Artificial agent distribution method and device and electronic equipment

Abstract

The invention relates to a manual seat distribution method and device and electronic equipment, and belongs to the technical field of intelligent customer service, and the distribution method comprises the steps: obtaining the seat score of each seat through a scoring algorithm according to the historical chat content, historical voice data and customer satisfaction evaluation of the previous day, obtaining a seat priority coefficient according to the seat score, and obtaining the seat priority coefficient according to the seat priority coefficient; carrying out call distribution according to the seat priority coefficient; according to the technical scheme, the service level of the seats is combined during seat allocation, and the customer experience can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent customer service technology, and in particular to a method, device and electronic equipment for allocating artificial seats. Background Art

[0002] A call center is an organization that handles customer calls in a centralized manner, and is responsible for answering, processing, and transferring customer calls. Initially, call center technology mainly relied on traditional telephone exchange systems. With the development of the Internet and cloud computing technologies, call center systems based on IP communications and cloud platforms have gradually become mainstream. Modern call centers usually integrate speech recognition (ASR), natural language processing (NLP), and text-to-speech (TTS) technologies to achieve intelligent customer service and automatic answering. Problems that cannot be solved by intelligent customer service and automatic answering are finally handed over to human agents for communication.

[0003] The existing method of allocating manual seats generally allocates seats based on the agent status. The agent status includes online and available to make or receive calls, online but unable to make or receive calls, on a call, and ringing. An agent whose status is online and available to make or receive calls can receive new calls for communication. The existing method of allocating manual seats is unable to accurately determine the service level of manual agents due to insufficient evaluation of manual agents, resulting in poor customer experience. Summary of the invention

[0004] To this end, the present invention provides a method, device and electronic device for allocating artificial seats to solve the problem of poor customer experience caused by the inability to accurately determine the service level of artificial agents in the prior art.

[0005] In a first aspect, a method for allocating artificial seats is provided, the method comprising: Acquire the agent data of each agent; the agent data includes the historical chat content, historical voice data and customer satisfaction evaluation of the previous day; Obtaining a seat score for each seat through a scoring algorithm based on the seat data; Obtaining an agent priority coefficient according to the agent score; Call allocation is performed according to the agent priority coefficient.

[0006] Furthermore, the seat score is obtained by a scoring formula according to the seat data, and the scoring formula is: ; Among them, Score 座席 Score is the seat score; 关键词 Score is the keyword matching score obtained based on the historical chat content; 情绪 Score is the emotion score obtained by analyzing historical voice data through artificial intelligence;流畅度 Score is the communication fluency score obtained based on historical voice data; 停顿 Score is the chat pause score obtained based on historical voice data; 满意度 The score for customer satisfaction evaluation.

[0007] Furthermore, the keyword matching score obtained based on the historical chat content includes: Obtaining the total number of keywords and the number of matching keywords after identifying the historical chat content; The keyword matching score is obtained according to the total number of keywords and the number of matching keywords through a matching formula, and the matching formula is: ; Among them, Score 关键词 is the keyword matching score; C 匹配 is the number of matching keywords; C 总数 is the total number of keywords.

[0008] Furthermore, the emotion score obtained by analyzing the historical voice data through artificial intelligence includes: Analyzing the historical voice data through artificial intelligence to obtain the emotional state of the agent; the emotional state includes anger, calmness and happiness; A corresponding preset score is used as the emotion score according to the emotion state.

[0009] Furthermore, the communication fluency score obtained based on the historical voice data includes: Analyze the historical voice data through artificial intelligence to obtain the actual speaking speed of the agent; The communication fluency score is obtained according to the actual speaking speed through the fluency formula, and the fluency formula is: ; Among them, Score 流畅度 Y is the communication fluency score; 实际 is the actual speaking speed; Y 理想 is the preset ideal speaking speed; k is a constant.

[0010] Furthermore, the chat pause score obtained based on the historical voice data includes: Analyze the historical voice data through artificial intelligence to obtain the number of pauses and interruptions of the agent; The chat pause score is obtained according to the number of pauses and the number of interruptions through a pause formula, and the pause formula is: ; Among them, Score 停顿 Score for chat pause; C停顿 is the number of pauses; C 中断 is the number of interruptions; l is a constant term.

[0011] Further, obtaining the seat priority coefficient according to the seat score includes: Sort the seats from high to low according to the seat scores to obtain a seat list; The seat priority coefficient of the agent is obtained according to the sequence number of the agent in the seat list.

[0012] Furthermore, the allocation algorithm includes: Obtain a target agent through an allocation formula, obtain the latest call and allocate the latest call to the target agent; The allocation formula is: ; Among them, P i The priority coefficient of agent i; M ij Indicates whether the skills of agent i match the current call type j; Q i Characterizes the call queue length of agent i; W i represents the average waiting time of agent i; C i Indicates the cumulative number of calls made by agent i; S i Characterizes the service level of agent i; α, β, γ, δ and λ are all constant terms.

