CN115660363A - Dialogue processing method and device, electronic equipment and storage medium - Google Patents
Dialogue processing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115660363A CN115660363A CN202211379020.5A CN202211379020A CN115660363A CN 115660363 A CN115660363 A CN 115660363A CN 202211379020 A CN202211379020 A CN 202211379020A CN 115660363 A CN115660363 A CN 115660363A
- Authority
- CN
- China
- Prior art keywords
- customer service
- determining
- target
- sub
- historical
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure provides a conversation processing method and device, electronic equipment and a storage medium, and relates to the fields of intelligent cloud, computer vision, natural language processing, deep learning, cloud computing and the like. The specific implementation scheme is as follows: classifying the dialog to be processed to obtain a target category, wherein the target category is used for indicating a target product related to the dialog to be processed and/or a target field direction related to the target product; determining the degree of adaptation of at least one candidate customer service to a target class; determining a target customer service from at least one candidate customer service according to the adaptation degree; and allocating the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service. Therefore, the target customer service is familiar with the target product related to the to-be-processed conversation and/or the target field direction related to the target product, and the condition that the customer service forwards and reprocesses the conversation due to the fact that the customer service is not familiar with the product related to the conversation and/or the segment field direction related to the product can be avoided.
Description
Technical Field
The present disclosure relates to the field of artificial intelligence, and in particular, to the technical fields of smart cloud, computer vision, natural language processing, deep learning, cloud computing, and the like, and in particular, to a method and an apparatus for processing a dialog, an electronic device, and a storage medium.
Background
With the rapid development of internet technology and networked transaction business scenarios, tens of thousands of network consumption platforms (or network sales platforms) are emerging. When a customer browses products and services by using the network consumption platform, the customer cannot be relieved to purchase the products and services due to the fact that the customer lacks knowledge about the quality, performance, services and the like of some products, and the network consumption platform builds a conversation type customer service system platform in order to strengthen the knowledge of the customer about the products and services.
The interactive customer service system platform mainly provides main functions of customer service information management, customer consultation conversation distribution, conversation information content recording, customer evaluation feedback, historical conversation information inquiry and analysis and the like, wherein the most core function is the customer consultation conversation distribution function.
For a relatively large-scale network consumption platform, millions of clients can initiate product and service consultation conversations every day, correspondingly, tens of thousands of professional customer service personnel are needed to carry out conversation responses to accept the consultation services of the clients, so that the problems of the clients are solved to the maximum extent, and the trade of the products and the services is promoted. Therefore, how to achieve an optimal allocation of the client consultation sessions is very important.
Disclosure of Invention
The disclosure provides a dialogue processing method, a dialogue processing device, an electronic device and a storage medium.
According to an aspect of the present disclosure, there is provided a conversation processing method including: obtaining a dialog to be processed, and determining a target category to which the dialog to be processed belongs, wherein the target category is used for indicating a target product related to the dialog to be processed and/or a target field direction related to the target product; determining the adaptation degree of at least one candidate customer service to the target category; determining target customer service from the at least one candidate customer service according to the adaptability of the at least one candidate customer service; and distributing the to-be-processed dialog to the target customer service so as to reply the to-be-processed dialog through the target customer service.
According to another aspect of the present disclosure, there is provided a conversation processing apparatus including: the acquisition module is used for acquiring the dialog to be processed; the first determination module is used for determining a target category to which the to-be-processed conversation belongs, wherein the target category is used for indicating a target product related to the to-be-processed conversation and/or a target field direction related to the target product; the second determining module is used for determining the adaptation degree of at least one candidate customer service and the target category; the third determining module is used for determining target customer service from the at least one candidate customer service according to the adaptation degree of the at least one candidate customer service; and the distribution module is used for distributing the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service.
According to still another aspect of the present disclosure, there is provided an electronic device including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to perform a dialog processing method according to the above aspect of the present disclosure.
According to still another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium of computer instructions for causing a computer to perform a dialog processing method set forth in the above-described aspect of the present disclosure.
According to yet another aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the dialog processing method set forth in the above-described aspect of the present disclosure.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
fig. 1 is a schematic flowchart of a dialog processing method according to a first embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a dialog processing method according to a second embodiment of the present disclosure;
fig. 3 is a schematic flowchart of a dialog processing method according to a third embodiment of the present disclosure;
fig. 4 is a schematic flowchart of a dialog processing method according to a fourth embodiment of the present disclosure;
fig. 5 is a schematic flowchart of a dialog processing method according to a fifth embodiment of the present disclosure;
fig. 6 is a schematic flowchart of a dialog processing method according to a sixth embodiment of the present disclosure;
FIG. 7 is a schematic flow chart of OCR recognition provided by the embodiments of the present disclosure;
FIG. 8 is a schematic diagram illustrating a training process and a prediction process of a target classification model according to an embodiment of the present disclosure;
FIG. 9 is a schematic diagram illustrating a calculation process of the fitness according to an embodiment of the present disclosure;
fig. 10 is a schematic structural diagram of a session processing apparatus according to a seventh embodiment of the present disclosure;
FIG. 11 shows a schematic block diagram of an example electronic device that may be used to implement embodiments of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
The consultation processing result and the processing efficiency of customer service can influence the shopping experience of a client, and can influence the decision of the client whether to perform consumption attempt on a product on a network consumption platform. Under the circumstances, higher and higher requirements are put forward on aspects of professional business level, problem processing efficiency, comprehensive service capacity and the like of customer service.
The network consumption platform may relate to a large number of products, for example, by 8 months in 2022, taking the network consumption platform as XX smart cloud as an example, the XX smart cloud has online 300 cloud products, and each product may also relate to a plurality of subdivided domain directions, taking the XX smart cloud data governance platform product as an example, the data governance platform product may relate to subdivided domain directions including: metadata collection, data development, data analysis, data operation, data service, data quality and other fields.
However, each customer service person may not be familiar with each domain direction of each product, the art has a special interest, and the customer service person can only be familiar with the specific domain direction (such as data job & data service & data quality) to improve the processing efficiency of the customer consultation problem. And several conditions are common:
first, customers often do not know the exact classification of a product (e.g., direct input data services) but rather know what problem they have encountered and want to know and solve the problem. Taking the example of a client session, assuming that the data source is the database MySQL (relational database management system), the problem of the client may be "how to improve the processing efficiency when processing big data import tasks? "or" how to solve the task faster to complete when processing the big data import task? ".
Secondly, the customer may not input text and only send a picture of the product abnormality, and at this time, if the customer is randomly assigned with a customer service, the processing efficiency is often low, for example, the assigned customer service personnel is not familiar with the product or the direction of the segmentation field related to the product, and the customer service personnel is required to forward and reprocess the conversation.
The intelligent dialogue allocation is a complex multidimensional intelligent technical problem and a core problem for determining the contribution capacity of a network consumption platform.
In the related technology, conversation allocation is realized based on a simple idle allocation and random allocation method, when a new client initiates product consultation or service consultation, one customer service staff is randomly searched from an idle customer service staff queue, and the client conversation is allocated to the customer service staff.
In this way, it often happens that after understanding a specific product problem, a customer service person performs secondary distribution of a conversation, that is, the conversation is distributed to a customer service person in a direction of a professional field corresponding to the product, or the conversation is distributed to a customer service person with more experience, or even more, multiple redistribution of the conversation may occur, which seriously affects the shopping experience of a customer, leads the customer to question the service level of a network consumption platform, and reduces the possibility of a deal.
In addition, pictures are very common dialog scenario information content, and can rapidly and detailedly reproduce specific scenes encountered by customers for customer service personnel, while the current common dialog distribution technology is not processed and used based on the information content of the pictures, and inefficient dialog distribution is easier to occur. Generally, customer service personnel can be divided into different levels or grades such as primary level, middle level, high level and the like, and some difficult problems related to products and services can only be solved by the customer service personnel allocated to the high level.
In view of at least one of the above problems, the present disclosure provides a dialog processing method, apparatus, electronic device, and storage medium.
A dialogue processing method, a dialogue processing apparatus, an electronic device, and a storage medium according to embodiments of the present disclosure are described below with reference to the drawings.
Fig. 1 is a flowchart illustrating a dialog processing method according to a first embodiment of the disclosure.
The embodiment of the present disclosure is exemplified by the dialog processing method being configured in a dialog processing apparatus, which can be applied to any electronic device, so that the electronic device can execute a dialog processing function.
The electronic device may be any device with computing capability, for example, a PC (Personal Computer), a mobile terminal, a server, and the like, and the mobile terminal may be a hardware device with various operating systems, touch screens, and/or display screens, such as an in-vehicle device, a mobile phone, a tablet Computer, a Personal digital assistant, and a wearable device.
As shown in fig. 1, the dialog processing method may include the steps of:
In the embodiment of the present disclosure, the pending dialog may be a dialog input by the user, and the input manner includes, but is not limited to, touch input (such as sliding, clicking, etc.), keyboard input, voice input, and the like. The dialog to be processed may include at least one of text information, picture information, audio information, and video information.
