Consultation business volume prediction method and device, electronic equipment and computer readable medium

Abstract

The embodiment of the invention discloses a consultation business volume prediction method and device, electronic equipment and a computer readable medium. A specific embodiment of the method comprises the steps of obtaining a historical consultation business volume sequence group, wherein a historical consultation business volume sequence corresponds to a consultation category; determining whether the prediction time period meets a business change point detection condition or not; for each historical consultation business volume sequence, in response to determining that the prediction time period satisfies a business change point detection condition, performing business change point processing on the historical consultation business volume sequence to obtain the historical consultation business volume sequence after business change point processing as a sample sequence; determining a consultation category corresponding to the historical consultation business volume sequence as a target consultation category; and according to the sample sequence, generating a predicted consultation business volume sequence corresponding to the prediction time period and the target consultation category. The embodiment is related to intelligent customer service, and the comprehensiveness of covered services, the consultation service volume prediction efficiency and the predicted consultation service volume accuracy are improved.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technologies, and particularly to a method, an apparatus, an electronic device, and a computer-readable medium for predicting the volume of consulting services. Background Art

[0002] With the development of Internet technologies and online business platforms, the demand for online consulting services has gradually increased, and it is necessary to predict the volume of consulting services in advance. Currently, when predicting the volume of consulting services, the commonly adopted method is as follows: for services with a large volume of business, at regular intervals, the person in charge of each level of business department submits, communicates, adjusts, and summarizes various prediction indicators through an excel template, and then analyzes to obtain the result.

[0003] However, the inventors have found that when predicting the volume of consulting services by the above method, the following technical problems often exist: when predicting the volume of consulting services for services with a large volume of business, the comprehensiveness of the covered services is poor, and data needs to be manually summarized and analyzed at regular intervals, resulting in a large amount of manual work, low efficiency in predicting the volume of consulting services. In addition, there are many internal data calibers. When the caliber of the manually exported data is incorrect, the accuracy of the predicted volume of consulting services is low.

[0004] The above information disclosed in this background art section is only used to enhance the understanding of the background of the inventive concept, and thus, it may include information that does not form the prior art known to those of ordinary skill in the art in this country. Summary of the Invention

[0005] This summary of the disclosure is used to introduce concepts in a brief form, and these concepts will be described in detail in the subsequent detailed implementation section. This summary of the disclosure is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to be used to limit the scope of the claimed technical solution.

[0006] Some embodiments of the present disclosure propose a method, an apparatus, an electronic device, a computer-readable medium, and a computer program product for predicting the volume of consulting services to solve one or more of the technical problems mentioned in the above background art section.

[0007] In a first aspect, some embodiments of the present disclosure provide a method for predicting consulting business volume. The method includes: obtaining a historical consulting business volume sequence group, where the historical consulting business volume sequences in the historical consulting business volume sequence group correspond to consulting categories; determining whether a prediction time period meets the business change point detection condition; for each historical consulting business volume sequence in the historical consulting business volume sequence group, performing the following steps: in response to determining that the prediction time period meets the business change point detection condition, performing business change point processing on the historical consulting business volume sequence to obtain the historical consulting business volume sequence after business change point processing as a sample sequence; determining the consulting category corresponding to the historical consulting business volume sequence as the target consulting category; generating a predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence.

[0008] Optionally, before generating a predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence, the method further includes: in response to determining that the prediction time period does not meet the business change point detection condition, determining the historical consulting business volume sequence as the sample sequence.

[0009] Optionally, performing business change point processing on the historical consulting business volume sequence to obtain the historical consulting business volume sequence after business change point processing as a sample sequence includes: determining a target business change point according to the historical consulting business volume sequence; determining each historical consulting business volume arranged after the target business change point in the historical consulting business volume sequence as the sample sequence.

