally, after the step of determining the quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index, the method further includes:
sending the quality inspection result and the session information to a quality inspector so that the quality inspector can perform manual quality inspection operation based on the session information and generate a manual quality inspection result;
and receiving the manual quality inspection result sent by the quality inspection personnel, and sending the quality inspection result and/or the manual quality inspection result to the customer service end so that the customer service end performs complaint operation on the basis of the quality inspection result and/or the manual quality inspection result and generates a complaint result.
In order to achieve the above object, the present invention also provides a session information quality inspection apparatus, including: the system comprises a memory, a processor and a session information quality inspection program which is stored on the memory and can run on the processor, wherein when the computer program is executed by the processor, the steps of the session information quality inspection method are realized.
In addition, to achieve the above object, the present invention further provides a computer-readable storage medium having a session information quality inspection program stored thereon, which when executed by a processor, implements the steps of the session information quality inspection method as described above.
According to the session information quality inspection method, the session information quality inspection device and the computer readable storage medium, session information between a client and a customer service end is obtained, and quality inspection indexes are extracted according to the session information, wherein the quality inspection indexes comprise session indexes, emotion indexes and feedback indexes; determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree; and determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index, determining the score corresponding to each quality inspection index by matching the quality inspection index corresponding to the session information with a preset quality inspection index, and determining the quality inspection result according to the score, so that the problem of low efficiency of manual quality inspection is solved, and the accuracy of quality inspection is improved.
Drawings
Fig. 1 is a schematic diagram of a terminal structure of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a session information quality inspection method according to a first embodiment of the present invention;
FIG. 3 is a flowchart illustrating a detailed process of step S10 of the session information quality inspection method according to the first embodiment of the present invention;
FIG. 4 is a flowchart illustrating a detailed process of step S20 of the session information quality inspection method according to the first embodiment of the present invention;
FIG. 5 is a flowchart illustrating a detailed process of step S30 of the session information quality inspection method according to the first embodiment of the present invention;
fig. 6 is a flowchart illustrating a detailed process of step S33 of the session information quality inspection method according to the first embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The main solution of the embodiment of the invention is as follows: acquiring session information between a client and a customer service terminal, and extracting quality inspection indexes according to the session information, wherein the quality inspection indexes comprise session indexes, conversational indexes, emotional indexes and feedback indexes; determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree; and determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index.
As shown in fig. 1, fig. 1 is a schematic terminal structure diagram of a hardware operating environment according to an embodiment of the present invention.
The terminal of the embodiment of the invention can be a PC, and can also be terminal equipment such as a smart phone, a tablet computer, a portable computer and the like.
As shown in fig. 1, the terminal may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory (e.g., a magnetic disk memory). The memory 1005 may alternatively be a storage device separate from the processor 1001.
Those skilled in the art will appreciate that the terminal structure shown in fig. 1 is not intended to be limiting and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a kind of computer storage medium, may include therein an operating system, a network communication module, a user interface module, and a session information quality inspection program.
In the terminal shown in fig. 1, the network interface 1004 is mainly used for connecting the backend server and the backend server
The server carries out data communication; the user interface 1003 is mainly used for connecting a client (user side) and performing data communication with the client; and the processor 1001 may be configured to call the session information quality inspection program stored in the memory 1005, and perform the following operations:
acquiring session information between a client and a customer service terminal, and extracting quality inspection indexes according to the session information, wherein the quality inspection indexes comprise session indexes, conversational indexes, emotional indexes and feedback indexes;
determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree;
and determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index.
Further, the processor 1001 may call the session information quality inspection program stored in the memory 1005, and further perform the following operations:
acquiring index data corresponding to each preset quality inspection index, wherein the index data comprise an index range and an index score;
comparing each quality inspection index with an index range corresponding to a preset quality inspection index to obtain a matching degree;
and determining the scores of the quality inspection indexes according to the matching degree and the index scores.
Further, the processor 1001 may call the session information quality inspection program stored in the memory 1005, and further perform the following operations:
determining the weight corresponding to each quality inspection index according to the index data of the preset quality inspection indexes, wherein the index data comprises the weight;
performing weighted summation operation according to the weight and the fraction to generate a weighted sum value;
and determining the quality inspection result according to the weighted sum value.