As an application scenario, the dialog service system platform applied to the network consumption platform in the method is exemplified, and the to-be-processed dialog may be a dialog input by a client in the dialog service system platform.
In the embodiment of the present disclosure, the number of the target categories may be one, or may also be multiple, and the present disclosure does not limit this.
In the embodiment of the present disclosure, a dialog to be processed may be classified, and a target category to which the dialog to be processed belongs may be determined, where the target category is used to indicate a target product related to the dialog to be processed, and/or indicate a specific domain direction (denoted as a target domain direction in the present disclosure) under the target product related to the dialog to be processed.
The method is applied to a network consumption platform, the network consumption platform is an XX intelligent cloud for exemplary explanation, and if a target product related to a to-be-processed conversation is a data management platform product, target field directions can include field directions such as metadata collection, data development, data analysis, data operation, data service and data quality.
In the embodiment of the present disclosure, the candidate service may be an idle service, and/or the candidate service may also be a non-idle service (for example, a service with a relatively small number of currently allocated sessions), which is not limited by the present disclosure. For example, the candidate customer service may be a customer service that replied to historical sessions belonging to the target category.
In the embodiment of the present disclosure, for any one candidate customer service, the degree of adaptation between the candidate customer service and the target category may be determined according to the dialog replied by the candidate customer service history and/or the dialog currently assigned. For example, the more number of conversations replied by the candidate customer service history belongs to the target category, the greater the degree of adaptation between the candidate customer service and the target category, and for example, the less the number of conversations currently allocated by the candidate customer service, the greater the degree of adaptation between the candidate customer service and the target category.
And 103, determining the target customer service from the at least one candidate customer service according to the adaptation degree of the at least one candidate customer service.
In the embodiment of the present disclosure, the target customer service may be determined from each candidate customer service according to the degree of adaptation of each candidate customer service.
As an example, the candidate customer service with the highest suitability may be used as the target customer service.
As another example, a candidate customer service whose suitability is higher than a set threshold may be taken as a target customer service.
And 104, distributing the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service.
In the disclosed embodiment, the pending conversation may be assigned to the target customer service so that the pending conversation may be replied to by the target customer service.
The dialog processing method provided by the embodiment of the disclosure classifies dialogs to be processed to obtain a target category, wherein the target category is used for indicating a target product related to the dialog to be processed and/or a target field direction related to the target product; determining the adaptation degree of at least one candidate customer service and a target category, and determining the target customer service from the at least one candidate customer service according to the adaptation degree of the at least one candidate customer service; and allocating the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service. Therefore, the target customer service is selected from the customer services according to the degree of adaptation between the customer services and the target category, the target customer service is familiar with the target product related to the to-be-processed conversation and/or the target field direction related to the target product, the to-be-processed conversation is distributed to the target customer service, the situation that the customer service forwards and retreats the conversation due to the fact that the customer service is not familiar with the product related to the conversation and/or the subdivision field direction related to the product can be avoided, the processing efficiency and the processing quality of the conversation are improved, and the use experience of a user is improved.
In the technical scheme of the present disclosure, the processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user are all performed under the premise of obtaining the consent of the user, and all meet the regulations of the relevant laws and regulations without violating the customs of the public order.
In order to clearly illustrate how the above embodiments of the present disclosure determine the degree of adaptation between each candidate customer service and the target category, the present disclosure further provides a dialog processing method.
Fig. 2 is a flowchart illustrating a dialog processing method according to a second embodiment of the disclosure.
As shown in fig. 2, the dialog processing method may include the steps of:
For the explanation of step 201, reference may be made to relevant descriptions in any embodiment of the present disclosure, and details are not described herein.
In the embodiment of the present disclosure, for any one candidate customer service, a first sub-suitability of the candidate customer service with the target category may be determined according to the session processed by the candidate customer service history. And the first sub-adaptability is used for representing the professional processing capacity of the candidate customer service on the historical conversation belonging to the target category.
And step 203, determining a second sub-adaptation degree of the candidate customer service and the target class, wherein the second sub-adaptation degree represents the processing efficiency of the candidate customer service on the historical conversation belonging to the target class.
In the embodiment of the present disclosure, a second sub-suitability of the candidate customer service with the target category may be determined according to the session processed by the candidate customer service history. And the second sub-adaptability is used for representing the processing efficiency of the candidate customer service on the historical conversation belonging to the target category.
And 204, determining a third sub-adaptation degree of the candidate customer service and the target category, wherein the third sub-adaptation degree represents the processing quality of the candidate customer service on the historical conversation belonging to the target category.
In the embodiment of the present disclosure, a third sub-suitability of the candidate customer service with the target category may be determined according to the session processed by the candidate customer service history. And the third sub-adaptability is used for representing the processing quality of the candidate customer service on the historical conversation belonging to the target category.
And step 205, determining the adaptation degree of the candidate customer service and the target category according to at least one of the first sub-adaptation degree, the second sub-adaptation degree and the third sub-adaptation degree.
In the embodiment of the disclosure, the degree of adaptation of the candidate customer service to the target category may be determined by at least one of the first sub-degree of adaptation, the second sub-degree of adaptation, and the third sub-degree of adaptation.
As a possible implementation manner, a maximum value of the first sub-suitability, the second sub-suitability, and the third sub-suitability may be used as the suitability of the candidate customer service to the target category.
As another possible implementation manner, the minimum value of the first sub-suitability, the second sub-suitability, and the third sub-suitability may be used as the suitability of the candidate customer service with the target category.
As another possible implementation manner, a median value of the first sub-suitability, the second sub-suitability, and the third sub-suitability may be used as the suitability of the candidate customer service to the target category.
As another possible implementation manner, the degree of adaptation between the candidate customer service and the target category may be determined according to multiple items of the first sub-degree of adaptation, the second sub-degree of adaptation, and the third sub-degree of adaptation.
As an example, the degree of adaptation of the candidate customer service to the target category may be determined according to the first sub-degree of adaptation and the second sub-degree of adaptation. For example, the first sub-fitness and the second sub-fitness may be weighted and summed to obtain the fitness of the candidate customer service to the target category.
As another example, the degree of adaptation of the candidate customer service to the target category may be determined according to the first sub-degree of adaptation and the third sub-degree of adaptation. For example, the first sub-suitability and the third sub-suitability may be weighted and summed to obtain the suitability of the candidate customer service to the target category.
As yet another example, the degree of adaptation of the candidate customer service to the target category may be determined according to the second sub-degree of adaptation and the third sub-degree of adaptation. For example, the second sub-suitability and the third sub-suitability may be weighted and summed to obtain the suitability of the candidate customer service to the target category.
As a further example, the degree of adaptation of the candidate customer service to the target category may be determined according to the first sub-degree of adaptation, the second sub-degree of adaptation, and the third sub-degree of adaptation.
For example, the first sub-fitness, the second sub-fitness and the third sub-fitness may be weighted and summed to obtain the fitness of the candidate customer service to the target category.
For another example, a first value may be determined according to the first sub-suitability, where the first value is in a positive correlation with the first sub-suitability, a second value is determined according to the second sub-suitability, where the second value is in a positive correlation with the second sub-suitability, and a third value is determined according to the third sub-suitability, where the third value is in a positive correlation with the third sub-suitability, so that the first value, the second value, and the third value may be subjected to weighted summation to obtain the suitability of the candidate customer service and the target category.
The weight corresponding to the first value, the weight corresponding to the second value, and the weight corresponding to the third value may be preset, or, to improve the flexibility and applicability of the method, the weight corresponding to the first value, the weight corresponding to the second value, and the weight corresponding to the third value may also be determined according to the target category, that is, a plurality of weights adapted to the target category may be obtained, and the weight corresponding to the first value, the weight corresponding to the second value, and the weight corresponding to the third value are determined according to the plurality of weights.
For example, for a core and strong-branded fist product, the recovery quality may be set as the first significance, and thus, the weight of the third value may be set to be relatively large, and the weight of the first value and/or the weight of the second value may be set to be relatively small, for example, for a high-quality and high-guest weight protection product, in order to ensure that the problem of the key customer is handled properly, the professional business level may be set as the first significance, and thus, the weight of the first value may be set to be relatively large, the weight of the second value and/or the weight of the third value may be set to be relatively small, for example, for a non-fist product or a non-weight protection product, a quick response may be set as the first significance, and thus, the weight of the second value may be set to be relatively large, and the weight of the first value and/or the weight of the third value may be set to be relatively small.