[0010] Optionally, determining a target business change point according to the historical consulting business volume sequence includes: determining whether there is a consulting business volume in the historical consulting business volume sequence that meets the abnormal change condition; in response to determining that there is a historical consulting business volume in the historical consulting business volume sequence that meets the abnormal change condition, determining each historical consulting business volume that meets the abnormal change condition as an abnormal consulting business volume set; selecting an abnormal consulting business volume that meets the preset time sequence condition from the abnormal consulting business volume set as the target business change point.

[0011] Optionally, generating a predicted consulting business volume sequence corresponding to the predicted time period and the target consulting category according to the above sample sequence includes: adjusting each first model parameter of the initial first consulting business volume prediction model according to the above sample sequence to train the first consulting business volume prediction model corresponding to the predicted time period and the target consulting category; adjusting each second model parameter of the initial second consulting business volume prediction model according to the above sample sequence to train the second consulting business volume prediction model corresponding to the predicted time period and the target consulting category; generating a first predicted consulting business volume sequence corresponding to the predicted time period and the target consulting category according to the first consulting business volume prediction model and the above sample sequence; generating a second predicted consulting business volume sequence corresponding to the predicted time period and the target consulting category according to the second consulting business volume prediction model and the above sample sequence; generating a predicted consulting business volume sequence corresponding to the predicted time period and the target consulting category according to the first predicted consulting business volume sequence and the second predicted consulting business volume sequence.

[0012] Optionally, adjusting each first model parameter of the initial first consulting business volume prediction model according to the above sample sequence to train the first consulting business volume prediction model corresponding to the predicted time period and the target consulting category includes: based on the initial first consulting business volume prediction model, performing the following parameter optimization steps: generating a parameter verification information set corresponding to the first parameter space through the above sample sequence and the initial first consulting business volume prediction model, where the parameter verification information in the parameter verification information set includes accuracy and uncertainty information, and the first parameter space corresponds to each of the first model parameters; determining each first model adjustment parameter corresponding to each of the first model parameters according to the parameter verification information set; updating the initial first consulting business volume prediction model according to each first model adjustment parameter; generating model verification information according to the updated initial first consulting business volume prediction model and the above sample sequence; in response to determining that the model verification information meets the preset convergence condition, determining the updated initial first consulting business volume prediction model as the first consulting business volume prediction model.

[0013] Optionally, the parameter optimization steps further include: in response to determining that the model verification information does not meet the preset convergence condition, performing the above parameter optimization steps again according to the updated initial first consulting business volume prediction model and the adjusted first parameter space.

[0014] Optionally, adjusting each second model parameter of the initial second consulting business volume prediction model according to the above sample sequence to train a second consulting business volume prediction model corresponding to the above prediction time period and the above target consulting category includes: constructing each second consulting business volume prediction model to be screened through a preset second parameter space corresponding to each second model parameter; for each second consulting business volume prediction model to be screened constructed, generating model metric information corresponding to the second consulting business volume prediction model to be screened according to the above sample sequence; and selecting, from each of the second consulting business volume prediction models to be screened, the second consulting business volume prediction model to be screened whose corresponding model metric information meets the preset model metric conditions as the second consulting business volume prediction model.

[0015] Optionally, generating a predicted consulting business volume sequence corresponding to the above prediction time period and the above target consulting category according to the above first predicted consulting business volume sequence and the above second predicted consulting business volume sequence includes: performing a harmonic mean process on the above first predicted consulting business volume sequence and the above second predicted consulting business volume sequence to obtain a predicted consulting business volume sequence corresponding to the above prediction time period and the above target consulting category.

[0016] Optionally, the method further includes: for each predicted consulting business volume sequence generated, performing the following steps: obtaining a set of schedulable customer service user information corresponding to the consulting category of the above predicted consulting business volume sequence; generating a total unit business processing volume according to the above set of schedulable customer service user information; for each predicted consulting business volume in the above predicted consulting business volume set, performing the following steps: in response to determining that the above predicted consulting business volume is greater than the above total unit business processing volume, determining the difference between the above predicted consulting business volume and the above total unit business processing volume as the total unit volume to be processed; determining the above set of schedulable customer service user information and the above total unit volume to be processed as customer service user scheduling information; and generating an initial customer service user scheduling table according to each determined customer service user scheduling information.