[0013] In a second aspect, a device for allocating artificial seats is provided, the device comprising: An acquisition module is used to acquire the agent data of each agent; the agent data includes the historical chat content, historical voice data and customer satisfaction evaluation of the previous day; A scoring module, used to obtain a seat score for each seat through a scoring algorithm based on the seat data; A screening module, used for obtaining a seat priority coefficient according to the seat score; The allocation module is used to allocate calls according to the agent priority coefficient.

[0014] In a third aspect, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the above-mentioned methods for allocating manual seats.

[0015] The present invention adopts the above technical solution and has at least the following beneficial effects: Provided are a method, device and electronic device for allocating artificial seats. According to the historical chat content, historical voice data and customer satisfaction evaluation of the previous day, the seat score of each seat is obtained through a scoring algorithm, and the agent priority coefficient is obtained by executing an elimination and promotion strategy based on the agent score. Call allocation is performed based on the agent priority coefficient. This technical solution combines the service level of the agent when allocating seats, which can effectively improve customer experience.

[0016] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 is a flow chart of a method for allocating artificial seats shown in an exemplary embodiment of the present invention; Figure 2 is a schematic block diagram of a device for allocating artificial seats according to an exemplary embodiment of the present invention; Figure 3 It is a schematic block diagram of an electronic device shown in an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] The method and device in this application are described below through specific embodiments.

[0021] See also Figure 1 , Figure 1 is a flow chart of a method for allocating artificial seats according to an exemplary embodiment of the present invention, see Figure 1 , the method comprising: Step S11, obtaining the agent data of each agent; the agent data includes the historical chat content, historical voice data and customer satisfaction evaluation of the previous day; Step S12: Obtain the seat score of each seat through a scoring algorithm based on the seat data; Step S13, executing the elimination and promotion strategy according to the seat scores to obtain the seat priority coefficient; Step S14: Allocate calls based on the agent priority coefficient.

[0022] It should be noted that the technical solution provided in this embodiment can be added to the existing call center system in the form of a small program in specific practice, or it can also be in the form of an independent application to provide an interface to the outside world with an agent allocation function; applicable scenarios include but are not limited to: call center agent allocation.

[0023] It can be understood that the method provided in this embodiment obtains the agent score of each agent through a scoring algorithm based on the historical chat content, historical voice data and customer satisfaction evaluation of the previous day, executes the elimination and promotion strategy based on the agent score to obtain the agent priority coefficient, and performs call allocation based on the agent priority coefficient; this technical solution combines the service level of the agent when allocating seats, which can effectively improve customer experience.

[0024] In specific practice, the "scoring algorithm" in step S12 includes: obtaining the seat score through a scoring formula based on the seat data, and the scoring formula is: ; Among them, Score 座席 Score is the seat score; 关键词 Score is the keyword matching score obtained based on the historical chat content; 情绪 Score is the emotion score obtained by analyzing historical voice data through artificial intelligence; 流畅度 Score is the communication fluency score obtained based on historical voice data; 停顿 Score is the chat pause score obtained based on historical voice data; 满意度 The score for customer satisfaction evaluation.

[0025] It should be noted that the agent score is valid for the day, and calls for the day are allocated based on the agent score of the day.

[0026] In practice, agents can also be graded according to their scores. Specifically, after agents are ranked according to their scores, the top 20% of agents are graded as Level 2, the top 20%-50% of agents are graded as Level 1, and the top 50%-100% of agents are graded as Level 0. The higher the agent's grade, the higher the authority. High authority for low authority includes: monitoring calls of agents with low grades, inserting calls of agents with low grades, muting agents with low grades, hanging up agents with low grades, and hanging up the entire call. At the beginning, all agents are defaulted to Level 0. Agent grades are re-ranked every day.

[0027] Specifically, the keyword matching score obtained based on the historical chat content includes: obtaining the total number of keywords and the number of matching keywords after identification from the historical chat content; obtaining the keyword matching score based on the total number of keywords and the number of matching keywords through a matching formula, and the matching formula is: Among them, Score 关键词 is the keyword matching score; C 匹配 is the number of matching keywords; C 总数 is the total number of keywords.

[0028] It should be noted that the keywords in the keyword library are identified from the historical chat content, and the total number of keywords is counted; the matching keywords are keywords obtained through preset topics or customer needs, and the matching keywords are identified from the historical chat content and counted to obtain the number of matching keywords.

[0029] Specifically, the emotional score obtained by analyzing historical voice data through artificial intelligence includes: obtaining the emotional state of the agent by analyzing historical voice data through artificial intelligence; the emotional state includes anger, calmness and happiness; and using the corresponding preset score as the emotional score based on the emotional state.