Further, the processor 1001 may call the session information quality inspection program stored in the memory 1005, and further perform the following operations:
comparing the weighted sum value with a preset quality inspection result interval to determine a target interval corresponding to the weighted sum value;
and generating the quality inspection result according to the quality inspection grade corresponding to the target interval.
Further, the processor 1001 may call the session information quality inspection program stored in the memory 1005, and further perform the following operations:
determining a quality inspection rule according to the type of the session information;
and according to the quality inspection rule, executing the steps of determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree.
Further, the processor 1001 may call the session information quality inspection program stored in the memory 1005, and further perform the following operations:
performing speech rate detection operation, silence detection operation, volume detection operation and conversation duration detection operation according to the conversation information to extract a conversation index corresponding to the conversation information;
determining text information corresponding to the session information according to a preset voice-to-text technology;
performing role separation operation on the text information to acquire first text information and second text information;
and performing preset processing operation based on the first text information and the second text information to obtain a conversational index, an emotional index and a feedback index corresponding to the conversational information, wherein the preset processing operation comprises sensitive word detection operation, knowledge base matching operation, regular expression rule matching operation and keyword fuzzy query operation.
First embodiment
Referring to fig. 2, a first embodiment of a session information quality inspection method according to the present invention provides a session information quality inspection method, including:
step S10, obtaining session information between the client and the customer service end, and extracting quality inspection indexes according to the session information, wherein the quality inspection indexes comprise session indexes, dialect indexes, emotion indexes and feedback indexes;
step S20, determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree;
and step S30, determining a quality inspection result corresponding to the session information according to the scores corresponding to the quality inspection indexes.
In this embodiment, the session information is session information between the client and the client, and the session information may be voice session information in a voice form or text session information in a text form. Optionally, the quality inspection indexes include session indexes, conversational indexes, emotion indexes and feedback indexes, the session indexes include speech speed information, volume information, session duration, mute duration and mute duty ratio, the conversational indexes include sensitive word information, reply matching degree, reply standard degree and flow matching degree, the emotion indexes include client emotion change parameters and customer service emotion change parameters, and the feedback indexes include evaluation results. Optionally, the quality inspection indexes include, but are not limited to, the above indexes, and may further include a call snatching index, where the call snatching index is used to determine whether the customer service interrupts the customer, and may further include a wrongly written word index, where the wrongly written word index is used to indicate the number of wrongly written words or the proportion of wrongly written words.
Optionally, after the session information is acquired, the quality inspection rules corresponding to different types are different, based on which, after the session information is acquired, the quality inspection rule is determined according to the type of the session information, and according to the quality inspection rule, the step of extracting the quality inspection index according to the session information is performed. Optionally, different types of session information correspond to different operation flows, for example, if the type of the session information is a fund query, the operation flow may be to confirm a question, check basic customer information, output a response result of the fund query, and if the type of the session information is a complaint handling, the operation flow may be to confirm a question, ask a reason, transfer a complaint department, or different types of session information correspond to different standard conversation techniques, or different types of session information correspond to different sensitive words. Based on this, in the embodiment of the application, the session information is obtained, and a corresponding quality inspection rule is determined for the type based on the session information, where the quality inspection rule includes preset quality inspection indexes and index data corresponding to each preset quality inspection index, where the index data includes an index range and an index score, and the index data may further include a weighted value corresponding to each preset quality inspection index, that is, the different preset quality inspection indexes have different degrees of influence on the session information quality inspection, and the higher the degree of influence is, the larger the weighted value is, the smaller the degree of influence is, and the smaller the weighted value is, for example, the degree of influence corresponding to the sensitive word information index is higher than the degree of influence of the voice information index, and then the weighted value corresponding to the sensitive word information index is higher than the weighted value corresponding to the voice information index.
Optionally, after obtaining the quality inspection rule, extracting a corresponding quality inspection index according to the quality inspection rule, referring to fig. 3, where, when the session information is voice session information, the S10 includes:
step S11, according to the conversation information, carrying out speech speed detection operation, silence detection operation, volume detection operation and conversation duration detection operation to extract a conversation index corresponding to the conversation information;
step S12, determining text information corresponding to the session information according to a preset voice-to-text technology;
step S12, performing role separation operation on the text information to obtain first text information and second text information;
step S14, based on the first text information and the second text information, performing preset processing operation,
and the preset processing operation comprises sensitive word detection operation, knowledge base matching operation, regular expression rule matching operation and keyword fuzzy query operation.