[0017] Optionally, after generating the initial customer service user scheduling table according to each determined customer service user scheduling information, the method further includes: displaying the above initial customer service user scheduling table, where each customer service user scheduling information corresponds to a table row of the above initial customer service user scheduling table; in response to detecting an external customer service user information input operation acting on any table row including the total unit volume to be processed, adding each input external customer service user information to the above any table row; and in response to detecting a save operation acting on the above initial customer service user scheduling table, determining the initial customer service user scheduling table added with each external customer service user information as the customer service user scheduling table.

[0018] Second aspect, some embodiments of the present disclosure provide a consulting business volume prediction device, the device includes: an acquisition unit configured to acquire a historical consulting business volume sequence group, wherein the historical consulting business volume sequences in the historical consulting business volume sequence group correspond to consulting categories; a determination unit configured to determine whether a prediction time period satisfies a business change point detection condition; an execution unit configured to, for each historical consulting business volume sequence in the historical consulting business volume sequence group, perform the following steps: in response to determining that the prediction time period satisfies the business change point detection condition, perform business change point processing on the historical consulting business volume sequence to obtain a historical consulting business volume sequence after business change point processing as a sample sequence; determine the consulting category corresponding to the historical consulting business volume sequence as a target consulting category; generate a predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence.

[0019] Optionally, before generating the predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence, the consulting business volume prediction device further includes: a sample sequence determination unit configured to, in response to determining that the prediction time period does not satisfy the business change point detection condition, determine the historical consulting business volume sequence as the sample sequence.

[0020] Optionally, the execution unit is further configured to: determine a target business change point according to the historical consulting business volume sequence; determine each historical consulting business volume arranged after the target business change point in the historical consulting business volume sequence as the sample sequence.

[0021] Optionally, the execution unit is further configured to: determine whether there is a historical consulting business volume in the historical consulting business volume sequence that satisfies an abnormal change condition; in response to determining that there is a historical consulting business volume in the historical consulting business volume sequence that satisfies the abnormal change condition, determine each historical consulting business volume that satisfies the abnormal change condition as an abnormal consulting business volume set; select an abnormal consulting business volume that satisfies a preset time sequence condition from the abnormal consulting business volume set as the target business change point.

[0022] Optionally, the execution unit is further configured to: according to the above sample sequence, adjust each first model parameter of the initial first consultation traffic prediction model to train the first consultation traffic prediction model corresponding to the above prediction time period and the above target consultation category; according to the above sample sequence, adjust each second model parameter of the initial second consultation traffic prediction model to train the second consultation traffic prediction model corresponding to the above prediction time period and the above target consultation category; according to the above first consultation traffic prediction model and the above sample sequence, generate a first predicted consultation traffic sequence corresponding to the above prediction time period and the above target consultation category; according to the above second consultation traffic prediction model and the above sample sequence, generate a second predicted consultation traffic sequence corresponding to the above prediction time period and the above target consultation category; according to the above first predicted consultation traffic sequence and the above second predicted consultation traffic sequence, generate a predicted consultation traffic sequence corresponding to the above prediction time period and the above target consultation category.

[0023] Optionally, the execution unit is further configured to: based on the initial first consultation traffic prediction model, perform the following parameter optimization steps: generate a set of parameter verification information corresponding to the first parameter space through the above sample sequence and the initial first consultation traffic prediction model, where the parameter verification information in the above set of parameter verification information includes accuracy and uncertainty information, and the above first parameter space corresponds to each of the above first model parameters; determine each first model adjustment parameter corresponding to each of the above first model parameters according to the set of parameter verification information; update the initial first consultation traffic prediction model according to each first model adjustment parameter; generate model verification information according to the updated initial first consultation traffic prediction model and the above sample sequence; in response to determining that the model verification information meets the preset convergence condition, determine the updated initial first consultation traffic prediction model as the first consultation traffic prediction model.