For example, if the mark adaptation degree is P, the adaptation degree P can be calculated by the following formula:
P＝C 7 *g 1 (f 1 )+C 8 *g 2 (f 2 )+C 9 *g 3 (f 3 )；
wherein, f 1 Is referred to as the first sub-adaptation, function g 1 The value of (a) is in positive correlation with the value of the independent variable, g 1 (f 1 ) Is referred to as the first value. In the same way, f 2 Is the second sub-adaptation, function g 2 The value of (a) is in positive correlation with the value of the independent variable, g 2 (f 2 ) It is meant that the second value is,f 3 is referred to as the third sub-adaptation degree, function g 3 The value of (a) is in positive correlation with the value of the independent variable, g 3 (f 3 ) Is referred to as the third value. C 7 Is the weight corresponding to the first sub-adaptation degree or the first value, C 8 Is the weight corresponding to the second sub-adaptation degree or the second value, C 9 The third sub-adaptation degree or the weight corresponding to the third value is referred to.
Wherein the function g 1 、g 2 、g 3 May be in the form of a numerical value, base, logarithm, exponent, etc., and the disclosure is not limited thereto.
Therefore, the method can determine the adaptation degree of the candidate customer service and the target category according to different modes, and can improve the flexibility and the applicability of the method.
In addition, the weight corresponding to each sub-adaptation degree or value is determined according to the target category to which the dialog to be processed belongs, so that the weight corresponding to each sub-adaptation degree can be determined according to the category to which the dialog belongs, and the reliability of the adaptation degree calculation result is improved.
In any embodiment of the present disclosure, when the number of the target categories is multiple, the degree of adaptation of the candidate customer service corresponding to each target category may be determined in the above manner, and the degree of adaptation of the candidate customer service is determined according to the degree of adaptation of each target category.
For example, the average of the suitability of each target category may be used as the suitability of the candidate customer service.
For another example, the maximum value, the minimum value, or the median of the suitability of each target category may be used as the suitability of the candidate customer service.
For another example, the suitability degrees of the target categories may be sorted from large to small, and the average of the previously sorted set number of suitability degrees may be used as the suitability degree of the candidate customer service.
For another example, the target category with the degree of adaptation greater than the set degree of adaptation threshold may be selected from the target categories, and the average of the degrees of adaptation of the selected target categories may be used as the degree of adaptation of the candidate customer service.
And step 206, determining the target customer service from the at least one candidate customer service according to the adaptability of the at least one candidate customer service.
For the explanation of steps 206 to 207, reference may be made to the related description in any embodiment of the present disclosure, which is not described herein again.
The dialogue processing method disclosed by the embodiment of the disclosure can determine the adaptation degree of the candidate customer service and the target category based on various factors, and can improve the reliability of the adaptation degree determination result, so that the customer service is selected according to the reliable adaptation degree, and the optimized allocation of the dialogue can be realized.
In order to clearly illustrate how the first sub-suitability of each candidate customer service and the target category is determined in the above embodiments of the present disclosure, the present disclosure further provides a dialog processing method.
Fig. 3 is a flowchart illustrating a dialog processing method according to a third embodiment of the present disclosure.
As shown in fig. 3, the dialog processing method may include the steps of:
The target category is used for indicating a target product related to the to-be-processed dialog and/or a target field direction related to the target product.
For the explanation of step 301, reference may be made to relevant descriptions in any embodiment of the present disclosure, and details are not described herein.
In the embodiment of the present disclosure, the setting level is a preset level or level, for example, the setting level may include: for example, the set level may include: first-level, second-level, third-level, fourth-level, etc., wherein the levels from high to low are respectively: first, second, third, fourth, etc. It should be noted that, the present disclosure does not limit the partition granularity of the hierarchy, for example, the number of the set hierarchies may be 3, 4, 5, and the like, and in practical application, the partition granularity of the hierarchy may be set according to the requirement of practical application, and the hierarchy may be partitioned according to the partition granularity.
In the embodiment of the disclosure, for any one candidate customer service, a target level where the candidate customer service is located may be determined from a plurality of set levels. For example, the target tier at which the candidate customer service is located may be high.
Step 303 determines that a first number of first historical sessions belonging to the objective category have been assigned to the candidate customer service.
In the embodiment of the present disclosure, a first history session belonging to the target category may be determined from history dialogs assigned to the candidate customer service, and the number of the first history session is counted, which is denoted as the first number in the present disclosure.
In the embodiment of the present disclosure, a second history session may be determined from each first history session, where the second history session may be a session that the candidate customer service cannot reply to and transfer to other customer services.
The first sub-suitability degree represents professional processing capacity of the candidate customer service on the historical conversation belonging to the target category.
In an embodiment of the present disclosure, the number of the second historical sessions, denoted as the second number in the present disclosure, may be counted, and the ratio of the second number to the first number may be determined, for example, the first number is denoted as N 1 The second number is N 2 The ratio of the second number to the first number is: n is a radical of hydrogen 2 /N 1 。
In the embodiment of the present disclosure, the first sub-adaptation degree may be determined according to a ratio of the second number to the first number and a target level.
The higher the ratio of the second number to the first number is, the higher the conversation transfer rate of the candidate customer service is, the lower the professional processing capacity or professional service level of the candidate customer service on the historical conversation belonging to the target category is, the lower the first sub-adaptation degree is, and conversely, the lower the ratio of the second number to the first number is, the lower the conversation transfer rate of the candidate customer service is, the higher the professional processing capacity or professional service level of the candidate customer service on the historical conversation belonging to the target category is, the higher the first sub-adaptation degree is.
The first sub-adaptation degree and the target level are in a positive correlation relationship, namely the higher the target level is, the higher the first sub-adaptation degree is, and conversely, the lower the target level is, the lower the first sub-adaptation degree is.
In a possible implementation manner of the embodiment of the present disclosure, the calculation manner of the first sub-adaptation degree may be: a third history session belonging to the target category and being the first reply to the candidate customer service can be determined from the first history sessions, a first difference (such as a difference value, an absolute value of the difference value, and the like) between the reply time of the third history session and the current time can be determined, and a score matched with the target level can be queried according to the target level, so that a first sub-suitability can be determined according to the score, the first difference, and the ratio of the second number to the first number.
As an example, the first sub-adaptation degree f 1 Can be determined according to the following formula:
f 1 ＝level-C 1 *rate+C 4 *year 1 ；
wherein, level refers to a score determined according to a target level where the candidate customer service is located, wherein the higher the target level is, the larger the level value is, otherwise, the lower the target level is, the smaller the level value is, and rate refers to a conversation transfer rate, year determined according to the ratio of the second number to the first number 1 The number of years between the reply time of the history session belonging to the target category and the current time of the first reply of the candidate customer service is referred to, namely the number of years from the reply time of the third history session to the current time。C 1 And C 4 Is a constant.
Therefore, the professional processing capacity of the candidate customer service on the historical conversation belonging to the target category can be determined based on the time element, namely the time length from the first reply of the candidate customer service to the conversation belonging to the target category to the current time, the growth factor of the candidate customer service can be considered, and the reliability of the first sub-fitness calculation result can be improved.
And step 306, determining a second sub-adaptation degree of the candidate customer service and the target category, wherein the second sub-adaptation degree represents the processing efficiency of the candidate customer service on the historical conversation belonging to the target category.
And 307, determining a third sub-adaptation degree of the candidate customer service and the target class, wherein the third sub-adaptation degree represents the processing quality of the candidate customer service on the historical conversation belonging to the target class.
And 308, determining the adaptation degree of the candidate customer service and the target class according to at least one of the first sub-adaptation degree, the second sub-adaptation degree and the third sub-adaptation degree.
And 310, distributing the to-be-processed dialog to the target customer service so as to reply the to-be-processed dialog through the target customer service.
For the explanation of steps 306 to 310, reference may be made to the related description in any embodiment of the present disclosure, which is not described herein again.
The conversation processing method of the embodiment of the disclosure can realize the comprehensive target level of the candidate customer service and the transfer rate of the candidate customer service to the historical conversation belonging to the target category to determine the professional processing capacity of the candidate customer service to the historical conversation belonging to the target category, and can improve the accuracy and reliability of the determination result.
In order to clearly illustrate how to determine the second sub-suitability of each candidate customer service and the target class in any embodiment of the disclosure, the disclosure further provides a dialogue processing method.
Fig. 4 is a flowchart illustrating a dialog processing method according to a fourth embodiment of the present disclosure.
As shown in fig. 4, the dialog processing method may include the steps of:
The target category is used for indicating a target product related to the to-be-processed dialogue and/or a target field direction related to the target product.
The first sub-suitability degree represents professional processing capacity of the candidate customer service on the historical conversation belonging to the target category.
For explanation of steps 401 to 402, reference may be made to relevant descriptions in any embodiment of the present disclosure, and details are not described herein.
In step 403, an average reply time length of each fourth historical conversation replied by the candidate customer service within the set time length is determined.
In the embodiment of the present disclosure, the set time duration is a preset time duration, for example, the set time duration may be one day, one week, and the like.
In the embodiment of the present disclosure, for any one candidate customer service, an average reply time length of each fourth history session that the candidate customer service replies within a set time length may be determined. For example, taking the set time length as one day for example, the fourth historical sessions that the candidate customer service replies in a certain day may be determined, and the reply time length of each fourth historical session may be determined, so that the average value of the reply time lengths of the fourth historical sessions may be obtained, and the average reply time length of each dialog may be obtained.