[0030] It should be noted that the preset scores corresponding to anger, calmness and happiness are -1, 0 and 1 respectively, and the preset scores can be set according to specific business needs.

[0031] Specifically, the communication fluency score obtained based on the historical voice data includes: analyzing the historical voice data through artificial intelligence to obtain the actual speaking speed of the agent; and obtaining the communication fluency score based on the actual speaking speed through the fluency formula, and the fluency formula is: Among them, Score 流畅度 Y is the communication fluency score; 实际 is the actual speaking speed; Y 理想 is the preset ideal speaking speed; k is a constant.

[0032] It should be noted that the actual speaking speed of the agent is obtained by analyzing the existing artificial intelligence technology; the preset ideal speaking speed can be set according to the specific business; and the constant term k represents the weight of the communication fluency score.

[0033] Specifically, the chat pause score obtained based on the historical voice data includes: analyzing the historical voice data through artificial intelligence to obtain the number of pauses and interruptions of the agent; and obtaining the chat pause score through the pause formula based on the number of pauses and interruptions. The pause formula is: Among them, Score 停顿 Score for chat pause; C 停顿 is the number of pauses; C 中断 is the number of interruptions; l is a constant term.

[0034] It should be noted that the number of pauses and interruptions of the agent is analyzed using existing artificial intelligence technology, and the constant term l represents the weight of the chat pause score.

[0035] In specific practice, "obtaining an agent priority coefficient according to an agent score" in step S13 includes: sorting the agents from high to low according to the agent scores to obtain an agent list; and obtaining the agent priority coefficient of the agent according to the sequence number of the agent in the agent list.

[0036] It should be noted that the serial number in each agent list is pre-set with a corresponding agent priority coefficient. When the agents are sorted according to the agent scores, each agent has a corresponding serial number, and thus each agent has a corresponding agent priority coefficient.

[0037] In specific practice, the step S14 of “allocating calls according to the agent priority coefficient” includes: obtaining a target agent through an allocation formula, obtaining the latest call and allocating the latest call to the target agent; the allocation formula is: ; Among them, P i The priority coefficient of agent i; M ij Indicates whether the skills of agent i match the current call type j; Q i Characterizes the call queue length of agent i; W i represents the average waiting time of agent i; C i Indicates the cumulative number of calls made by agent i; S i Characterizes the service level of agent i; α, β, γ, δ and λ are all constant terms.

[0038] It should be noted that the allocation formula combines the existing minimum queue length allocation, shortest average waiting time allocation and load balancing allocation, and also combines the service level optimization allocation and priority allocation based on agent scores; α, β, γ, δ and λ represent the weights of minimum queue length allocation, shortest average waiting time allocation, load balancing allocation, service level optimization allocation and priority allocation respectively; α, β, γ, δ and λ are set according to specific needs.

[0039] Specifically, the goal of the minimum queue length allocation is to allocate new calls to the agent with the least number of people waiting in the current queue. The formula is: ; Among them, Q i Indicates the call queue length of agent i.

[0040] Specifically, the goal of the shortest average waiting time allocation is to allocate to the agent who can guarantee the shortest waiting time. The formula is: ; Among them, W i It represents the average waiting time of agent i.

[0041] Specifically, the goal of load balancing is to balance the call volume of each agent and avoid overloading some agents. The formula is: Among them, C i Indicates the cumulative number of calls made by agent i.

[0042] Specifically, the service level optimization allocation goal is to allocate calls to the most suitable agent according to the agent's service level to improve service quality. The formula is: , where S i It represents the service level of agent i, that is, the agent score of agent i on that day.

[0043] Specifically, priority allocation target: Allocation based on agent priority coefficient and call type, the formula is: , where P i Indicates the agent priority coefficient, M ij Indicates whether the skills of agent i match the current call type j. The original skill coefficient of the agent is 1. M ij The value is mainly affected by the following situations: 1. In AI intent recognition, if there is a problem in the last 10 calls that the current agent cannot solve, -0.5; 2. The higher the agent's ranking in the agent ranking, the higher the value is +0.2 for each ranking.

[0044] See also Figure 2 , Figure 2 is a schematic block diagram of a human agent allocation device shown in an exemplary embodiment of the present invention, see Figure 2 , the manual seat allocation device 100 includes: The acquisition module 101 is used to acquire the agent data of each agent; the agent data includes the historical chat content, historical voice data and customer satisfaction evaluation of the previous day; Scoring module 102, used to obtain the seat score of each seat through a scoring algorithm according to the seat data; A screening module 103 is used to obtain a seat priority coefficient according to the seat score; The allocation module 104 is used to allocate calls according to the agent priority coefficient.

[0045] It should be noted that the technical solution provided in this embodiment can be applied to scenarios including, but not limited to, call center agent allocation in specific practice.