[0024] Optionally, the above parameter optimization steps further include: in response to determining that the model verification information does not meet the above preset convergence condition, perform the above parameter optimization steps again according to the updated initial first consultation traffic prediction model and the adjusted first parameter space.

[0025] Optionally, the execution unit is further configured to: construct each to-be-screened second consultation traffic prediction model through a preset second parameter space corresponding to each of the above second model parameters; for each constructed to-be-screened second consultation traffic prediction model, generate model index information corresponding to the to-be-screened second consultation traffic prediction model according to the above sample sequence; select the to-be-screened second consultation traffic prediction model whose corresponding model index information meets the preset model index condition from the above to-be-screened second consultation traffic prediction models as the second consultation traffic prediction model.

[0026] Optionally, the execution unit is further configured to: perform a harmonic mean process on the first predicted consultation traffic volume sequence and the second predicted consultation traffic volume sequence to obtain a predicted consultation traffic volume sequence corresponding to the predicted time period and the target consultation category.

[0027] Optionally, the consultation traffic volume prediction device further includes: a step execution unit configured to, for each generated predicted consultation traffic volume sequence, perform the following steps: obtain a set of schedulable customer service user information corresponding to the consultation category of the predicted consultation traffic volume sequence; generate a total unit business processing volume according to the set of schedulable customer service user information; for each predicted consultation traffic volume in the predicted consultation traffic volume set, perform the following steps: in response to determining that the predicted consultation traffic volume is greater than the total unit business processing volume, determine the difference between the predicted consultation traffic volume and the total unit business processing volume as the total unit to-be-processed volume; determine the set of schedulable customer service user information and the total unit to-be-processed volume as customer service user scheduling information; and generate an initial customer service user scheduling table according to the determined customer service user scheduling information.

[0028] Optionally, after generating the initial customer service user scheduling table according to the determined customer service user scheduling information, the step execution unit further includes: a display unit, an adding unit, and a customer service user scheduling table determination unit. The display unit is configured to display the initial customer service user scheduling table, where each customer service user scheduling information corresponds to a table row of the initial customer service user scheduling table. The adding unit is configured to, in response to detecting an external customer service user information input operation on any table row including the total unit to-be-processed volume, add the input external customer service user information to the any table row. The customer service user scheduling table determination unit is configured to, in response to detecting a save operation on the initial customer service user scheduling table, determine the initial customer service user scheduling table added with the external customer service user information as the customer service user scheduling table.

[0029] In a third aspect, some embodiments of the present disclosure provide an electronic device, including: one or more processors; a storage device storing one or more programs thereon, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the first aspect.

[0030] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium storing a computer program thereon, where the program, when executed by a processor, implements the method described in any implementation manner of the first aspect.

[0031] Fifth aspect, some embodiments of the present disclosure provide a computer program product, including a computer program, which when executed by a processor, implements the method described in any implementation manner of the above first aspect.

