Resource scheduling method and device

Abstract

The present invention discloses a resource scheduling method and device, and relates to the field of computer technology. A specific implementation of the method includes: receiving a consultation request initiated by a user, and predicting the object of receiving the consultation request based on the user's historical behavior data and user portrait data; if the object is a robot customer service, sending the consultation request to the robot customer service for reception, obtaining a reply message and returning it, or obtaining an unanswerable message to trigger a transfer to manual instructions; if the object is a manual customer service or a transfer to manual instructions is received, determining the user's consultation intention, matching the user's consultation intention with the intention portrait that each manual customer service group can receive; determining the target manual customer service group with the highest matching degree, sending the consultation request to the target manual customer service group for reception, obtaining a reply message and returning it. This implementation method uses data such as user historical behavior data and portrait data, the real-time acceptance volume and intention portrait of manual customer service, etc. to accurately perform customer service scheduling and improve customer service reception efficiency and satisfaction.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a resource scheduling method and device. Background Art

[0002] The customer service system is a dialogue system that can provide users with one-on-one online consultation and answering services. At present, the mainstream customer service systems in the industry generally provide answering services by providing robot customer service and manual customer service. After the user initiates a consultation online, the robot customer service will be given priority to receive the consultation. If the robot customer service can successfully answer the user's question, the robot customer service will receive the consultation. Otherwise, manual customer service will be required to intervene to provide more flexible customer service.

[0003] In the process of implementing the present invention, the inventors found that the existing customer service resource scheduling mode has the following problems:

[0004] 1. By default, robot customer service is used to handle all consultation traffic. Only after the user actively initiates a request to transfer to a human customer service can the human customer service handle it. As a result, high-value and high-risk user consultations cannot be assigned to human customer service with higher answering effects in the first place, which directly affects the consultation satisfaction of these users;

[0005] 2. The user's real-time consultation situation is not taken into consideration, the scheduling flexibility is low, and the resource utilization rate is poor; the traffic received by manual customer service cannot be accurately allocated to the most suitable customer service, resulting in the need to re-explore user intentions and manually transfer the call again after the manual call is picked up. Summary of the invention

[0006] In view of this, an embodiment of the present invention provides a resource scheduling method and device, which can at least solve the problem in the prior art that robot customer service is used by default to handle all consultation traffic, the accuracy of user intention allocation is low, and customer service resources cannot be scheduled differently.

[0007] To achieve the above object, according to one aspect of an embodiment of the present invention, a resource scheduling method is provided, comprising:

[0008] Receive consultation requests initiated by users, and predict who will receive the consultation requests based on user historical behavior data and user profile data;

[0009] If the object is a robot customer service, the consultation request is sent to the robot customer service reception, and a reply message is received and returned, or a message that the answer cannot be answered is received to trigger a manual instruction;

[0010] If the object is a manual customer service or receives an instruction to transfer to a manual service, the user's consultation intention is determined, and the user's consultation intention is matched with the intention profile that each manual customer service group can receive;

[0011] The target manual customer service group with the highest matching degree is determined, the consultation request is sent to the target manual customer service group for reception, and a reply message is obtained and returned.

[0012] Optionally, predicting the object receiving the consultation request based on the user's historical behavior data and user portrait data includes:

[0013] Calculate the weight of each data item in each user portrait data and the historical behavior data, and add them up to obtain the total weight;

[0014] If the total weight is greater than the preset weight, the result of manual customer service reception is returned, otherwise the result of robot customer service reception is returned.

[0015] Optionally, the calculating the weight of each data item in each user portrait data and the historical behavior data includes:

[0016] Determine the channel information through which the user initiates the consultation request, and obtain the real-time behavior data of the user's current operation from the channel information;

[0017] Calculate the weight of each data item in each user portrait data, the historical behavior data, and the real-time behavior data.

[0018] Optionally, sending the consultation request to a robot customer service receptionist includes:

[0019] According to the real-time data of communication with the user, the user's consultation intention is predicted in real time, and the reply information corresponding to the user's consultation intention and the real-time data is queried.

[0020] Optionally, the method further includes: after returning the reply information transmitted by the robot customer service to the user, if a transfer request initiated by the user is received, triggering a transfer instruction to a human operator; or

[0021] If a message that cannot be answered is received from the robot customer service, a prompt message indicating whether to transfer to a human operator will be returned to the user, thereby triggering a transfer to a human operator instruction based on the transfer option selected by the user.