Optionally, after receiving the session information, performing a speech rate detection operation, a silence detection operation, a volume detection operation, and a session duration detection operation on the session information, where the speech rate detection operation is used to detect speech rate information of a client and speech rate information of a client, the speech information of the client includes a speech rate of each first dialog statement output by the client, and the speech information of the client includes a speech rate of each second dialog statement output by the client, where the speech rate information may be average speech rate information in a session process or speech rate information greater than or equal to a preset speech rate threshold; the volume detection operation user detects first volume information of the client and second volume information of the customer service side, wherein the volume information comprises the volume of each first dialog sentence and the volume of each second dialog sentence, and also comprises average volume information, and also can be volume information which is larger than or equal to a preset volume threshold; the silence detection operation is used for detecting a silence duration in a conversation process, the conversation duration detection operation is used for detecting a start timestamp and an end timestamp of a conversation, the conversation duration corresponding to the conversation process is determined according to the start timestamp and the end timestamp, optionally, after the silence duration and the conversation duration are determined, the silence ratio is determined according to the ratio of the silence duration to the conversation duration, the silence ratio is the proportion of the silence duration to the complete conversation duration, the silence duration can also be a reply duration corresponding to each conversation sentence, when a second conversation sentence is output by a client, the timestamp is a first time point, when the server replies to the second conversation sentence, the timestamp is a second time point, and the reply duration is a difference value between the second time point and the first time point, the mute duration may also be an average reply duration based on the respective reply durations.
Optionally, the voice conversation information is converted into text information according to a preset voice-to-text technology to determine text information corresponding to the conversation information, and the text information is subjected to role separation operation to obtain first text information and second text information, wherein the first text information includes text information corresponding to each first pair of utterances output by the customer service terminal, and the second text information includes text information corresponding to each second pair of utterances output by the client terminal. Optionally, after the first text information and the second text information are obtained, a preset processing operation is performed based on the first text information and the second text information to obtain a conversational index, an emotional index and a feedback index corresponding to the conversational information, where the preset processing operation includes a sensitive word detection operation, a knowledge base matching operation, a regular expression rule matching operation, and a keyword fuzzy query operation.
Optionally, the sensitive word detecting operation is configured to detect sensitive word information in the text information, where the sensitive word information includes a sensitive word, a number of sensitive words, and the like, the sensitive word may be an illegal word,
such as words with abusive nature, or words that are not suitable for public use, such as private information (identification numbers, home addresses, etc.).
Optionally, the knowledge base matching operation is configured to match session information in a session process with a knowledge base to obtain a reply matching degree, a reply standard degree, and a process matching degree, where the reply matching degree is configured to determine whether a first spoken sentence output by a customer service end matches a second spoken sentence output by a customer, and optionally, a manner of determining whether the first spoken sentence output by the customer service end matches the second spoken sentence output by the customer service end includes that a first conversational sentence output by the customer service end is consistent with a standard conversational reply of the knowledge base, for example, when "how much" a current commission is output by the customer ", a standard conversational reply corresponding to the knowledge base is: if the customer service end outputs 'please inquire about a customer manager', determining that a first dialogue sentence output by the customer service end is not matched with a second dialogue sentence output by the customer, and setting the reply matching degree to be 0, wherein the current commission is XX; optionally, the reply standard degree is used to determine whether the first spoken sentence output by the customer service end is consistent with a standard speech technology in the knowledge base, for example, if the standard speech technology is "this side is for you to communicate with the superior," and the first spoken sentence is "you call XX directly," then it is determined whether the first spoken sentence output by the customer service end is inconsistent with the standard speech technology in the knowledge base, then the reply standard degree is low, and the reply standard degree may be set to 0; the process matching degree is used for judging whether the operation process is consistent with a standard operation process of the knowledge base, the process matching degree comprises the contact ratio of the operation process and the standard operation process, and exemplarily, in an after-sales scene, if the standard operation process is: obtaining customer information, determining a problem, confirming a warranty date, confirming a customer address, and negotiating after-sale maintenance time, wherein the operation flow of the customer service end is as follows: obtaining customer information, determining a problem, transferring after-sales service, where the operation flow is inconsistent with the standard flow, the flow matching degree is low, optionally, the flow matching degree may be a determination of overlap ratio of the operation flow and the standard flow, where the matching degree is determined according to the number of matched flows and the number of total flows, for example, the number of matched flows is 5, the number of total flows is 10, the overlap ratio is 5/10=0.5, and the flow matching degree is 0.5.