At step 404, a third number of sessions to be replied to that are allocated and not replied to by the candidate service is determined.
In the embodiment of the present disclosure, the candidate customer service may be determined to currently allocate and not reply to the dialog, which is recorded as the dialog to be replied in the present disclosure, and the number of the dialogs to be replied is counted, which is recorded as the third number in the present disclosure.
In the embodiment of the present disclosure, the reply waiting duration of each to-be-replied session may be estimated according to the third number, where the reply waiting duration and the third number are in a positive correlation, that is, the larger the third number is, the longer the reply waiting duration is, and conversely, the smaller the third number is, the shorter the reply waiting duration is.
And 406, determining a second sub-adaptation degree of the candidate customer service and the target class according to the average reply duration and the waiting reply duration.
And the second sub-suitability characterizes the processing efficiency of the candidate customer service on the historical conversation belonging to the target category.
In the embodiment of the present disclosure, the second sub-adaptation degree may be determined according to the average reply duration and the wait reply duration.
The second sub-adaptation degree and the average reply duration are in a negative correlation relationship, that is, the longer the average reply duration is, the lower the second sub-adaptation degree is, otherwise, the shorter the average reply duration is, the higher the second sub-adaptation degree is.
The second sub-adaptation degree and the reply waiting time length are in a negative correlation relationship, that is, the longer the reply waiting time length is, the lower the second sub-adaptation degree is, otherwise, the shorter the reply waiting time length is, the higher the second sub-adaptation degree is.
In a possible implementation manner of the embodiment of the present disclosure, the calculation manner of the second sub-adaptation degree may be: a sixth historical session belonging to the target category and being the first reply to the candidate customer service may be determined from the fifth historical session belonging to the target category and being the first reply to the candidate customer service, and a second difference (such as a difference value, an absolute value of the difference value, and the like) between the reply time of the sixth historical session and the current time may be determined, so that the second sub-adaptation degree may be determined according to the average reply time length, the reply waiting time length, and the second difference.
As an example, the second sub-adaptation degree f 2 Can be determined according to the following formula:
f 2 ＝1/time 1 +C 2 /time 2 +C 5 *year 2 ；
wherein, time 1 Means averaging backDuration of recovery, time 2 Means waiting for a recovery time, year 2 The number of years between the reply time of the history session belonging to the target category and the current time of the first reply of the candidate customer service is referred to, namely the number of years from the reply time of the sixth history session to the current time. C 2 And C 5 Is a constant.
Therefore, the processing efficiency of the candidate customer service on the historical conversation belonging to the target class can be determined based on the time element, namely the time length from the first reply of the candidate customer service to the conversation belonging to the target class to the current time, and the reliability of the second sub-fitness calculation result can be improved by considering the growth factor of the candidate customer service.
And the third sub-suitability characterizes the processing quality of the candidate customer service on the historical conversation belonging to the target category.
And step 408, determining the adaptation degree of the candidate customer service and the target category according to at least one of the first sub-adaptation degree, the second sub-adaptation degree and the third sub-adaptation degree.
And step 409, determining the target customer service from the at least one candidate customer service according to the adaptability of the at least one candidate customer service.
And step 410, distributing the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service.
For the explanation of steps 407 to 410, reference may be made to the related description in any embodiment of the present disclosure, which is not described herein again.
The dialog processing method of the embodiment of the disclosure can realize the average reply duration of the candidate customer service to each dialog and the reply waiting duration of the dialog to be replied, so as to determine the processing efficiency of the candidate customer service to the historical dialog belonging to the target category, and improve the accuracy and reliability of the determination result.
In order to clearly illustrate how the third sub-suitability of each candidate customer service and the target category is determined in the above embodiments of the present disclosure, the present disclosure further provides a dialog processing method.
Fig. 5 is a flowchart illustrating a dialog processing method according to a fifth embodiment of the present disclosure.
As shown in fig. 5, the dialog processing method may include the steps of:
The target category is used for indicating a target product related to the to-be-processed dialogue and/or a target field direction related to the target product.
The first sub-suitability characterizes professional processing capacity of the candidate customer service on the historical conversation belonging to the target category.
And the second sub-suitability characterizes the processing efficiency of the candidate customer service on the historical conversation belonging to the target category.
For explanation of steps 501 to 503, reference may be made to relevant descriptions in any embodiment of the present disclosure, and details are not repeated herein.
And step 504, obtaining an evaluation score of the seventh history session belonging to the target category and replied by the candidate customer service, wherein the evaluation score is used for indicating the processing quality of the seventh history session by the candidate customer service.
In the embodiment of the present disclosure, for any one candidate customer service, a seventh history session belonging to the target category and replied by the candidate customer service may be determined, and an evaluation score of each seventh history session is obtained, where the evaluation score is used to indicate the processing quality of the seventh history session by the candidate customer service.
For example, after a customer service replies to a conversation, the customer may score or rate the quality of the customer service's processing to obtain a rating score for the conversation.
And step 505, determining an eighth history session of complaints from the seventh history sessions.
In the disclosed embodiment, an eighth historical session that is complained by the customer may be determined from the seventh historical sessions.
And the third sub-suitability characterizes the processing quality of the candidate customer service on the historical dialogue belonging to the target category.
In the disclosed embodiment, the number of the seventh history session, which is denoted as the fifth number in the present disclosure, and the number of the eighth history session, which is denoted as the fourth number in the present disclosure, may be counted, and the ratio of the fourth number and the fifth number may be determined, for example, the fourth number is marked as N 4 The fifth number is N 5 The ratio of the fourth number to the fifth number is: n is a radical of 4 /N 5 。
In the embodiment of the disclosure, the third sub-suitability of the candidate customer service and the target category may be determined according to the ratio of the fourth number and the fifth number and the evaluation score of each seventh historical session.
As an example, a mean value of the evaluation scores of the seventh historical sessions may be counted, and a third sub-suitability of the candidate customer service and the target category may be determined according to a ratio of the fourth number and the fifth number and the mean value.
And the third sub-adaptation degree is in a negative correlation relation with the ratio of the fourth number to the fifth number, and the third sub-adaptation degree is in a positive correlation relation with the average value.
In a possible implementation manner of the embodiment of the present disclosure, a calculation manner of the third sub-adaptation degree may be: the ninth history session belonging to the target category and returned first by the candidate customer service may be determined from the seventh history sessions, and a third difference (for example, a difference value, an absolute value of the difference value, or the like) between the return time of the ninth history session and the current time may be determined, and an average value of the evaluation scores of the seventh history sessions may be determined, so that the third sub-suitability may be determined according to the third difference, the average value, and a ratio of the fourth number to the fifth number.
As an example, the third sub-adaptation degree f 3 Can be based onThe following equation determines:
f 3 ＝assess-C 3 *compliant+C 6 *year 3 ；
wherein, asset is the average of the evaluation scores of the seventh historical sessions, and complant is the conversation complaint proportion determined according to the ratio of the fourth number to the fifth number, year 3 The number of years between the reply time of the history session belonging to the target category and the current time of the first reply of the candidate customer service is referred to, namely the number of years from the reply time of the ninth history session to the current time. C 3 And C 6 Is a constant.
Therefore, the processing quality of the candidate customer service on the historical conversation belonging to the target class can be determined based on the time element, namely the time length from the first reply of the candidate customer service to the conversation belonging to the target class to the current time, and the reliability of the third sub-fitness calculation result can be improved by considering the growth factor of the candidate customer service.
And step 507, determining the adaptation degree of the candidate customer service and the target category according to at least one of the first sub-adaptation degree, the second sub-adaptation degree and the third sub-adaptation degree.
And step 508, determining the target customer service from the at least one candidate customer service according to the adaptation degree of the at least one candidate customer service.
For the explanation of steps 507 to 509, reference may be made to the related description in any embodiment of the present disclosure, and details are not repeated herein.
The dialog processing method provided by the embodiment of the disclosure can be used for determining the processing quality of the candidate customer service on the historical dialog belonging to the target category by integrating the evaluation score of the historical dialog belonging to the target category and the complaint proportion of the historical dialog belonging to the target category and replied by the candidate customer service, and can improve the reliability and accuracy of the determination result.
In order to clearly illustrate how to determine the target category to which the dialog to be processed belongs in any embodiment of the disclosure, the disclosure also provides a dialog processing method.
Fig. 6 is a schematic flowchart of a dialog processing method according to a sixth embodiment of the present disclosure.
As shown in fig. 6, the dialog processing method may include the steps of:
For the explanation of step 601, reference may be made to the related description in any embodiment of the present disclosure, which is not described herein again.
In the embodiment of the present disclosure, the target file may include files such as a picture, a PDF (Portable Document Format), and a video.
In the embodiment of the present disclosure, the set format is a preset format, for example, when the target document is a picture, the set format may be an image format such as png (Portable Network Graphics), jpg (Joint Photographic Experts Group), and the like.