[0022] Optionally, determining the user's consultation intention includes:

[0023] In the case where the object is a manual customer service, predicting the user's consultation intention based on the user's historical behavior data and the user portrait data; or

[0024] When receiving an instruction to transfer to manual service, the user's consultation intention is obtained from the instruction to transfer to manual service; wherein, the user's consultation intention is predicted by the robot customer service through analyzing the real-time data of communication with the user.

[0025] Optionally, sending the consultation request to the target manual customer service group for reception includes:

[0026] Obtain a manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and filter at least one manual customer service whose score is greater than a preset score;

[0027] Obtaining the real-time reception volume of each manual customer service of the at least one manual customer service, and predicting the probability of each manual customer service receiving the consultation request in combination with the reception capacity score;

[0028] The consultation request is sent to a human customer service representative with the highest probability.

[0029] Optionally, before loading the reception capability score of each manual customer service, the method further includes:

[0030] Obtaining a user consultation intention set corresponding to the target manual customer service reception group;

[0031] For each human customer service representative, we obtain historical reception data for each user’s consultation intention, and conduct a reception capability portrait based on the user’s consultation intention and reception evaluation satisfaction, thus obtaining the reception capability score for each user’s consultation intention.

[0032] Optionally, sending the consultation request to the target manual customer service group for reception includes:

[0033] Obtain the manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and determine the manual customer service with the largest score;

[0034] Predicting the probability of the one manual customer service receiving the consultation request according to the real-time reception volume and reception capacity score of the one manual customer service;

[0035] If the probability is greater than or equal to the preset probability, the consultation request is sent to the one manual customer service representative;

[0036] Otherwise, another manual customer service with the second highest score is determined, and the above steps are repeated to obtain a target manual customer service with a probability greater than or equal to a preset probability, and the consultation request is sent to the target manual customer service for reception.

[0037] To achieve the above object, according to another aspect of an embodiment of the present invention, a resource scheduling device is provided, comprising:

[0038] The human-machine diversion decision module is used to receive consultation requests initiated by users and predict the object of receiving the consultation request based on the user's historical behavior data and user portrait data;

[0039] A machine reception module is used to send the consultation request to the robot customer service reception if the object is a robot customer service, obtain a reply message and return it, or obtain an unanswerable message to trigger a manual instruction;

[0040] The manual reception module is used to determine the user's consultation intention if the object is a manual customer service or receives a transfer instruction to a manual service, and match the user's consultation intention with the intention portrait that can be received by each manual customer service group; determine the target manual customer service group with the highest matching degree, send the consultation request to the target manual customer service group for reception, obtain the reply information and return it.

[0041] Optionally, the human-machine diversion decision module is used to:

[0042] Calculate the weight of each data item in each user portrait data and the historical behavior data, and add them up to obtain the total weight;

[0043] If the total weight is greater than the preset weight, the result of manual customer service reception is returned, otherwise the result of robot customer service reception is returned.

[0044] Optionally, the human-machine diversion decision module is used to:

[0045] Determine the channel information through which the user initiates the consultation request, and obtain the real-time behavior data of the user's current operation from the channel information;

[0046] Calculate the weight of each data item in each user portrait data, the historical behavior data, and the real-time behavior data.

[0047] Optionally, the machine reception module is used to:

[0048] According to the real-time data of communication with the user, the user's consultation intention is predicted in real time, and the reply information corresponding to the user's consultation intention and the real-time data is queried.

[0049] Optionally, the machine reception module is also used to:

[0050] After the reply information transmitted by the robot customer service is returned to the user, if a transfer request initiated by the user is received, the transfer instruction is triggered; or

[0051] If a message that cannot be answered is received from the robot customer service, a prompt message indicating whether to transfer to a human operator will be returned to the user, thereby triggering a transfer to a human operator instruction based on the transfer option selected by the user.

[0052] Optionally, the manual reception module is used to:

[0053] In the case where the object is a manual customer service, predicting the user's consultation intention based on the user's historical behavior data and the user portrait data; or

[0054] When receiving an instruction to transfer to manual service, the user's consultation intention is obtained from the instruction to transfer to manual service; wherein, the user's consultation intention is predicted by the robot customer service through analyzing the real-time data of communication with the user.

[0055] Optionally, the manual reception module is used to:

[0056] Obtain a manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and filter at least one manual customer service whose score is greater than a preset score;

[0057] Obtaining the real-time reception volume of each manual customer service of the at least one manual customer service, and predicting the probability of each manual customer service receiving the consultation request in combination with the reception capacity score;

[0058] The consultation request is sent to a human customer service representative with the highest probability.