Optionally, the regular expression rule matching operation is configured to match the first dialog statement with a knowledge base to detect whether the customer service outputs a standard dialogues or not, so as to obtain the reply standard level, where if the standard dialogues include an opening word, a closing word, and an identity confirmation word, if the opening word does not exist in the first dialog statement, and only the closing word and the identity confirmation word exist, the reply standard level is 2/3.
Optionally, the keyword fuzzy query operation is configured to obtain a question keyword of a second dialog statement, determine a standard dialog corresponding to the keyword according to the question keyword, and further determine a reply matching degree, a reply standard degree, and a process matching degree according to whether a reply to the question keyword in a first dialog statement matches the standard dialog.
Optionally, the keyword fuzzy query operation is further configured to obtain an emotion keyword in session information, determine the client emotion change parameter and the customer service emotion change parameter according to the emotion keyword, specifically, determine an emotion level according to the emotion keyword, determine the client emotion change parameter and the customer service emotion change parameter according to the emotion level, where the emotion level includes a normal level, a pleasure level, and an anger level. Exemplarily, reference is made to the following dialogs:
customer: you are good, i want to consult a repair reporting procedure. (Normal grade)
Customer service: you go well and trouble to provide your order information. I query for you immediately. (Normal grade)
Customer: good, troubled you, order information is xxx. (Happy grade)
Customer service: you are good, then, according to your order information, the user looks up that your order has passed the guarantee date, so you need to repair the user, and then, you assign a corresponding maintenance engineer. (Normal grade)
Customer: according to what, the user buys the product, the product is not maintained for free, and the user complains! What play! (angry level)
Customer service: the love is Za zha bar, complaints are complained, and the person is what enjoyment! (angry rating)
The customer emotion change parameter is a normal rating-a pleasure rating-an anger rating, and the customer service emotion change parameter is a normal rating-an anger rating.
Optionally, the manner of determining the emotion index may also be to perform determination by combining speech rate information, volume information, and emotion keywords, and when the speech rate is converted from a normal speech rate to a fast speech rate, or the volume is changed from a normal volume to a large volume, the emotion change parameter may be determined to be a normal level — an anger level. Optionally, determining the emotion change parameter includes, but is not limited to, the above.
Optionally, the keyword fuzzy query operation is further configured to obtain an evaluation keyword in the session information, and determine a feedback index according to the evaluation keyword, where the evaluation result includes an excellent level, a medium level, and a bad level, and for example, when the client outputs "you are too happy and thank you for me", the evaluation result is the excellent level, and when the client outputs "i want to complain and you have no right solution", the evaluation result is the bad level.
Optionally, the manner of obtaining the feedback index may also be that after the session is ended, an evaluation page is output, and the feedback index is determined according to information input by the client for the evaluation page.
Optionally, after the quality inspection indexes are determined, the matching degree of the quality inspection indexes and preset quality inspection indexes is determined, and the scores corresponding to the quality inspection indexes are determined according to the matching degree, wherein the higher the matching degree is, the higher the score is, the lower the matching degree is and the lower the score is.
Optionally, when the session information is text session information, the quality inspection index includes sensitive word information, reply matching degree, reply standard degree, process matching degree, client emotion change parameter, client service emotion change parameter, and evaluation result.
Alternatively, referring to fig. 4, the step S20 includes:
step S21, acquiring index data corresponding to each preset quality inspection index, wherein the index data comprises an index range and an index score;
step S22, comparing each quality inspection index with an index range corresponding to a preset quality inspection index to obtain a matching degree;
and step S23, determining the scores of the quality inspection indexes according to the matching degree and the index scores.