In the embodiment of the present disclosure, in a case that the target file is included in the to-be-processed dialog, the target file may be subjected to Optical Character Recognition based on an OCR (Optical Character Recognition) technology to obtain a Character Recognition result.
And step 604, classifying the dialog text information and the character recognition result except the target file in the dialog to be processed by adopting the target classification model to obtain a target class.
The target category is used for indicating a target product related to the to-be-processed dialogue and/or a target field direction related to the target product. It should be noted that the explanation of the target category in the foregoing embodiment is also applicable to this embodiment, and details are not described herein.
Wherein the target classification model has learned the correspondence between the text and the category.
In the embodiment of the present disclosure, a trained target classification model may be used to classify the dialog text information and the character recognition result except for the target file in the dialog to be processed, so as to obtain the target class.
As an example, the training manner of the target classification model may be: the method comprises the steps of obtaining a sample conversation, generating training text information according to conversation text information in the sample conversation and a character recognition result corresponding to target content, carrying out class marking on the training text information to obtain a marked class, classifying the training text information by adopting an initial classification model to obtain a prediction class, and training the initial classification model according to the difference between the prediction class and the marked class to obtain a target classification model.
For example, a loss value may be generated according to a difference between the prediction category and the labeling category, and model parameters in the initial classification model may be adjusted according to the loss value to minimize the loss value.
Wherein the loss value is in positive correlation with the difference, namely the loss value is smaller when the difference is smaller, and the loss value is larger when the difference is larger.
It should be noted that, the above example is performed by only taking the termination condition of the model training as the minimization of the loss value, and when the method is actually applied, other termination conditions may also be set, for example, the number of times of training reaches the set number, the length of time of training reaches the set length, the loss value converges, and the like, which is not limited by the disclosure.
In the embodiment of the present disclosure, when the to-be-processed dialog does not include the target file, the to-be-processed dialog may be classified by using the target classification model to obtain the target class.
It should be noted that, in practical applications, the to-be-processed dialog may further include audio information, and the audio information may also be subjected to speech recognition based on a speech recognition technology to obtain audio text information, so that, in the case where the to-be-processed dialog further includes a target file, the audio text information, the dialog text information, and the character recognition result may be classified by using the target classification model to obtain a target class. And under the condition that the target file is not included in the dialog to be processed, the audio text information and the dialog text information can be classified by adopting a target classification model so as to obtain a target class.
It should be noted that steps 603-604 and step 605 are implemented in parallel, and only one of them needs to be executed in actual application.
For the explanation of steps 606 to 608, reference may be made to the related description in any embodiment of the present disclosure, which is not described herein again.
The dialog processing method of the embodiment of the disclosure can realize that the character information in the target content is recognized based on the OCR technology to obtain the character recognition result under the condition that the dialog contains the target content, so that the category of the dialog is recognized based on the character recognition result and the dialog text information, and the accuracy of the recognition result can be improved. In addition, the dialogs to be processed are classified based on the deep learning technology, the target classes to which the dialogs to be processed belong are obtained, and the accuracy of classification results can be improved.
In any embodiment of the present disclosure, intelligent allocation of conversations may be achieved by the following steps:
in a first step, when target content (such as a picture and/or a video) is included in a conversation, the target content may be recognized based on OCR technology in the field of artificial intelligence to obtain a character recognition result, the target content is taken as an example of a picture, and the picture may be subjected to optical character recognition based on OCR technology to obtain a character recognition result (i.e., picture information). And, the character recognition result can be used as the input of semantic corpus.
In order to realize intelligent and optimal allocation of most suitable customer service to the conversation, the picture information in the conversation can be processed to assist the allocation decision of the conversation intelligence, especially in the current conversation scene of the online trading platform, under the background that the picture information has higher and higher proportion. The OCR technology in artificial intelligence is introduced into a conversational customer service system platform, and the problems that picture information must be manually identified and cannot be used for intelligent conversational distribution can be solved.
Currently, each mainstream cloud platform manufacturer opens an Application Programming Interface (API) of the OCR technology, and can call the API (i.e., a character recognition function Interface) every time to automatically recognize picture information in a conversation. For example, accessing the OCR API capability of the cloud platform to the conversational customer service system platform may be implemented by the OCR recognition system architecture and flow as shown in fig. 7.
The character recognition result (i.e. picture character recognition result) obtained based on the OCR API recognition can be used for displaying in the dialog system interface, and can also play an important role in the subsequent dialog distribution system. The text information in the picture is displayed in a dialogue system in a text form, so that the dialogue system has great convenience, for example, aiming at a scene of electronic product maintenance and consumption, if the dialogue includes a picture of product abnormity sent by a client, at the moment, product information such as the version, the model and the date of delivery of the product can be automatically identified through an OCR technology. With the continuous development of the OCR technology, the accuracy of OCR recognition results is higher and higher, and a customer service can quickly match corresponding maintenance schemes or search in a knowledge platform conveniently according to corresponding product information according to accurate OCR recognition results, so that the finishing efficiency of conversation is improved.
In fig. 7, before OCR recognition is performed on a picture, the format, pixels, and the like of the picture may be verified, and only when the verification is passed, OCR recognition is performed on the picture. That is, different OCR APIs only support processing pictures in a set format and/or with pixels within a set range.
And secondly, classifying the dialogue text information and the character recognition result based on a multi-classification training algorithm in a deep learning technology in a machine learning technology to obtain a target class (or called as a keyword, wherein the keyword is used for indicating products related to the dialogue and the field direction related to the products, and is subsequently referred to as products & directions for short) which can be recognized by the system.
The conversation distribution between the clients and the customer service is carried out, the relation between the client consultation problem and the professional direction of the customer service needs to be established, and the adaptability of the conversation distribution can be analyzed by a keyword-based model method. The intelligent allocation of the conversation can be divided into the following two core steps to process: 1. analyzing keywords from dialogue information input by a client; 2. based on the keywords, the most suitable customer service is found.
Products of the network consumption platform are limited, and the direction of the product subdivision field is also limited, which is a classic multi-classification problem in the field of machine learning in the artificial intelligence technology. A conversational customer service system platform can generate new conversational consultation of a certain magnitude (for example, tens of millions or more) every day, and the consultation expression modes of customers aiming at the same product in the same field direction are often changeable, so that the language and the rules of the specific field direction are difficult to intuitively grope.
Based on text information (denoted as dialog text information in the present disclosure) and picture information (denoted as character recognition result in the present disclosure) in a dialog, a corresponding product & direction, such as a data quality direction of a data governance product, is predicted through an algorithm model (denoted as a target classification model in the present disclosure). The training process and the classification prediction process of the target classification model may be as shown in fig. 8. Obviously, it is common for more than one person familiar with each field to be present, and the degree of adaptation analysis is required for the specific service to which it is assigned.
The above 1 st core step mainly solves which group of customer services the new session should be assigned to (i.e. customer services familiar with product & direction), and the 2 nd core step solves which optimal customer service the new session should be assigned to.
And thirdly, training a comprehensive capacity user portrait model of each customer service based on a multi-dimensional data analysis method.
The model is used for accurately expressing the adaptation degree of each customer service to a specific keyword. The dialogue type customer service system platform stores dialogue information with an ultra-large data magnitude every day, each customer service can generate hundreds of dialogue information contents every day, the dialogue information is increased explosively as time goes on, and the dialogue information can be used for training a customer service comprehensive capability user portrait model and provides an algorithm model basis for intelligent allocation of dialogue.
The present disclosure can depict a capability user profile model from three dimensions as shown in table 1, with specific elements analyzed from three levels, respectively. It should be noted that elements may be added for different applications.
TABLE 1 correspondence between dimensions and elements
Wherein f is 1 = level (customer service level) -C 1 *rate+C 4 *year 1 ；(1)
f 2 ＝1/time 1 +C 2 /time 2 +C 5 *year 2 ； (2)
f 3 ＝assess-C 3 *compliant+C 6 *year 3 ； (3)
Alternatively, to apply the capability user representation model, a numerical scaling constant may be appropriately valued. One way of reference is to calculate all the values of all customer services in the global domain and then take the average value, for example, the conversation transfer rate of each customer service is (r) 1 ,r 2 ,r 3 ,r 4 ,...,r n ) An average value S = avg (r 1, r2, r3, r 4.., rn) is obtained, and then 1/S is used as a parameter constant C corresponding to the session transfer rate 1 Other constants are similar in concept.
And calculating the sub-adaptation degree of each optional customer service in the current idle queue or the sequencing queue in three dimensions according to the keywords predicted by the target classification model in the second step.
And fourthly, determining an allocation reference based on a multi-factor model (professional, efficiency, quality and the like) algorithm, and selecting the most appropriate customer service personnel from the idle customer service personnel queue for conversation allocation to realize multi-objective optimal allocation.