[0059] Optionally, the manual reception module is further used to:

[0060] Obtaining a user consultation intention set corresponding to the target manual customer service reception group;

[0061] For each human customer service representative, we obtain historical reception data for each user’s consultation intention, and conduct a reception capability portrait based on the user’s consultation intention and reception evaluation satisfaction, thus obtaining the reception capability score for each user’s consultation intention.

[0062] Optionally, the manual reception module is used to:

[0063] Obtain the manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and determine the manual customer service with the largest score;

[0064] Predicting the probability of the one manual customer service receiving the consultation request according to the real-time reception volume and reception capacity score of the one manual customer service;

[0065] If the probability is greater than or equal to the preset probability, the consultation request is sent to the one manual customer service representative;

[0066] Otherwise, another manual customer service with the second highest score is determined, and the above steps are repeated to obtain a target manual customer service with a probability greater than or equal to a preset probability, and the consultation request is sent to the target manual customer service for reception.

[0067] To achieve the above objective, according to another aspect of an embodiment of the present invention, a resource scheduling electronic device is provided.

[0068] The electronic device of an embodiment of the present invention includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement any of the resource scheduling methods described above.

[0069] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by a processor, any of the resource scheduling methods described above is implemented.

[0070] According to the solution provided by the present invention, one embodiment of the above invention has the following advantages or beneficial effects: in response to the defects of existing intelligent customer service scheduling, a new solution is provided that uses user historical behavior data, user portrait data, real-time acceptance volume of manual customer service, group intention portrait, and combines multi-dimensional data such as user consultation intention to dynamically and intelligently schedule human and machine resources, which can accurately realize the dynamic allocation and scheduling of machine and manual customer service, and improve customer service reception efficiency and satisfaction.

[0071] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0073] Figure 1 A schematic diagram of the main process of a resource scheduling method according to an embodiment of the present invention;

[0074] Figure 2 is a flow chart of an optional resource scheduling method according to an embodiment of the present invention;

[0075] Figure 3 It is a schematic diagram of matching the intent profiles that the manual customer service team can receive with the user's inquiry intent;

[0076] Figure 4 It is a schematic diagram of the intelligent scheduling process of human-machine resources in the customer service system according to the present invention;

[0077] Figure 5 is a schematic diagram of main modules of a resource scheduling device according to an embodiment of the present invention;

[0078] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0079] Figure 7 It is a schematic diagram of the structure of a computer system of a mobile device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0081] At present, the mainstream customer service systems in the industry generally provide robot customer service and manual customer service to provide answering services. These two customer service modes have the following characteristics:

[0082] 1) Robot customer service: Advantages - high efficiency, large reception volume, fast expansion speed, and low resource utilization cost; Disadvantages - low language comprehension ability, low user satisfaction, and poor user experience;

[0083] 2) Manual customer service: Advantages - good user intent exploration effect, high reception satisfaction, and excellent user experience; Disadvantages - high resource utilization cost, low reception efficiency, and small reception volume.

[0084] Based on the above characteristics, robot customer service, due to its efficient response speed and high reception capacity, is mainly used to provide scripted responses, quick question navigation, user consultation intention exploration, etc.; while manual customer service, due to its good language comprehension ability, is usually used to provide complex scenario customer service services such as answering difficult questions, advanced user consultation, member consultation, and user complaints.

[0085] The existing customer service system resource scheduling in the market can be divided into three subsystems: scheduling system, robot customer service, and manual customer service. The scheduling system is mainly responsible for the allocation of consultation relationships between robot customer service and manual customer service, and provides static scheduling capabilities based on the degree of manual reception saturation, user-initiated transfer to manual service, and fixed consultation rules; robot customer service is mainly responsible for receiving conversations assigned by the scheduling system and answering simple questions; manual customer service is mainly responsible for receiving conversations assigned by the scheduling system, and manual reception is the final reception link to close the loop of user consultation.

[0086] See also Figure 1 , which shows a main flow chart of a resource scheduling method provided by an embodiment of the present invention, comprising the following steps:

[0087] S101: receiving a consultation request initiated by a user, and predicting a party to receive the consultation request based on the user's historical behavior data and user portrait data;

[0088] S102: If the object is a robot customer service, the consultation request is sent to the robot customer service reception, and a reply message is received and returned, or a message that the answer cannot be answered is received to trigger a manual instruction;

[0089] S103: If the object is a manual customer service or receives a transfer instruction to a manual customer service, determine the user's consultation intention, and match the user's consultation intention with the intention profile that each manual customer service group can receive;

[0090] S104: Determine the target manual customer service group with the highest matching degree, send the consultation request to the target manual customer service group for reception, obtain reply information and return it.