[0032] The above embodiments of the present disclosure have the following beneficial effects: Through the consultation business volume prediction method of some embodiments of the present disclosure, the comprehensiveness of covered services, the prediction efficiency of consultation business volume, and the accuracy of predicted consultation business volume are improved. Specifically, the reasons for the poor comprehensiveness of covered services, the low prediction efficiency of consultation business volume, and the low accuracy of predicted consultation business volume are as follows: When predicting the consultation business volume for services with a large business volume, the comprehensiveness of covered services is poor, and data needs to be manually summarized and analyzed at regular intervals, resulting in a large manual workload and low prediction efficiency of consultation business volume. In addition, there are many internal data calibers. When the caliber of manually exported data is incorrect, the accuracy of predicted consultation business volume is low. Based on this, the consultation business volume prediction method of some embodiments of the present disclosure first obtains a historical consultation business volume sequence group. Among them, the historical consultation business volume sequences in the above historical consultation business volume sequence group correspond to consultation categories. Thus, the time series of historical consultation business volumes of different consultation categories can be uniformly obtained. Then, it is determined whether the prediction time period meets the service change point detection condition. Thus, it can be pre-determined whether it is necessary to perform service change point detection on each historical consultation business volume sequence when predicting the consultation business volume in the future time period. After that, for each historical consultation business volume sequence in the above historical consultation business volume sequence group, the following steps are executed: First step, in response to determining that the above prediction time period meets the above service change point detection condition, perform service change point processing on the above historical consultation business volume sequence to obtain the historical consultation business volume sequence after service change point processing as the sample sequence. Thus, when the prediction time period meets the above service change point detection condition, service change point detection can be performed on the historical consultation business volume sequence to generate a sample sequence for predicting the consultation business volume. Second step, determine the consultation category corresponding to the above historical consultation business volume sequence as the target consultation category. Thus, the consultation category of the currently polled consultation business volume sequence can be determined. Third step, according to the above sample sequence, generate a predicted consultation business volume sequence corresponding to the above prediction time period and the above target consultation category. Thus, the time series of the consultation business volume of the target consultation category within the prediction time period can be automatically predicted based on the determined sample sequence. Also, because each historical consultation business volume in the historical consultation business volume sequence is uniformly obtained, the caliber error of data acquisition can be avoided, thereby improving the accuracy of predicted consultation business volume. Also, because the consultation business volume is predicted by consultation category, the consultation business volume of different consultation categories can be automatically predicted, thereby improving the comprehensiveness of covered services. Also, because the acquisition and processing of historical consultation business volume sequences and the prediction of consultation business volume are all automatically executed without manual participation, the prediction efficiency of consultation business volume is improved. Thus, the comprehensiveness of covered services, the prediction efficiency of consultation business volume, and the accuracy of predicted consultation business volume are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic and the elements and elements are not necessarily drawn to scale.

[0034] Figure 1 is a schematic diagram of an application scenario of a consultation traffic prediction method according to some embodiments of the present disclosure;

[0035] Figure 2 is a schematic diagram of another application scenario of a consultation traffic prediction method according to some embodiments of the present disclosure;

[0036] Figure 3 is a flowchart of some embodiments of a consultation traffic prediction method according to the present disclosure;

[0037] Figure 4 is a flowchart of some other embodiments of a consultation traffic prediction method according to the present disclosure;

[0038] Figure 5 is a schematic structural diagram of some embodiments of a consultation traffic prediction device according to the present disclosure;

[0039] Figure 6 is a schematic structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure. Specific Embodiments

[0040] The embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0041] In addition, it should be noted that for the sake of convenience of description, only the parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present disclosure and the features in the embodiments can be combined with each other.

[0042] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence relationship of the functions performed by these devices, modules or units.

[0043] It should be noted that the modifiers "one" and "multiple" mentioned in this disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless clearly specified otherwise in the context, it should be understood as "one or more".

[0044] The names of the messages or information exchanged between multiple devices in the embodiments of this disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0045] Regarding the operations of collecting, storing, using, etc. of the user's personal information (such as dispatchable customer service user information) involved in this disclosure, before performing the corresponding operations, relevant organizations or individuals shall fulfill obligations including conducting a personal information security impact assessment, fulfilling the obligation of notification to the personal information subject, and obtaining the prior authorization and consent of the personal information subject.

[0046] The following will detail this disclosure with reference to the accompanying drawings and in conjunction with embodiments.

[0047] Figure 1 It is a schematic diagram of an application scenario of a consultation business volume prediction method according to some embodiments of this disclosure.

[0048] In Figure 1 the application scenario, first, the computing device 101 can obtain the historical consultation business volume sequence group 102. Among them, the historical consultation business volume sequences in the above historical consultation business volume sequence group 102 correspond to consultation categories. Then, the computing device 101 can determine whether the prediction time period 103 meets the business change point detection condition. After that, for each historical consultation business volume sequence (such as the historical consultation business volume sequence 1021) in the above historical consultation business volume sequence group 102, the computing device 101 can perform the following steps: First step, in response to determining that the above prediction time period 103 meets the above business change point detection condition, perform business change point processing on the above historical consultation business volume sequence 1021 to obtain the historical consultation business volume sequence after business change point processing as the sample sequence 104. Second step, determine the consultation category corresponding to the above historical consultation business volume sequence 1021 as the target consultation category 105. Third step, according to the above sample sequence 104, generate a predicted consultation business volume sequence 106 corresponding to the above prediction time period 103 and the above target consultation category 105.