And determining the final adaptation degree of each customer service based on the multi-factor decision information, and realizing the optimal distribution of the conversation. The overall idea of the corresponding dialog assignment is not consistent for different products, even different domain directions of the same product. For example, for some basic products (non-fist products, non-re-insurance products) of the network consumption platform, a quick response may be the first important point; for another example, for the most core and most brandy product (fist product) of the network consumption platform, the quality may be the first important, so as to ensure the continuous, comprehensive and good appreciation and high quality of the fist product of the consumption platform; for another example, for a high-quality and high-price product (a product with heavy insurance) of a network consumption platform, in order to ensure that the problem of the key customer is properly handled, a professional business level may be the first significance. Obviously, although different products have different emphasis points, different requirements are put forward on the overall strategy, and model parameters matched with keywords in the directions of various products and fields can be newly added in the conversational customer service platform system. For example, the model parameters may be as shown in Table 2.
TABLE 2 correspondence between dimensions and parameters
The final calculation formula of the fitness P may be:
P＝C 7 *g 1 (f 1 )+ C 8 *g 2 (f 2 )+ C 9 *g 3 (f 3 )； (4)
calculation of the suitability degree the core flow of calculation of the suitability degree may be as shown in fig. 9, after calculating the suitability degree for the new dialog, the new dialog may be assigned to the customer service with the highest suitability degree. In addition to the calculation of the user portrayal capability model and the fitness of each customer service in the idle customer service queue, each customer service in the non-idle customer service queue needs to be considered. For example, for the keywords of services and products with high importance to professional business level, it is likely that the high-level professional customer service will soon all have scheduled the conversation, and at this time, it is not possible to directly assign a new conversation. At this time, an assigned task queue may be maintained for the customer service in each docked session, but the task queue is not a constant one because the system can only estimate the specific consultation end time of the customer service for each session according to the average reply duration, but actually the specific consultation end time of each customer service for each session is completely uncertain, and the current customer service may become idle soon or may become idle for a long time.
For non-idle customer service, the tasks in the task queue need to be continuously refreshed through the real-time computing engine, and whether the tasks are still continuously allocated to the customer service or are allocated to idle customer service or are allocated to non-idle customer service with less tasks in other task queues is determined. Obviously, the occurrence of idle conditions can be reduced by increasing the importance of the value of the result of the expected waiting time duration (i.e. the reply waiting time duration) in the conversation processing efficiency, and the specific setting can be subject to the comprehensive consideration of the product.
In summary, the image information in the dialog is extracted based on the OCR technology in the artificial intelligence, the text semantics of the dialog and the keywords in the image corpus are extracted through the deep learning algorithm in the machine learning technology, the user portrait model of the dialog capability of the customer service is obtained through the multidimensional data analysis method, the allocation benchmark is determined through the multi-factor model (professional, efficiency, quality and the like) algorithm, the intelligent dialog allocation is finally realized, the optimal allocation under the target condition of the allocation benchmark is realized, and the contribution capacity of the network consumption platform can be improved.
Corresponding to the dialog processing method provided in the embodiment of fig. 1 to 6, the present disclosure also provides a dialog processing apparatus, and since the dialog processing apparatus provided in the embodiment of the present disclosure corresponds to the dialog processing method provided in the embodiment of fig. 1 to 6, the embodiment of the dialog processing method is also applicable to the dialog processing apparatus provided in the embodiment of the present disclosure, and will not be described in detail in the embodiment of the present disclosure.
Fig. 10 is a schematic structural diagram of a session processing apparatus according to a seventh embodiment of the disclosure.
As shown in fig. 10, the dialog processing device 1000 may include: an acquisition module 1001, a first determination module 1002, a second determination module 1003, a third determination module 1004, and an assignment module 1005.
The obtaining module 1001 is configured to obtain a dialog to be processed.
The first determining module 1002 is configured to determine a target category to which the to-be-processed dialog belongs, where the target category is used to indicate a target product related to the to-be-processed dialog and/or a target domain direction related to the target product.
A second determining module 1003, configured to determine a degree of adaptation between the at least one candidate customer service and the target category.
A third determining module 1004, configured to determine the target customer service from the at least one candidate customer service according to the fitness of the at least one candidate customer service.
A distributing module 1005, configured to distribute the pending conversation to the target customer service, so as to reply the pending conversation through the target customer service.
In a possible implementation manner of the embodiment of the present disclosure, the second determining module 1003 may include:
the first determining unit is used for determining a first sub-adaptation degree of the candidate customer service and the target class aiming at any candidate customer service in the at least one candidate customer service, wherein the first sub-adaptation degree represents the professional processing capacity of the candidate customer service on the historical conversation belonging to the target class.
And the second determining unit is used for determining a second sub-adaptation degree of the candidate customer service and the target class, wherein the second sub-adaptation degree represents the processing efficiency of the candidate customer service on the historical conversation belonging to the target class.
And the third determining unit is used for determining a third sub-adaptation degree of the candidate customer service and the target class, wherein the third sub-adaptation degree represents the processing quality of the candidate customer service on the historical conversation belonging to the target class.
And the fourth determining unit is used for determining the adaptation degree of the candidate customer service and the target class according to at least one of the first sub-adaptation degree, the second sub-adaptation degree and the third sub-adaptation degree.
In a possible implementation manner of the embodiment of the present disclosure, the first determining unit is configured to: determining a target level where the candidate customer service is located from a plurality of set levels; determining that a first number of first historical conversations belonging to a target category have been assigned to a candidate customer service; determining a second history session from the first history sessions, wherein the second history session is a session for transferring the candidate customer service to other customer services; and determining a first sub-adaptation degree according to the ratio of the second number to the first number of the second historical conversation and the target level.
In a possible implementation manner of the embodiment of the present disclosure, the first determining unit is configured to: determining a third history session from the first history sessions, wherein the third history session is a session which is first replied by the candidate customer service and belongs to the target category; inquiring the score matched with the target level according to the target level; determining a first difference between a reply time of the third history session and a current time; and determining a first sub-adaptation degree according to the score, the first difference and the ratio of the second number to the first number.
In a possible implementation manner of the embodiment of the present disclosure, the second determining unit is configured to determine an average reply duration of each fourth history session that the candidate customer service replies within a set duration; determining a third number of the candidate customer service-allocated and unanswered conversations to be replied; determining the reply waiting time length of each session to be replied according to the third number; and determining a second sub-adaptation degree according to the average reply duration and the reply waiting duration.
In a possible implementation manner of the embodiment of the present disclosure, the second determining unit is configured to: determining a sixth historical conversation of the first reply from the fifth historical conversations of the candidate customer service replies, which belong to the target category; determining a second difference between a reply time of the sixth historical session and a current time; and determining a second sub-adaptation degree according to the average reply duration, the reply waiting duration and the second difference.
In a possible implementation manner of the embodiment of the present disclosure, the third determining unit is configured to: obtaining an evaluation score of a seventh historical conversation of the candidate customer service reply, wherein the evaluation score is used for indicating the processing quality of the seventh historical conversation by the candidate customer service; determining an eighth historical session of complaints from the seventh historical sessions; and determining a third sub-suitability degree according to the ratio of the fourth number of the eighth historical conversations to the fifth number of the seventh historical conversations and the evaluation score of each seventh historical conversation.
In a possible implementation manner of the embodiment of the present disclosure, the third determining unit is configured to: determining a ninth historical session of a first reply from the seventh historical session; determining a third difference between a reply time of the ninth historical session and the current time; determining the average value of the evaluation scores of the seventh historical conversations; and determining a third sub-adaptation degree according to the third difference, the mean value and the ratio of the fourth number to the fifth number.
In a possible implementation manner of the embodiment of the present disclosure, the fourth determining unit is configured to: acquiring a plurality of weights adapted to the target category; determining a first value according to the first sub-adaptation degree, wherein the first value and the first sub-adaptation degree are in a positive correlation relationship; determining a second value according to the second sub-suitability, wherein the second value and the second sub-suitability are in a positive correlation relationship; determining a third value according to the third sub-adaptation degree, wherein the third value and the third sub-adaptation degree are in a positive correlation; and according to the multiple weights, carrying out weighted summation on the first value, the second value and the third value to obtain the degree of adaptation of the candidate customer service to the target category.
In a possible implementation manner of the embodiment of the present disclosure, the first determining module 1002 is configured to: identifying whether the dialog to be processed contains a target file with a set format or not; under the condition that the dialog to be processed contains a target file, carrying out optical character recognition on the target file to obtain a character recognition result; classifying the dialog text information and the character recognition result except the target file in the dialog to be processed by adopting a target classification model to obtain a target class; wherein the target classification model has learned the correspondence between the text and the category.
The dialog processing device of the embodiment of the disclosure classifies dialogs to be processed to obtain a target category, wherein the target category is used for indicating a target product related to the dialog to be processed and/or a target field direction related to the target product; determining the adaptation degree of at least one candidate customer service and a target category, and determining the target customer service from the at least one candidate customer service according to the adaptation degree of the at least one candidate customer service; and allocating the pending conversation to the target customer service so as to reply the pending conversation through the target customer service. Therefore, the target customer service is selected from the customer services according to the degree of adaptation between the customer services and the target category, the target customer service is familiar with the target product related to the to-be-processed conversation and/or the target field direction related to the target product, the to-be-processed conversation is distributed to the target customer service, the situation that the customer service forwards and retreats the conversation due to the fact that the customer service is not familiar with the product related to the conversation and/or the subdivision field direction related to the product can be avoided, the processing efficiency and the processing quality of the conversation are improved, and the use experience of a user is improved.