[0091] In the above implementation, for step S101, the user initiates a consultation request to the customer service system, and the intelligent scheduling system preliminarily undertakes the traffic. The intelligent scheduling system uses an intelligent human-machine diversion algorithm to query the user's historical behavior data in the past 7 days (only for example, actual adjustable) according to the user's login account, IP address and region of the incoming line, including historical purchase records, consultation records and merchant marketing activity records, etc. Each user's account may have more than one label, so the loaded user portrait data may also be multiple, such as high value, high risk, high purchase intention, etc.

[0092] The linear regression algorithm is used to calculate the weight of each data item in each user portrait data and historical behavior data. For each data item, the number of feature values ​​calculated by the algorithm may not be limited to the number of its own feature values. For example, if the order status is completed, the calculated feature values ​​include "probability of consulting after-sales, personal reputation value, operation status", etc.

[0093] The linear regression algorithm here is obtained through training. Specifically, historical behavior data and user portrait data of multiple users are obtained in advance, and the weight of each data item is labeled in advance using a manual labeling method. Then, they are input into the linear regression algorithm model for training to obtain a trained linear regression algorithm.

[0094] For example, input items: Order status = dispute, User identity = plus, Historical consultation = active consultation, Consultation channel = after-sales entrance, Account level = l5, Manual reception record = l3

[0095] Output items: Dispute status manual reception weight 0.17, plus user manual reception weight 0.12, active consultation manual reception weight 0.11, after-sales entrance manual reception weight 0.09, account level manual reception weight 0.08, manual reception record weight 0.03

[0096] The total weight is 0.61 after adding up the weights. If it exceeds the preset weight of 0.5, the result of manual customer service reception is returned. Otherwise, it means that robot customer service reception is required.

[0097] In addition to the user's historical behavior data and user portrait data, the user's current real-time behavior data can also be obtained based on the channel information of the user's consultation request, such as the system's real-time hot issues in the market and the recent marketing activities of the merchants corresponding to the consultation products. Different weights are assigned to each data item in the user's historical behavior data and real-time behavior data, and the user portrait data, and then the reception weight weighted prediction is performed.

[0098] Furthermore, in order to filter out data with low relevance and reduce the amount of calculated data, a weight threshold can be preset to adaptively filter out user portrait data and data items with weights lower than the preset weight threshold.

[0099] For step S102, if the prediction is for a robot customer service, the consultation request is assigned to the robot customer service for reception and processing. The robot customer service reception stage belongs to real-time intention exploration, client interaction with the user, process response mining, and analysis of the data communicated with the user to predict the actual intention of the user's consultation. For example, during the robot reception stage, the user will be asked what questions they want to consult, what kind of goods and orders they need to consult, etc. If the intention explored by the robot customer service exceeds its reception scope or there is no corresponding reply information, it means that it cannot be answered and needs to be transferred to manual processing.

[0100] In addition to situations where the robot customer service cannot answer, there are also some other manually configured rules to trigger instructions to transfer to human operators, such as the rules for receiving high-value users. When such users come in for consultation, if the robot customer service finds that their user profile is a high-value user, the instructions to transfer to human operators will be automatically triggered. Or the robot customer service queries for reply information to answer the consultation, but after the intelligent scheduling system returns the reply information to the user, the user is dissatisfied with the reply and actively initiates a request to transfer to human operators. Furthermore, for situations where the robot customer service cannot answer, it can also return a prompt message to the user whether to transfer to human operators. The instruction to transfer to human operators is only triggered after the user selects the option to transfer to human operators.

[0101] For step S103, for situations that clearly tend to require manual processing, such as users with complaint intentions and messages triggering risk control, they will not be assigned to robot customer service for reception, but will be received by manual customer service.

[0102] Since there are a large number of manual customer service representatives, they can be grouped according to their skills in advance. Each manual customer service group can handle a variety of skills, that is, the user's consultation intentions can be diverse, and the two are in a one-to-many relationship. For situations that require manual customer service, the intelligent scheduling system summarizes the user's consultation intentions in real time to match the appropriate manual customer service group, and then selects the appropriate target manual customer service representative from them, who will then conduct a closed-loop reception.

[0103] The method provided in the above embodiment analyzes user historical behavior data and user portrait data, and can accurately realize the dynamic allocation and scheduling of machine and manual customer service. Regardless of whether the robot customer service can answer, the system can provide services to the outside world normally, thereby improving customer service reception efficiency and user satisfaction.