[0049] It should be noted that the above computing device 101 can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster composed of multiple servers or terminal devices, or as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the above-listed hardware devices. It can be implemented as, for example, multiple software or software modules for providing distributed services, or as a single software or software module. No specific limitation is made here.

[0050] It should be understood that Figure 1 the number of computing devices in

[0051] Figure 2 is a schematic diagram of another application scenario of the consulting business volume prediction method according to some embodiments of the present disclosure.

[0052] In Figure 2In the application scenario, first, the computing device 101 can obtain the historical consultation traffic volume sequence group 102. Among them, the historical consultation traffic volume sequences in the historical consultation traffic volume sequence group 102 correspond to consultation categories. Then, the computing device 101 can determine whether the prediction time period 103 meets the service change point detection condition. After that, for each historical consultation traffic volume sequence (such as the historical consultation traffic volume sequence 1021) in the historical consultation traffic volume sequence group 102, the computing device 101 can perform the following steps: First step, in response to determining that the prediction time period 103 meets the service change point detection condition, perform service change point processing on the historical consultation traffic volume sequence 1021 to obtain the historical consultation traffic volume sequence after service change point processing as the sample sequence 104. Second step, determine the consultation category corresponding to the historical consultation traffic volume sequence 1021 as the target consultation category 105. Third step, according to the sample sequence 104, adjust each first model parameter of the initial first consultation traffic volume prediction model 107 to train the first consultation traffic volume prediction model 108 corresponding to the prediction time period 103 and the target consultation category 105. Fourth step, according to the sample sequence 104, adjust each second model parameter of the initial second consultation traffic volume prediction model 109 to train the second consultation traffic volume prediction model 110 corresponding to the prediction time period 103 and the target consultation category 105. Fifth step, according to the first consultation traffic volume prediction model 108 and the sample sequence 104, generate the first predicted consultation traffic volume sequence 111 corresponding to the prediction time period 103 and the target consultation category 105. Sixth step, according to the second consultation traffic volume prediction model 110 and the sample sequence 104, generate the second predicted consultation traffic volume sequence 112 corresponding to the prediction time period 103 and the target consultation category 105. Seventh step, according to the first predicted consultation traffic volume sequence 111 and the second predicted consultation traffic volume sequence 112, generate the predicted consultation traffic volume sequence 106 corresponding to the prediction time period 103 and the target consultation category 105.

[0053] It should be noted that the above computing device 101 can be hardware or software. When the computing device is hardware, it can be implemented as a distributed cluster composed of multiple servers or terminal devices, or can be implemented as a single server or a single terminal device. When the computing device is embodied as software, it can be installed in the above-listed hardware devices. It can be implemented as, for example, multiple software or software modules for providing distributed services, or can be implemented as a single software or software module. No specific limitation is made here.

[0054] It should be understood that Figure 2 the number of computing devices in

[0055] Continue to refer to Figure 3 , which shows the process 300 of some embodiments of the consulting business volume prediction method according to the present disclosure. The consulting business volume prediction method includes the following steps:

[0056] Step 301, obtain a historical consulting business volume sequence group.

[0057] In some embodiments, the execution subject of the consulting business volume prediction method (such as Figure 1 the computing device shown) can obtain the historical consulting business volume sequence group from the database through a wired connection or a wireless connection. Among them, the historical consulting business volume sequence group can be each historical consulting business volume sequence corresponding to each consulting category. Each historical consulting business volume sequence in the above historical consulting business volume sequence group corresponds to a consulting category. The consulting category can be the category corresponding to the customer service business. For example, each consulting category can include, but is not limited to: beauty products category, clothing category, book category, electronic product category. The historical consulting business volume sequence can be the time sequence of the historical consulting business volume within a preset historical time period. There is no limitation on the specific setting of the preset historical time period. For example, the preset historical time period can be the historical time period corresponding to the past 3 years. Each historical consulting business volume in the historical consulting business volume sequence can be the historical consulting business volume at the granularity of days. The historical consulting business volume can be the customer service consultation volume received in one day of historical time.