Optionally, the preset quality inspection indexes corresponding to different quality inspection indexes are different, the preset quality inspection indexes include a preset session index, a preset emotion index and a preset feedback index, when the quality inspection index is the session index, the preset quality inspection index corresponding to the session index is the preset session index, optionally, the index data corresponding to different preset quality inspection indexes are different, the index data includes an index range and an index score, exemplarily, the index data corresponding to the preset speech rate index in the preset session index is [20 words/min, 50 words/min ], the corresponding index score is 10, the index data corresponding to the preset volume index is [20dB, 60dB ], the corresponding index score is 10, and the index data corresponding to the preset silence ratio index includes: the index range is [0%, 20% ], the corresponding index score is 10 points, the index range is [20%, 40% ], the corresponding index score is 5 points, the index range is [40%, 100% ], the corresponding index score is 0 points; the index range corresponding to the preset sensitive word index comprises the following steps: the number of the sensitive words is less than 2, the corresponding index score is 10 points and exceeds 2, and the corresponding index score is 0 point; the index range corresponding to the preset reply matching degree index comprises [0%,100% ], and the corresponding index score is [0 score, 100 score ].
Optionally, each quality inspection index is compared with an index range corresponding to a preset quality inspection index to obtain a matching degree, and then the score of the quality inspection index is determined according to the matching degree and the index score, wherein different matching degrees correspond to different scores.
Optionally, the matching degree may be a matching probability, and the manner of determining the matching probability may be determining whether the quality inspection index is within an index range, if so, the matching probability is 100%, and if not, the matching probability is 0%, for example, an index range of index data corresponding to a preset speech rate index in the preset session index is: [20 words/minute, 50 words/minute ], the corresponding index score is 10 points, the speech rate information is 25 words/minute, the matching probability is 100%, the matching degree is 100%, the score corresponding to the speech rate information is 10 points, in addition, the speech rate of each first dialogue sentence output by the customer service end can be compared with the index range to obtain the matching degree corresponding to each first dialogue sentence, the score corresponding to each first dialogue sentence is further determined, and the score corresponding to the preset information is determined according to the score corresponding to each first dialogue sentence.
Alternatively, the matching probability may also be determined by determining a ratio of the quality inspection indicator to the indicator range, for example, the indicator range [0%,100% ] corresponding to the reply matching degree is preset, the corresponding indicator score is [0 point, 100 points ], when the reply matching degree is 50%, 50%/100% =50%, the matching probability is 50%, the matching degree is 50%, and the indicator score is 50 points.
Optionally, the index data may further include each index range and an index score corresponding to each index range, where different index ranges correspond to different index scores, specifically, the quality inspection index is compared with each index range to determine an index range corresponding to the quality inspection index, the index range corresponding to the quality inspection index is determined as the matching degree, and then the score of the quality inspection index is determined according to the index score corresponding to the index range corresponding to the quality inspection index, for example, when the quality inspection index is a client emotion change parameter, the corresponding preset quality inspection index is a preset emotion change index, and the index data corresponding to the preset emotion change index includes: changing a customer emotion change parameter from a normal grade to an anger grade, corresponding index scores being 0, changing a customer emotion change parameter from a normal grade to a pleasure grade, corresponding index scores being 6, changing a customer emotion change parameter from an anger grade to a normal grade, corresponding index scores being 8, changing a customer emotion change parameter from an anger grade to a pleasure grade, and corresponding index scores being 10, when a customer emotion change parameter is changed from a normal grade to an anger grade and then to a pleasure grade, determining an index score to be a combination of an index score corresponding to a normal grade changed from an angry grade to an angry grade and an index score corresponding to a pleasure grade changed from an anger grade to an angry grade, that is 0 score +10 =10, and when a customer emotion change parameter is changed from an angry grade to a normal grade and then to a pleasure grade, determining an index score to be a combination of an index score corresponding to a normal grade changed from an angry grade to a normal grade and an index score corresponding to a grade changed from a normal grade to a pleasant grade I.e. 8+6=14 points. It will be appreciated that the above description is by way of example only.
Optionally, after determining the score of each quality inspection index, determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index, and optionally, referring to fig. 4, the step S30 includes:
step S31, determining the weight corresponding to each quality inspection index according to the index data of the preset quality inspection index, wherein the index data comprises the weight;
step S32, carrying out weighted summation operation according to the weight and the fraction to generate a weighted sum value;
and step S33, determining the quality inspection result according to the weighted sum value.
Optionally, after the score corresponding to each quality inspection index is obtained, the weight corresponding to each quality inspection index is determined based on the index data corresponding to each preset quality inspection index, a weighted sum operation is performed according to each score and each weight to generate a weighted sum value, and then the quality inspection result is determined according to the weighted sum value. Optionally, the higher the weighted sum value is, the higher the service quality of the customer service end is represented, and after determining the weighted sum value, the method further includes:
step S331, comparing the weighted sum value with a preset quality inspection result interval to determine a target interval corresponding to the weighted sum value;
and S332, generating the quality inspection result according to the quality inspection grade corresponding to the target interval.