[0104] See also Figure 2 , shows a schematic diagram of an optional resource scheduling method according to an embodiment of the present invention, comprising the following steps:

[0105] S201: receiving a consultation request initiated by a user, and predicting a party to receive the consultation request based on the user's historical behavior data and user portrait data;

[0106] S202: If the object is a robot customer service, the consultation request is sent to the robot customer service reception, and a reply message is received and returned, or a message that the answer cannot be answered is received to trigger a manual instruction;

[0107] S203: When the object is a manual customer service, predict the user's consultation intention according to the user's historical behavior data and the user portrait data;

[0108] S204: when receiving the instruction to transfer to a human operator, obtaining the user's consultation intention from the instruction to transfer to a human operator; wherein the user's consultation intention is predicted by the robot customer service by analyzing the real-time data of communication with the user;

[0109] S205: Match the user's consultation intention with the intention profile that can be received by each manual customer service group;

[0110] S206: Determine the target manual customer service group with the highest matching degree, send the consultation request to the target manual customer service group for reception, obtain reply information and return it.

[0111] In the above embodiment, for steps S201 and S202, see Figure 1 The description of steps S101 and S102 is not repeated here.

[0112] For steps S203 to S206, each manual customer service group is responsible for different areas, which leads to different user consultation intentions that can be processed, such as manual customer service group A-price protection intention, manual customer service group B-complaint intention. A manual customer service group may include multiple manual customer service.

[0113] See also Figure 3As shown in the figure, for the decision-making diversion traffic, it is necessary to predict the user's consultation intention based on the user's historical behavior data and user portrait data, and dynamically match it with the intention portrait that each artificial customer service group can receive, to obtain the target artificial customer service group with the highest matching degree. Furthermore, the number of predicted user consultation intentions may be large, and these user consultation intentions can be Figure 1 The two lists are sent to users at the same time, and they can make their own choices to improve the accuracy of subsequent matching.

[0114] When a robot is transferred to a human, it will also carry the user consultation intent predicted by the robot customer service. Therefore, if the traffic is transferred from a machine to a human, it is necessary to dynamically match the user consultation intent predicted by the robot customer service with the intent profile that each human customer service group can receive, and obtain the target human customer service group with the highest matching degree. For example, if the robot customer service predicts the price protection intent, the human customer service group that can receive the price protection intent will be matched first in the future.

[0115] After obtaining the target manual customer service group based on the user's consultation intent, load the reception capability rank value of each manual customer service in the group and sort the scores from high to low. The higher the score, the better the reception effect. There are two ways to determine the reception of the consultation request in this group:

[0116] 1. First, filter out the manual customer service staff with scores greater than the preset score (such as 0.5) from the group, obtain the real-time reception volume of each manual customer service staff, and combine the two to predict the probability of each manual customer service staff receiving the consultation request. Finally, send the user's consultation request to the manual customer service staff with the highest probability.

[0117] 2. First, determine the human customer service with the highest score from the group, obtain the real-time reception volume of the human customer service, and calculate the probability of the human customer service receiving the consultation request. If the probability is greater than 50% (just an example), the human customer service will receive the consultation request. Otherwise, start from another human customer service ranked second, and repeat the above steps until the target human customer service with a probability value greater than 50% is selected. Finally, the user consultation request is sent to the selected target human customer service for processing.

[0118] In addition, the reception capacity rank value calculation of each manual customer service can first determine the user consultation intention that the manual customer service group can handle, obtain the customer data that has been received under each user consultation intention, that is, the historical reception data, and combine the customer's intention at the time of consultation and the reception evaluation satisfaction score to create a reception capacity profile. The higher the reception satisfaction of customers with a certain intention, the higher the reception capacity rank value.

[0119] The method provided in the above embodiment, for the scenario of reception by manual customer service, matches the user's consultation intention and calculates the reception capacity, and selects a manual customer service reception whose reception probability meets the requirements from the target manual skill customer service group, so as to improve the accuracy of the reply information.