[0058] It should be noted that the above wireless connection method can include, but is not limited to, 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other currently known or future-developed wireless connection methods.

[0059] Step 302, determine whether the prediction time period meets the business change point detection condition.

[0060] In some embodiments, the above execution subject can determine whether the prediction time period meets the business change point detection condition. Among them, the above prediction time period can be the future time period for which the consulting business volume needs to be predicted. The above business change point detection condition can be that the month corresponding to the above prediction time period does not include the value reduction month. The value reduction month can be the month in which the transfer value of the item can be reduced in the current month (for example, the value reduction months can be the big promotion months of June and November).

[0061] Step 303, for each historical consulting business volume sequence in the historical consulting business volume sequence group, perform the following steps:

[0062] Step 3031, in response to determining that the prediction time period meets the service change point detection condition, perform service change point processing on the historical consultation business volume sequence to obtain the historical consultation business volume sequence after service change point processing as the sample sequence.

[0063] In some embodiments, the above-mentioned execution entity may, in response to determining that the above-mentioned prediction time period meets the above-mentioned service change point detection condition, perform service change point processing on the above-mentioned historical consultation business volume sequence to obtain the historical consultation business volume sequence after service change point processing as the sample sequence. In practice, the above-mentioned execution entity may, in response to determining that there are no abnormal points in the above-mentioned consultation business volume sequence, determine the above-mentioned historical consultation business volume sequence as the sample sequence. An abnormal point may be the historical consultation business volume with abnormal distribution determined by an anomaly detection algorithm. For example, the anomaly detection algorithm may be a difference detection algorithm.

[0064] In some optional implementation manners of some embodiments, the above-mentioned execution entity may perform service change point processing on the above-mentioned historical consultation business volume sequence through the following steps to obtain the historical consultation business volume sequence after service change point processing as the sample sequence:

[0065] The first step is to determine the target service change point according to the above-mentioned historical consultation business volume sequence.

[0066] The second step is to determine each historical consultation business volume arranged after the above-mentioned target service change point in the above-mentioned historical consultation business volume sequence as the sample sequence. Thus, each historical consultation business volume after the service change point can be intercepted as the sample sequence to eliminate the influence of abnormal data before the service change point.

[0067] In some optional implementation manners of some embodiments, the above-mentioned execution entity may determine the target service change point according to the above-mentioned historical consultation business volume sequence through the following steps:

[0068] The first step is to determine whether there is a historical consultation business volume in the above-mentioned historical consultation business volume sequence that meets the abnormal change condition. Among them, the above-mentioned abnormal change condition may be that the historical consultation business volume is an outlier in the above-mentioned historical consultation business volume sequence. An outlier may be the abnormal historical consultation business volume determined by a change point detection method.

[0069] The second step is to, in response to determining that there is a historical consultation business volume in the above-mentioned historical consultation business volume sequence that meets the above-mentioned abnormal change condition, determine each historical consultation business volume that meets the above-mentioned abnormal change condition as the abnormal consultation business volume set.

[0070] Step 3: Select the abnormal consultation traffic volume that meets the preset time series condition from the above abnormal consultation traffic volume set as the service change point. Among them, the above preset time series condition can be that the time corresponding to the abnormal consultation traffic volume is the latest. Thus, the latest abnormal consultation traffic volume can be used as the target service change point to eliminate the influence of the abnormal data before the latest abnormal consultation traffic volume.

[0071] Step 3032: Determine the consultation category corresponding to the historical consultation traffic volume sequence as the target consultation category.

[0072] In some embodiments, the above execution subject may determine the consultation category corresponding to the above historical consultation traffic volume sequence as the target consultation category.