Optionally, the quality inspection grades corresponding to different preset quality inspection result intervals are different, for example, when the preset quality inspection result interval is (0,40), the quality inspection grade is poor, when the preset quality inspection result interval is (40,70), the quality inspection grade is medium, when the preset quality inspection result interval is (70,90), the quality inspection grade is good, and when the preset quality inspection result interval is (80,100), the quality inspection grade is excellent.
Optionally, the weighted sum is compared with preset quality inspection result intervals corresponding to all quality inspection grades to determine a target interval corresponding to the weighted sum, the target interval includes the weighted sum, the target interval is one of the preset quality inspection result intervals, and the quality inspection result is generated according to the quality inspection grade corresponding to the target interval. Optionally, the quality inspection result further includes a quality inspection detail table, where the quality inspection detail table includes scores corresponding to the quality inspection indexes, and matching degrees of the quality inspection indexes and preset quality inspection indexes, so that the customer service end can know the quality of service problem in the session in time according to the quality inspection result.
Optionally, after the step of determining the quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index, the method further includes:
sending the quality inspection result and the session information to a quality inspector so that the quality inspector can perform manual quality inspection operation based on the session information and generate a manual quality inspection result;
and receiving the manual quality inspection result sent by the quality inspection personnel, and sending the quality inspection result and/or the manual quality inspection result to the customer service end so that the customer service end performs complaint operation on the basis of the quality inspection result and/or the manual quality inspection result and generates a complaint result.
Optionally, after the quality inspection result is generated, the quality inspection result and the session information are sent to a quality inspector, so that the quality inspector performs manual quality inspection operation based on a quality inspection rule and a quality inspection experience corresponding to the session information, and generates a manual quality inspection result.
Optionally, in order to enable the customer service personnel to know the conversation quality inspection condition of the customer service personnel in time, after the manual quality inspection result is generated, the quality inspection result and/or the manual quality inspection result is sent to the customer service end, so that the customer service end performs a complaint operation on the basis of the quality inspection result and/or the manual quality inspection result and generates a complaint result, optionally, the complaint result includes a complaint reason, and exemplarily, the complaint reason may be a low score, a sensitive word recognition error, a keyword recognition error, and a standard language identification error.
Optionally, after receiving the complaint result, the manual quality inspection result and the quality inspection result may be compared according to the complaint result, and when the quality inspection result is greatly different from the manual quality inspection result, the quality inspection rule is rechecked or corrected.
In the embodiment of the application, after session information between a client and a customer service end is acquired, quality inspection indexes of multiple dimensions are extracted according to the session information, wherein the quality inspection indexes comprise session indexes, tactical indexes, emotion indexes and feedback indexes, the quality inspection indexes are matched with preset quality inspection results to determine scores corresponding to all the quality inspection indexes, weighting summation operation is performed according to weights corresponding to all the quality inspection indexes and weights corresponding to all the quality inspection indexes to determine scores corresponding to the session information, quality inspection results are determined according to the scores, automatic quality inspection is performed based on quality inspection rules, quality inspection efficiency is improved, quality inspection coverage is improved, and quality inspection cost is reduced.
In addition, an embodiment of the present invention further provides a computer-readable storage medium, where a session information quality inspection program is stored, and when executed by a processor, the session information quality inspection program implements the steps of the above-described embodiment.
It is noted that, in this document, the terms "comprises", "comprising" or any other variation thereof
It is intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) as described above and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A session information quality inspection method is characterized by comprising the following steps:
acquiring session information between a client and a customer service terminal, and extracting quality inspection indexes according to the session information, wherein the quality inspection indexes comprise session indexes, conversational indexes, emotional indexes and feedback indexes;
determining the matching degree of the quality inspection indexes and preset quality inspection indexes, and determining the corresponding scores of all the quality inspection indexes according to the matching degree;
and determining a quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index.
2. The method according to claim 1, wherein the session index includes speech rate information, volume information, session duration, mute duration, and mute duty ratio, the session index includes sensitive word information, reply matching degree, reply standard degree, and process matching degree, the emotion index includes client emotion change parameter and client emotion change parameter, and the feedback index includes evaluation result.