[0120] See also Figure 4 , which shows the intelligent scheduling process of human-machine resources in the customer service system of this solution:

[0121] 1. When receiving a user consultation request, first use intelligent human-machine diversion decision-making to predict whether it will be received by manual customer service by analyzing the user's historical behavior data and user portrait data;

[0122] 2. For scenarios that can be handled by robot customer service, the robot customer service analyzes the real-time communication data with the user, predicts the user's consultation intention, and determines whether the consultation intention can be handled. If it can be handled, it checks whether there is corresponding reply information, otherwise it transfers to manual service. In addition, if the user is not satisfied with the reply information, it can also trigger the transfer to manual service;

[0123] 3. For scenarios that require human customer service, first use intelligent intent diversion to select the target human customer service reception group that best matches the user's consultation intent from multiple human customer service reception groups;

[0124] 4. For the target manual customer service reception group, through intelligent customer service resource allocation, the target manual customer service whose real-time reception volume and reception score meet the requirements are screened out. The target manual customer service will receive the user's consultation, obtain the reply information and return it, thereby improving the accuracy of the reply information.

[0125] Compared with the prior art, the method provided by the embodiment of the present invention has at least the following beneficial effects:

[0126] 1. Using the intelligent human-machine diversion decision-making function, real-time dynamic and differentiated customer service resource allocation can be performed for user consultations. For example, for users with obvious complaint intentions and large emotional fluctuations, they can be automatically assigned to manual customer service for priority processing based on such intentions, effectively avoiding the situation where such traffic is assigned to robot customer service, which aggravates the user's negative emotions;

[0127] 2. Use the intention profile of the manual customer service group to dynamically match the user's consultation intention, and accurately assign the user's consultation to the manual customer service group with the highest matching degree. For example, in the pre-sales activity consultation in the e-commerce field, you can select the group with high proficiency in the activity in the profile to assign it, so that the user's consultation questions can be received in the first time, effectively avoiding the initiation of a second manual transfer during the manual consultation stage;

[0128] 3. For each manual customer service, a reception capability profile is created based on historical reception data, and the customer service reception capability is ranked and scored. Differentiated traffic is allocated according to the score, so that manual customer service with strong reception capability can give priority to receiving users, thereby improving user satisfaction.

[0129] See also Figure 5 , which shows a schematic diagram of main modules of a resource scheduling device 500 provided in an embodiment of the present invention, including:

[0130] The human-machine diversion decision module 501 is used to receive a consultation request initiated by a user and predict the object to receive the consultation request based on the user's historical behavior data and user portrait data;

[0131] The machine reception module 502 is used to send the consultation request to the robot customer service reception if the object is a robot customer service, get a reply message and return it, or get an unanswerable message to trigger a manual instruction;

[0132] The manual reception module 503 is used to determine the user's consultation intention if the object is a manual customer service or receives a transfer instruction to a manual service, and match the user's consultation intention with the intention portrait that can be received by each manual customer service group; determine the target manual customer service group with the highest matching degree, send the consultation request to the target manual customer service group for reception, obtain the reply information and return it.

[0133] In the implementation device of the present invention, the human-machine diversion decision module 501 is used to:

[0134] Calculate the weight of each data item in each user portrait data and the historical behavior data, and add them up to obtain the total weight;

[0135] If the total weight is greater than the preset weight, the result of manual customer service reception is returned, otherwise the result of robot customer service reception is returned.

[0136] In the implementation device of the present invention, the human-machine diversion decision module 501 is used to:

[0137] Determine the channel information through which the user initiates the consultation request, and obtain the real-time behavior data of the user's current operation from the channel information;

[0138] Calculate the weight of each data item in each user portrait data, the historical behavior data, and the real-time behavior data.

[0139] In the implementation device of the present invention, the machine reception module 502 is used to:

[0140] According to the real-time data of communication with the user, the user's consultation intention is predicted in real time, and the reply information corresponding to the user's consultation intention and the real-time data is queried.

[0141] In the implementation device of the present invention, the machine reception module 502 is also used for:

[0142] After the reply information transmitted by the robot customer service is returned to the user, if a transfer request initiated by the user is received, the transfer instruction is triggered; or

[0143] If a message that cannot be answered is received from the robot customer service, a prompt message indicating whether to transfer to a human operator will be returned to the user, thereby triggering a transfer to a human operator instruction based on the transfer option selected by the user.

[0144] In the implementation device of the present invention, the manual reception module 503 is used to:

[0145] In the case where the object is a manual customer service, predicting the user's consultation intention based on the user's historical behavior data and the user portrait data; or

[0146] When receiving an instruction to transfer to manual service, the user's consultation intention is obtained from the instruction to transfer to manual service; wherein, the user's consultation intention is predicted by the robot customer service through analyzing the real-time data of communication with the user.