[0046] It can be understood that the device provided in this embodiment obtains the agent score of each agent through a scoring algorithm based on the historical chat content, historical voice data and customer satisfaction evaluation of the previous day, executes the elimination and promotion strategy according to the agent score to obtain the agent priority coefficient, and performs call allocation according to the agent priority coefficient; this technical solution combines the service level of the agent when allocating seats, which can effectively improve customer experience.

[0047] See also Figure 3 , Figure 3 is a schematic block diagram of an electronic device according to an exemplary embodiment of the present invention, see Figure 3 , an electronic device 200, comprising: at least one processor 202; and A memory 201 is communicatively connected to at least one processor 202; wherein, The memory 201 stores instructions that can be executed by at least one processor 202. The instructions are executed by the at least one processor 202 so that the at least one processor 202 can execute any of the above-mentioned methods for allocating manual seats.

[0048] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

[0049] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0050] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0051] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for allocating artificial seats, characterized in that: The method comprises: Acquire the agent data of each agent; the agent data includes the historical chat content, historical voice data and customer satisfaction evaluation of the previous day; Obtaining a seat score for each seat through a scoring algorithm based on the seat data; Obtaining an agent priority coefficient according to the agent score; Call allocation is performed according to the agent priority coefficient.

2. The distribution method according to claim 1, characterized in that: The scoring algorithm comprises: The seat score is obtained by a scoring formula based on the seat data, and the scoring formula is: ； Among them, Score 座席 Score is the seat score; 关键词 Score is the keyword matching score obtained based on the historical chat content; 情绪 Score is the emotion score obtained by analyzing historical voice data through artificial intelligence; 流畅度 Score is the communication fluency score obtained based on historical voice data; 停顿 Score is the chat pause score obtained based on historical voice data; 满意度 The score for customer satisfaction evaluation.

3. The distribution method according to claim 2, characterized in that: The keyword matching scores obtained based on the historical chat content include: Obtaining the total number of keywords and the number of matching keywords after identifying the historical chat content; The keyword matching score is obtained according to the total number of keywords and the number of matching keywords through a matching formula, and the matching formula is: ； Among them, Score 关键词 is the keyword matching score; C 匹配 is the number of matching keywords; C 总数 is the total number of keywords.

4. The distribution method according to claim 2, characterized in that: The emotion score obtained by analyzing the historical voice data through artificial intelligence includes: Analyzing the historical voice data through artificial intelligence to obtain the emotional state of the agent; the emotional state includes anger, calmness and happiness; A corresponding preset score is used as the emotion score according to the emotion state.

5. The distribution method according to claim 2, characterized in that: The communication fluency score obtained based on the historical voice data includes: Analyze the historical voice data through artificial intelligence to obtain the actual speaking speed of the agent; The communication fluency score is obtained according to the actual speaking speed through the fluency formula, and the fluency formula is: ； Among them, Score 流畅度 Y is the communication fluency score; 实际 is the actual speaking speed; Y 理想 is the preset ideal speaking speed; k is a constant.

6. The distribution method according to claim 2, characterized in that: The chat pause score obtained based on the historical voice data includes: Analyze the historical voice data through artificial intelligence to obtain the number of pauses and interruptions of the agent; The chat pause score is obtained according to the number of pauses and the number of interruptions through a pause formula, and the pause formula is: ； Among them, Score 停顿 Score for chat pause; C 停顿 is the number of pauses; C 中断 is the number of interruptions; l is a constant term.

7. The distribution method according to claim 1, characterized in that: The step of obtaining the seat priority coefficient according to the seat score includes: Sort the seats from high to low according to the seat scores to obtain a seat list; The seat priority coefficient of the agent is obtained according to the sequence number of the agent in the seat list.

8. The distribution method according to claim 1, characterized in that: The call allocation according to the seat priority coefficient includes: Obtain a target agent through an allocation formula, obtain the latest call and allocate the latest call to the target agent; The allocation formula is: ； Among them, P i The priority coefficient of agent i; M ij Indicates whether the skills of agent i match the current call type j; Q i Characterizes the call queue length of agent i; W i represents the average waiting time of agent i; C i Indicates the cumulative number of calls made by agent i; S i Characterizes the service level of agent i; α, β, γ, δ and λ are all constant terms.

9. A device for allocating artificial seats, characterized in that: The device comprises: An acquisition module is used to acquire the agent data of each agent; the agent data includes the historical chat content, historical voice data and customer satisfaction evaluation of the previous day; A scoring module, used to obtain a seat score for each seat through a scoring algorithm based on the seat data; A screening module, used for obtaining a seat priority coefficient according to the seat score; The allocation module is used to allocate calls according to the agent priority coefficient.

10. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for allocating artificial seats according to any one of claims 1 to 6.

Images