3. The session information quality inspection method according to claim 1, wherein the step of determining the matching degree of the quality inspection indexes and preset quality inspection indexes and determining the scores corresponding to the quality inspection indexes according to the matching degree comprises:
acquiring index data corresponding to each preset quality inspection index, wherein the index data comprise an index range and an index score;
comparing each quality inspection index with an index range corresponding to a preset quality inspection index to obtain a matching degree;
and determining the score of each quality inspection index according to the matching degree and the index score.
4. The method according to claim 1, wherein the step of determining the quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index comprises:
determining the weight corresponding to each quality inspection index according to the index data of the preset quality inspection indexes, wherein the index data comprises the weight;
performing weighted summation operation according to the weight and the fraction to generate a weighted sum value;
and determining the quality inspection result according to the weighted sum value.
5. The session information quality inspection method according to claim 4, wherein the step of determining the quality inspection result according to the weighted sum value comprises:
comparing the weighted sum value with a preset quality inspection result interval to determine a target interval corresponding to the weighted sum value;
and generating the quality inspection result according to the quality inspection grade corresponding to the target interval.
6. The session information quality inspection method of claim 1, wherein the method further comprises:
determining a quality inspection rule according to the type of the session information;
and according to the quality inspection rule, executing a step of extracting a quality inspection index according to the session information.
7. The method according to claim 1, wherein when the session information is voice session information, the step of extracting the quality control indicator according to the session information comprises:
carrying out speech rate detection operation, silence detection operation, volume detection operation and conversation duration detection operation according to the conversation information so as to extract a conversation index corresponding to the conversation information;
determining text information corresponding to the session information according to a preset voice-to-text technology;
performing role separation operation on the text information to acquire first text information and second text information;
and performing preset processing operation based on the first text information and the second text information to obtain a conversational index, an emotional index and a feedback index corresponding to the conversational information, wherein the preset processing operation comprises sensitive word detection operation, knowledge base matching operation, regular expression rule matching operation and keyword fuzzy query operation.
8. The method for quality inspection of session information according to claim 1, wherein after the step of determining the quality inspection result corresponding to the session information according to the score corresponding to each quality inspection index, the method further comprises:
sending the quality inspection result and the session information to a quality inspector so that the quality inspector can perform manual quality inspection operation based on the session information and generate a manual quality inspection result;
and receiving the manual quality inspection result sent by the quality inspection personnel, and sending the quality inspection result and/or the manual quality inspection result to the customer service end so that the customer service end performs complaint operation on the basis of the quality inspection result and/or the manual quality inspection result and generates a complaint result.
9. A session information quality inspection apparatus, characterized by comprising: a memory, a processor, and a session information quality testing program stored on the memory and executable on the processor, the session information quality testing program when executed by the processor implementing the steps of the session information quality testing method according to any one of claims 1 to 8.
10. A computer-readable storage medium, wherein a session information quality inspection program is stored on the computer-readable storage medium, and when executed by a processor, the session information quality inspection program implements the steps of the session information quality inspection method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211044623.XA CN115115291A (en) | 2022-08-30 | 2022-08-30 | Session information quality inspection method, device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211044623.XA CN115115291A (en) | 2022-08-30 | 2022-08-30 | Session information quality inspection method, device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115115291A true CN115115291A (en) | 2022-09-27 |
Family
ID=83335986