[0147] In the implementation device of the present invention, the manual reception module 503 is used to:

[0148] Obtain a manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and filter at least one manual customer service whose score is greater than a preset score;

[0149] Obtaining the real-time reception volume of each manual customer service of the at least one manual customer service, and predicting the probability of each manual customer service receiving the consultation request in combination with the reception capacity score;

[0150] The consultation request is sent to a human customer service representative with the highest probability.

[0151] In the implementation device of the present invention, the manual reception module 503 is also used for:

[0152] Obtaining a user consultation intention set corresponding to the target manual customer service reception group;

[0153] For each human customer service representative, we obtain historical reception data for each user’s consultation intention, and conduct a reception capability portrait based on the user’s consultation intention and reception evaluation satisfaction, thus obtaining the reception capability score for each user’s consultation intention.

[0154] In the implementation device of the present invention, the manual reception module 503 is used to:

[0155] Obtain the manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and determine the manual customer service with the largest score;

[0156] Predicting the probability of the one manual customer service receiving the consultation request according to the real-time reception volume and reception capacity score of the one manual customer service;

[0157] If the probability is greater than or equal to the preset probability, the consultation request is sent to the one manual customer service representative;

[0158] Otherwise, another manual customer service with the second highest score is determined, and the above steps are repeated to obtain a target manual customer service with a probability greater than or equal to a preset probability, and the consultation request is sent to the target manual customer service for reception.

[0159] In addition, the specific implementation content of the device described in the embodiment of the present invention has been described in detail in the method described above, so the repeated content will not be described again here.

[0160] The intelligent human-machine resource scheduling device for the customer service system provided by the embodiment of the present invention can effectively solve the pain points of the existing customer service system, such as simple human-machine resource scheduling function and low scheduling allocation accuracy, and has a significant improvement in application effect and great market promotion value.

[0161] Figure 6 An exemplary system architecture 600 to which an embodiment of the present invention may be applied is shown, including terminal devices 601 , 602 , 603 , a network 604 and a server 605 (only an example).

[0162] Terminal devices 601, 602, 603 can be various electronic devices with display screens and supporting web browsing, and various communication client applications are installed. Users can use terminal devices 601, 602, 603 to interact with server 605 through network 604 to receive or send messages, etc.

[0163] The network 604 is used to provide a medium for communication links between the terminal devices 601, 602, 603 and the server 605. The network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0164] Server 605 may be a server that provides various services, and is used to receive user inquiries, predict the object that will receive the inquiry, and transfer the inquiry to the object for reception.

[0165] It should be noted that the method provided in the embodiment of the present invention is generally executed by the server 605 , and accordingly, the device is generally set in the server 605 .

[0166] It should be understood that Figure 6The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0167] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of a terminal device suitable for implementing an embodiment of the present invention. Figure 7 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0168] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0169] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0170] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0171] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0172] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0173] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor. For example, they may be described as: a processor includes a human-machine diversion decision module, a machine reception module, and a manual reception module. The names of these modules do not, in some cases, constitute a limitation on the modules themselves. For example, the manual reception module may also be described as a "manual reception consultation module."

[0174] As another aspect, the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes:

[0175] Receive consultation requests initiated by users, and predict who will receive the consultation requests based on user historical behavior data and user profile data;

[0176] If the object is a robot customer service, the consultation request is sent to the robot customer service reception, and a reply message is received and returned, or a message that the answer cannot be answered is received to trigger a manual instruction;

[0177] If the object is a manual customer service or receives an instruction to transfer to a manual service, the user's consultation intention is determined, and the user's consultation intention is matched with the intention profile that each manual customer service group can receive;

[0178] The target manual customer service group with the highest matching degree is determined, the consultation request is sent to the target manual customer service group for reception, and a reply message is obtained and returned.

[0179] According to the technical solution of the embodiments of the present invention, in order to address the defects of the existing intelligent customer service scheduling, a new solution is provided which uses user historical behavior data, user portrait data, real-time manual customer service acceptance volume, group intention portrait, and combines multi-dimensional data such as user consultation intention to dynamically and intelligently schedule human and machine resources. This solution can accurately realize the dynamic allocation and scheduling of machine and manual customer service, and improve customer service reception efficiency and satisfaction.