To implement the above embodiments, the present disclosure also provides an electronic device, which may include at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the dialog processing method according to any of the above embodiments of the present disclosure.
In order to implement the above embodiments, the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the dialog processing method proposed in any of the above embodiments of the present disclosure.
In order to implement the above embodiments, the present disclosure further provides a computer program product, which includes a computer program that, when executed by a processor, implements the dialog processing method proposed by any of the above embodiments of the present disclosure.
The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.
FIG. 11 shows a schematic block diagram of an example electronic device that may be used to implement embodiments of the present disclosure. The electronic device may include the server and the client in the above embodiments. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 11, the electronic device 1100 includes a computing unit 1101, which can perform various appropriate actions and processes in accordance with a computer program stored in a ROM (Read-Only Memory) 1102 or a computer program loaded from a storage unit 1108 into a RAM (Random Access Memory) 1103. In the RAM1103, various programs and data necessary for the operation of the electronic device 1100 may also be stored. The calculation unit 1101, the ROM 1102, and the RAM1103 are connected to each other by a bus 1104. An I/O (Input/Output) interface 1105 is also connected to the bus 1104.
A number of components in electronic device 1100 are connected to I/O interface 1105, including: an input unit 1106 such as a keyboard, mouse, or the like; an output unit 1107 such as various types of displays, speakers, and the like; a storage unit 1108 such as a magnetic disk, optical disk, or the like; and a communication unit 1109 such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 1109 allows the electronic device 1100 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.
The computing unit 1101 can be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of the computing Unit 1101 include, but are not limited to, a CPU (Central Processing Unit), a GPU (graphics Processing Unit), various dedicated AI (Artificial Intelligence) computing chips, various computing Units running machine learning model algorithms, a DSP (Digital Signal Processor), and any suitable Processor, controller, microcontroller, and the like. The calculation unit 1101 performs the respective methods and processes described above, such as the above-described dialogue processing method. For example, in some embodiments, the above-described dialog processing method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 1108. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 1100 via the ROM 1102 and/or the communication unit 1109. When the computer program is loaded into RAM1103 and executed by computing unit 1101, one or more steps of the dialog processing method described above may be performed. Alternatively, in other embodiments, the computing unit 1101 may be configured to perform the above-described dialog processing method by any other suitable means (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be realized in digital electronic circuitry, integrated circuitry, FPGAs (Field Programmable Gate arrays), ASICs (Application-Specific Integrated circuits), ASSPs (Application Specific Standard products), SOCs (System On Chip, system On a Chip), CPLDs (Complex Programmable Logic devices), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program code, when executed by the processor or controller, causes the functions/acts specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a RAM, a ROM, an EPROM (Electrically Programmable Read-Only-Memory) or flash Memory, an optical fiber, a CD-ROM (Compact Disc Read-Only-Memory), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a Display device (e.g., a CRT (Cathode Ray Tube) or LCD (Liquid Crystal Display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: LAN (Local Area Network), WAN (Wide Area Network), internet, and blockchain Network.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The Server may be a cloud Server, which is also called a cloud computing Server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of high management difficulty and weak service expansibility in a conventional physical host and a VPS (Virtual Private Server). The server may also be a server of a distributed system, or a server incorporating a blockchain.
It should be noted that artificial intelligence is a subject for studying a computer to simulate some human thinking process and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.), and has both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing, and the like; the artificial intelligence software technology mainly comprises a computer vision technology, a voice recognition technology, a natural language processing technology, machine learning/deep learning, a big data processing technology, a knowledge map technology and the like.
According to the technical scheme of the embodiment of the disclosure, a target category is obtained by classifying the dialog to be processed, wherein the target category is used for indicating a target product related to the dialog to be processed and/or a target field direction related to the target product; determining the degree of adaptation of at least one candidate customer service to the target category, and determining the target customer service from the at least one candidate customer service according to the degree of adaptation of the at least one candidate customer service; and allocating the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service. Therefore, the target customer service is selected from the customer services according to the degree of adaptation between the customer services and the target category, the target customer service is familiar with the target product related to the to-be-processed conversation and/or the target field direction related to the target product, the to-be-processed conversation is distributed to the target customer service, the situation that the customer service forwards and reprocesses the conversation due to the fact that the customer service is not familiar with the product related to the conversation and/or the subdivision field direction related to the product can be avoided, the processing efficiency and the processing quality of the conversation can be improved, and the use experience of a user can be improved.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel or sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions proposed in the present disclosure can be achieved.
The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.
Claims (23)
1. A method of dialog processing, the method comprising:
obtaining a dialog to be processed, and determining a target category to which the dialog to be processed belongs, wherein the target category is used for indicating a target product related to the dialog to be processed and/or a target field direction related to the target product;
determining the adaptation degree of at least one candidate customer service to the target category;
determining target customer service from the at least one candidate customer service according to the adaptability of the at least one candidate customer service;
and distributing the to-be-processed dialog to the target customer service so as to reply the to-be-processed dialog through the target customer service.
2. The method of claim 1, wherein said determining a degree of fit of at least one candidate customer service to the target category comprises:
for any candidate customer service in the at least one candidate customer service, determining a first sub-suitability of the candidate customer service with the target category, wherein the first sub-suitability characterizes professional processing capacity of the candidate customer service on historical sessions belonging to the target category;
determining a second sub-suitability of the candidate customer service and the target category, wherein the second sub-suitability characterizes the processing efficiency of the candidate customer service on the historical conversation belonging to the target category;
determining a third sub-suitability degree of the candidate customer service and the target class, wherein the third sub-suitability degree represents the processing quality of the candidate customer service on the historical conversation belonging to the target class;
and determining the adaptation degree of the candidate customer service and the target category according to at least one of the first sub-adaptation degree, the second sub-adaptation degree and the third sub-adaptation degree.
3. The method of claim 2, wherein said determining a first sub-fitness of said candidate customer service to said objective category comprises:
determining a target level where the candidate customer service is located from a plurality of set levels;
determining that a first number of first historical conversations belonging to the objective category have been assigned to the candidate customer service;
determining a second history session from each first history session, wherein the second history session is a session for transferring the candidate customer service to other customer services;
determining the first sub-suitability based on a ratio of a second number of the second historical sessions to the first number and the target level.
4. The method of claim 3, wherein the determining the first sub-suitability from a ratio of a second number of the second historical sessions to the first number, and the target level comprises:
determining a third history session from the first history sessions, wherein the third history session is a session which is replied by the candidate customer service for the first time and belongs to the target category;
inquiring a score matched with the target level according to the target level;
determining a first difference between a reply time of the third history session and a current time;
and determining the first sub-adaptation degree according to the score, the first difference and the ratio of the second number to the first number.
5. The method of claim 2, wherein said determining a second sub-fitness of the candidate customer service to the objective category comprises:
determining the average reply time length of each fourth historical conversation replied by the candidate customer service within the set time length;
determining a third number of the sessions to be replied, which are allocated by the candidate customer service and are not replied;
determining the reply waiting time of each session to be replied according to the third number;
and determining the second sub-adaptation degree according to the average reply duration and the reply waiting duration.
6. The method of claim 5, wherein the determining the second sub-adaptation degree according to the average reply duration and the reply waiting duration comprises:
determining a sixth historical session of a first reply from the fifth historical sessions of the candidate customer service replies which belong to the target category;
determining a second difference between a reply time of the sixth historical session and a current time;
and determining the second sub-adaptation degree according to the average reply duration, the reply waiting duration and the second difference.
7. The method of claim 2, wherein said determining a third sub-fitness of said candidate customer service to said objective category comprises:
obtaining an evaluation score of a seventh historical session belonging to the target category and returned by the candidate customer service, wherein the evaluation score is used for indicating the processing quality of the seventh historical session by the candidate customer service;
determining an eighth historical session of complaints from each of the seventh historical sessions;
and determining the third sub-suitability degree according to the ratio of the fourth number of the eighth historical conversations to the fifth number of the seventh historical conversations and the evaluation score of each seventh historical conversation.
8. The method of claim 6, wherein the determining the third sub-suitability based on a ratio of a fourth number of the eighth historical sessions and a fifth number of the seventh historical sessions, and a rating score for each of the seventh historical sessions comprises:
determining a ninth historical session of a first reply from the seventh historical session;
determining a third difference between a reply time of the ninth historical session and a current time;
determining an average of the evaluation scores of the seventh historical sessions;
determining the third sub-adaptation degree according to the third difference, the ratio of the mean value to the fourth number and the fifth number.