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211044623.XA Pending CN115115291A (en) | 2022-08-30 | 2022-08-30 | Session information quality inspection method, device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115115291A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104301554A (en) * | 2013-07-18 | 2015-01-21 | 中兴通讯股份有限公司 | Device and method used for detecting service quality of customer service staff |
| CN105184315A (en) * | 2015-08-26 | 2015-12-23 | 北京中电普华信息技术有限公司 | Quality inspection treatment method and system |
| CN108737243A (en) * | 2018-04-20 | 2018-11-02 | 平安科技（深圳）有限公司 | Conversation message quality detecting method and device |
| CN111314566A (en) * | 2020-01-20 | 2020-06-19 | 北京神州泰岳智能数据技术有限公司 | Voice quality inspection method, device and system |
| CN112804400A (en) * | 2020-12-31 | 2021-05-14 | 中国工商银行股份有限公司 | Customer service call voice quality inspection method and device, electronic equipment and storage medium |
| CN113361969A (en) * | 2021-07-07 | 2021-09-07 | 北京容联七陌科技有限公司 | Intelligent quality inspection system capable of flexibly configuring templates |
| CN113468296A (en) * | 2021-09-02 | 2021-10-01 | 杭州实在智能科技有限公司 | Model self-iteration type intelligent customer service quality inspection system and method capable of configuring business logic |
| CN113657773A (en) * | 2021-08-19 | 2021-11-16 | 中国平安人寿保险股份有限公司 | Method and device for testing speech technology, electronic equipment and storage medium |
| CN113849617A (en) * | 2021-09-23 | 2021-12-28 | 深圳追一科技有限公司 | Customer service quality inspection method and device, computer equipment and storage medium |
-
2022
- 2022-08-30 CN CN202211044623.XA patent/CN115115291A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104301554A (en) * | 2013-07-18 | 2015-01-21 | 中兴通讯股份有限公司 | Device and method used for detecting service quality of customer service staff |
| CN105184315A (en) * | 2015-08-26 | 2015-12-23 | 北京中电普华信息技术有限公司 | Quality inspection treatment method and system |
| CN108737243A (en) * | 2018-04-20 | 2018-11-02 | 平安科技（深圳）有限公司 | Conversation message quality detecting method and device |
| CN111314566A (en) * | 2020-01-20 | 2020-06-19 | 北京神州泰岳智能数据技术有限公司 | Voice quality inspection method, device and system |
| CN112804400A (en) * | 2020-12-31 | 2021-05-14 | 中国工商银行股份有限公司 | Customer service call voice quality inspection method and device, electronic equipment and storage medium |
| CN113361969A (en) * | 2021-07-07 | 2021-09-07 | 北京容联七陌科技有限公司 | Intelligent quality inspection system capable of flexibly configuring templates |
| CN113657773A (en) * | 2021-08-19 | 2021-11-16 | 中国平安人寿保险股份有限公司 | Method and device for testing speech technology, electronic equipment and storage medium |
| CN113468296A (en) * | 2021-09-02 | 2021-10-01 | 杭州实在智能科技有限公司 | Model self-iteration type intelligent customer service quality inspection system and method capable of configuring business logic |
| CN113849617A (en) * | 2021-09-23 | 2021-12-28 | 深圳追一科技有限公司 | Customer service quality inspection method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10832686B2 (en) | Method and apparatus for pushing information | |
| US10776580B2 (en) | Method for providing dialogue service with chatbot assisted by human agents | |
| CN107818798A (en) | Customer service quality evaluating method, device, equipment and storage medium | |
| US8145562B2 (en) | Apparatus and method for fraud prevention | |
| CN110580282B (en) | Method and device for interacting with customer service through simulation user | |
| US20110172989A1 (en) | Intelligent and parsimonious message engine | |
| CN107886951B (en) | Voice detection method, device and equipment | |
| KR101615848B1 (en) | Method and computer program of recommending dialogue sticker based on similar situation detection | |
| CN107133709B (en) | Quality inspection method, device and system for customer service | |
| CN108173835B (en) | Verification method, device, server and terminal | |
| CN110610705A (en) | Voice interaction prompter based on artificial intelligence | |
| CN113111658B (en) | Method, device, equipment and storage medium for checking information | |
| CN105869631B (en) | The method and apparatus of voice prediction | |
| CN115115291A (en) | Session information quality inspection method, device and computer readable storage medium | |
| CN114157763A (en) | Information processing method and device in interactive process, terminal and storage medium | |
| CN110765242A (en) | Method, device and system for providing customer service information | |
| CN113821625A (en) | Artificial intelligence based tactical recommendation method, device, equipment and medium | |
| CN113822029A (en) | Customer service assistance method, device and system | |
| CN112101035B (en) | Named entity identification method and device, readable storage medium and electronic equipment | |
| KR20200109995A (en) | A phising analysis apparatus and method thereof | |
| CN114065742B (en) | Text detection method and device | |
| US20200184352A1 (en) | Information output system, information output method, and recording medium | |
| CN114221920B (en) | Automatic contact method, device, computer equipment and medium based on artificial intelligence | |
| JP7353695B1 (en) | Information provision device | |
| CN113468297B (en) | Dialogue data processing method and device, electronic equipment and storage equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220927 |