[0180] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A resource scheduling method, characterized in that: include: Receive a consultation request initiated by a user, and predict the object that will receive the consultation request based on the user's historical behavior data and user portrait data; wherein the predicted object also includes: obtaining the real-time behavior data of the user's current operation based on the channel information through which the user initiated the consultation request; If the object is a robot customer service, the consultation request is sent to the robot customer service reception, and a reply message is received and returned, or a message that the answer cannot be answered is received to trigger a manual instruction; If the object is a manual customer service or a transfer instruction to a manual service is received, the user's consultation intention is determined, and the user's consultation intention is matched with the intention profile that each manual customer service group can receive; wherein, the determination of the user's consultation intention includes: when the object is a manual customer service, predicting the user's consultation intention based on the user's historical behavior data and the user profile data; or when a transfer instruction to a manual service is received, obtaining the user's consultation intention from the transfer instruction; the user's consultation intention is predicted by the robot customer service through analyzing the real-time data of communication with the user; The target manual customer service group with the highest matching degree is determined, the consultation request is sent to the target manual customer service group for reception, and a reply message is obtained and returned.

2. The method according to claim 1, characterized in that The predicting of the object receiving the consultation request based on the user's historical behavior data and user portrait data includes: Calculate the weight of each data item in each user portrait data and the historical behavior data, and add them up to obtain the total weight; If the total weight is greater than the preset weight, the result of manual customer service reception is returned, otherwise the result of robot customer service reception is returned.

3. The method according to claim 2, characterized in that The calculating the weight of each data item in each user portrait data and the historical behavior data includes: Determine the channel information through which the user initiates the consultation request, and obtain the real-time behavior data of the user's current operation from the channel information; Calculate the weight of each data item in each user portrait data, the historical behavior data, and the real-time behavior data.

4. The method according to claim 1, characterized in that The sending of the consultation request to the robot customer service reception comprises: According to the real-time data of communication with the user, the user's consultation intention is predicted in real time, and the reply information corresponding to the user's consultation intention and the real-time data is queried.

5. The method according to claim 1 or 4, characterized in that: Also includes: After the reply information transmitted by the robot customer service is returned to the user, if a transfer request initiated by the user is received, the transfer instruction is triggered; or If a message that cannot be answered is received from the robot customer service, a prompt message indicating whether to transfer to a human operator will be returned to the user, thereby triggering a transfer to a human operator instruction based on the transfer option selected by the user.

6. The method according to claim 1, characterized in that The sending the consultation request to the target manual customer service group for reception includes: Obtain a manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and filter at least one manual customer service whose score is greater than a preset score; Obtaining the real-time reception volume of each manual customer service of the at least one manual customer service, and predicting the probability of each manual customer service receiving the consultation request in combination with the reception capacity score; The consultation request is sent to a human customer service representative with the highest probability.

7. The method according to claim 6, characterized in that Before loading the reception capability score of each manual customer service, it also includes: Obtaining a user consultation intention set corresponding to the target manual customer service reception group; For each human customer service representative, we obtain historical reception data for each user’s consultation intention, and conduct a reception capability portrait based on the user’s consultation intention and reception evaluation satisfaction, thus obtaining the reception capability score for each user’s consultation intention.

8. The method according to claim 1, characterized in that The sending the consultation request to the target manual customer service group for reception includes: Obtain the manual customer service queue of the target manual customer service group, load the reception capacity score of each manual customer service, and determine the manual customer service with the largest score; Predicting the probability of the one manual customer service receiving the consultation request according to the real-time reception volume and reception capacity score of the one manual customer service; If the probability is greater than or equal to the preset probability, the consultation request is sent to the one manual customer service representative; Otherwise, another manual customer service with the second highest score is determined, and the above steps are repeated to obtain a target manual customer service with a probability greater than or equal to a preset probability, and the consultation request is sent to the target manual customer service for reception.

9. A resource scheduling device, characterized in that: include: The human-machine diversion decision module is used to receive the consultation request initiated by the user, and predict the object receiving the consultation request based on the user's historical behavior data and user portrait data; wherein the predicted object also includes: obtaining the real-time behavior data of the user's current operation based on the channel information of the user initiating the consultation request; A machine reception module is used to send the consultation request to the robot customer service reception if the object is a robot customer service, obtain a reply message and return it, or obtain an unanswerable message to trigger a manual instruction; The manual reception module is used to determine the user's consultation intention if the object is a manual customer service or a transfer instruction is received, and match the user's consultation intention with the intention profile that can be received by each manual customer service group; determine the target manual customer service group with the highest matching degree, send the consultation request to the target manual customer service group for reception, obtain reply information and return it; Among them, determining the user consultation intention includes: when the object is a manual customer service, predicting the user consultation intention based on the user historical behavior data and the user portrait data; or when receiving a transfer to manual instruction, obtaining the user consultation intention from the transfer to manual instruction; the user consultation intention is predicted by the robot customer service through analyzing the real-time data of communication with the user.

10. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 8.

11. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images