9. The method of claim 2, wherein said determining the degree of fit of the candidate customer service to the target category based on at least one of the first degree of sub-fit, the second degree of sub-fit, and the third degree of sub-fit comprises:
acquiring a plurality of weights adapted to the target category;
determining a first value according to the first sub-suitability, wherein the first value and the first sub-suitability are in a positive correlation relationship;
determining a second value according to the second sub-suitability, wherein the second value and the second sub-suitability are in a positive correlation;
determining a third value according to the third sub-adaptation degree, wherein the third value and the third sub-adaptation degree are in a positive correlation;
and according to the weights, carrying out weighted summation on the first value, the second value and the third value to obtain the degree of adaptation of the candidate customer service to the target category.
10. The method of any of claims 1-9, wherein the determining a target category to which the pending conversation belongs comprises:
identifying whether the dialog to be processed contains a target file with a set format or not;
under the condition that the target file is included in the dialog to be processed, carrying out optical character recognition on the target file to obtain a character recognition result;
classifying the dialog text information except the target file and the character recognition result in the dialog to be processed by adopting a target classification model to obtain the target category;
wherein the target classification model has learned correspondence between text and categories.
11. A conversation processing apparatus, the apparatus comprising:
the acquisition module is used for acquiring the dialog to be processed;
the first determination module is used for determining a target category to which the to-be-processed conversation belongs, wherein the target category is used for indicating a target product related to the to-be-processed conversation and/or a target field direction related to the target product;
the second determining module is used for determining the adaptation degree of at least one candidate customer service and the target category;
the third determining module is used for determining target customer service from the at least one candidate customer service according to the adaptation degree of the at least one candidate customer service;
and the distribution module is used for distributing the to-be-processed conversation to the target customer service so as to reply the to-be-processed conversation through the target customer service.
12. The apparatus of claim 11, wherein the second determining means comprises:
a first determining unit, configured to determine, for any candidate customer service in the at least one candidate customer service, a first sub-suitability of the candidate customer service with the target category, where the first sub-suitability characterizes professional processing capability of the candidate customer service on a historical conversation belonging to the target category;
a second determining unit, configured to determine a second sub-suitability degree of the candidate customer service and the target category, where the second sub-suitability degree represents a processing efficiency of the candidate customer service on a historical conversation belonging to the target category;
a third determining unit, configured to determine a third sub-suitability of the candidate customer service and the target category, where the third sub-suitability characterizes a processing quality of the candidate customer service on a historical dialog belonging to the target category;
a fourth determining unit, configured to determine a degree of adaptation between the candidate customer service and the target category according to at least one of the first sub-degree of adaptation, the second sub-degree of adaptation, and the third sub-degree of adaptation.
13. The apparatus of claim 12, wherein the first determining unit is configured to:
determining a target level where the candidate customer service is located from a plurality of set levels;
determining that a first number of first historical conversations belonging to the target category have been assigned to the candidate customer service;
determining a second history session from each first history session, wherein the second history session is a session for transferring the candidate customer service to other customer services;
determining the first sub-suitability according to a ratio of a second number of the second historical sessions to the first number and the target level.
14. The apparatus of claim 13, wherein the first determining unit is configured to:
determining a third history session from the first history sessions, wherein the third history session is a session which is replied by the candidate customer service for the first time and belongs to the target category;
inquiring a score matched with the target level according to the target level;
determining a first difference between a reply time of the third history session and a current time;
and determining the first sub-adaptation degree according to the score, the first difference and the ratio of the second number to the first number.
15. The apparatus of claim 12, wherein the second determining unit is configured to:
determining the average reply time length of each fourth historical conversation replied by the candidate customer service within the set time length;
determining a third number of the sessions to be replied, which are allocated by the candidate customer service and are not replied;
determining the reply waiting time of each session to be replied according to the third number;
and determining the second sub-adaptation degree according to the average reply duration and the reply waiting duration.
16. The apparatus of claim 15, wherein the second determining unit is configured to:
determining a sixth historical session of a first reply from the fifth historical sessions of the candidate customer service replies which belong to the target category;
determining a second difference between a reply time of the sixth historical session and a current time;
and determining the second sub-adaptation degree according to the average reply duration, the reply waiting duration and the second difference.
17. The apparatus of claim 12, wherein the third determining unit is configured to:
obtaining an evaluation score of a seventh historical session belonging to the target category and returned by the candidate customer service, wherein the evaluation score is used for indicating the processing quality of the seventh historical session by the candidate customer service;
determining an eighth historical session of complaints from each of the seventh historical sessions;
and determining the third sub-suitability degree according to the ratio of the fourth number of the eighth historical conversations to the fifth number of the seventh historical conversations and the evaluation score of each seventh historical conversation.
18. The apparatus of claim 16, wherein the third determining unit is configured to:
determining a ninth history session of a first reply from the seventh history session;
determining a third difference between a reply time of the ninth historical session and a current time;
determining an average of the evaluation scores of the seventh historical sessions;
determining the third sub-suitability based on the third difference, the mean, and a ratio of the fourth number and the fifth number.
19. The apparatus of claim 12, wherein the fourth determining unit is configured to:
acquiring a plurality of weights adapted to the target category;
determining a first value according to the first sub-adaptation degree, wherein the first value and the first sub-adaptation degree are in a positive correlation relationship;
determining a second value according to the second sub-adaptation degree, wherein the second value and the second sub-adaptation degree are in a positive correlation;
determining a third value according to the third sub-suitability, wherein the third value and the third sub-suitability are in a positive correlation;
and according to the weights, carrying out weighted summation on the first value, the second value and the third value to obtain the degree of adaptation of the candidate customer service to the target category.
20. The apparatus of any of claims 11-19, wherein the first determining means is to:
identifying whether the dialog to be processed contains a target file with a set format;
under the condition that the dialog to be processed contains the target file, carrying out optical character recognition on the target file to obtain a character recognition result;
classifying the dialog text information except the target file and the character recognition result in the dialog to be processed by adopting a target classification model to obtain the target category;
wherein the target classification model has learned correspondences between text and categories.
21. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein, the first and the second end of the pipe are connected with each other,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the dialog processing method of any of claims 1-10.
22. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the dialogue processing method according to any one of claims 1 to 10.
23. A computer program product comprising a computer program which, when executed by a processor, carries out the steps of the dialog processing method according to any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211379020.5A CN115660363A (en) | 2022-11-04 | 2022-11-04 | Dialogue processing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211379020.5A CN115660363A (en) | 2022-11-04 | 2022-11-04 | Dialogue processing method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115660363A true CN115660363A (en) | 2023-01-31 |
Family
ID=85016767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211379020.5A Pending CN115660363A (en) | 2022-11-04 | 2022-11-04 | Dialogue processing method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115660363A (en) |
-
2022
- 2022-11-04 CN CN202211379020.5A patent/CN115660363A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109710766B (en) | Complaint tendency analysis early warning method and device for work order data | |
| CN112560461A (en) | News clue generation method and device, electronic equipment and storage medium | |
| CN110826327A (en) | Emotion analysis method and device, computer readable medium and electronic equipment | |
| CN113392920A (en) | Method, apparatus, device, medium, and program product for generating cheating prediction model | |
| CN114037059A (en) | Pre-training model, model generation method, data processing method and data processing device | |
| JP7454630B2 (en) | Training method and device for label recommendation model, label acquisition method and device | |
| CN116468479A (en) | Method for determining page quality evaluation dimension, and page quality evaluation method and device | |
| CN113010782B (en) | Demand acquisition method, device, electronic equipment and computer readable medium | |
| CN114882388A (en) | Method, device, equipment and medium for training and predicting multitask model | |
| CN115660363A (en) | Dialogue processing method and device, electronic equipment and storage medium | |
| CN114240250A (en) | Intelligent management method and system for vocational evaluation | |
| CN114511022A (en) | Feature screening, behavior recognition model training and abnormal behavior recognition method and device | |
| CN113806541A (en) | Emotion classification method and emotion classification model training method and device | |
| CN112925913A (en) | Method, apparatus, device and computer-readable storage medium for matching data | |
| CN112862305A (en) | Method, device, equipment and storage medium for determining risk state of object | |
| CN115273854B (en) | Service quality determining method and device, electronic equipment and storage medium | |
| CN117131197B (en) | Method, device, equipment and storage medium for processing demand category of bidding document | |
| CN116340831B (en) | Information classification method and device, electronic equipment and storage medium | |
| KR20190104745A (en) | Issue interest based news value evaluation apparatus and method, storage media storing the same | |
| CN115795023B (en) | Document recommendation method, device, equipment and storage medium | |
| CN113656393B (en) | Data processing method, device, electronic equipment and storage medium | |
| CN113642495B (en) | Training method, apparatus, and program product for evaluating model for time series nomination | |
| CN118035445A (en) | Work order classification method and device, electronic equipment and storage medium | |
| CN117651167A (en) | Resource recommendation method, device, equipment and storage medium | |
| CN113886541A (en) | Demand evaluation information generation method, demand evaluation information display 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 |