[0073] Optionally, before step 3033, the above execution subject may also, in response to determining that the above prediction time period does not meet the above service change point detection condition, determine the above historical consultation traffic volume sequence as the sample sequence. Thus, when the prediction time period does not meet the service change point detection condition, the historical consultation traffic volume sequence can be directly used as the sample sequence.

[0074] Step 3033: Generate a predicted consultation traffic volume sequence corresponding to the prediction time period and the target consultation category according to the sample sequence.

[0075] In some embodiments, the above execution subject may generate a predicted consultation traffic volume sequence corresponding to the above prediction time period and the above target consultation category according to the above sample sequence. In practice, the above execution subject may input the above sample sequence into a consultation traffic volume prediction model corresponding to the above prediction time period and the above target consultation category that has been pre-trained to obtain a predicted consultation traffic volume sequence corresponding to the above prediction time period and the above target consultation category. The consultation traffic volume prediction model can be a neural network model. For example, the consultation traffic volume prediction model can be a long short-term memory network (LSTM, Long Short-Term Memory).

[0076] Optionally, for each generated predicted consultation traffic volume sequence, the above execution subject may perform the following steps:

[0077] Step 1: Obtain a set of schedulable customer service user information corresponding to the above-mentioned predicted consultation business volume sequence for the consultation category. Among them, the above-mentioned set of schedulable customer service user information can be the user-related information of each customer service user corresponding to the above-mentioned consultation category. The schedulable customer service user information in the set of schedulable customer service user information can include, but is not limited to, at least one of the following: customer service user identifier, customer service user nickname, consultation category, and daily business processing volume. The daily business processing volume can be the total amount of consultation business that a customer service user can handle in a day. In practice, the above-mentioned execution entity can obtain the set of schedulable customer service user information corresponding to the above-mentioned predicted consultation business volume sequence for the consultation category from a database.

[0078] Step 2: Generate a total unit business processing volume based on the above-mentioned set of schedulable customer service user information. In practice, the above-mentioned execution entity can determine the sum of the daily business processing volumes included in the above-mentioned set of schedulable customer service user information as the total unit business processing volume.

[0079] Step 3: For each predicted consultation business volume in the above-mentioned predicted consultation business volume set, perform the following steps:

[0080] First sub-step: In response to determining that the above-mentioned predicted consultation business volume is greater than the above-mentioned total unit business processing volume, determine the difference between the above-mentioned predicted consultation business volume and the above-mentioned total unit business processing volume as the total unit to-be-processed volume. The total unit to-be-processed volume can represent the consultation business volume within a single day for which additional customer service users need to be supplemented for processing.

[0081] Second sub-step: Determine the above-mentioned set of schedulable customer service user information and the above-mentioned total unit to-be-processed volume as customer service user scheduling information.

[0082] Optionally, the above-mentioned execution entity can also, in response to determining that the above-mentioned predicted consultation business volume is less than or equal to the above-mentioned total unit business processing volume, select each piece of schedulable customer service user information that meets the preset consultation business processing conditions from the above-mentioned set of schedulable customer service user information as the customer service user scheduling information. The preset consultation business processing conditions can be that the sum of the daily business processing volumes included in each piece of selected schedulable customer service user information is greater than or equal to the above-mentioned predicted consultation business volume, and the number of each piece of selected schedulable customer service user information is the least.

[0083] Step 4: Generate an initial customer service user scheduling table based on the determined customer service user scheduling information. In practice, the above-mentioned execution entity can store the determined customer service user scheduling information in an empty table to obtain the initial customer service user scheduling table. The initial customer service user scheduling table can include a schedulable customer service user information field, a total unit to-be-processed volume field, and a predicted date field. The predicted date field can represent the future time corresponding to the predicted consultation business volume. Thus, an initial customer service user scheduling table that needs to be further scheduled can be automatically generated.

[0084] Optionally, after generating the initial customer service user scheduling table according to the determined various customer service user scheduling information, the execution entity may further perform the following steps:

[0085] First, display the initial customer service user scheduling table. Each customer service user scheduling information corresponds to a table row of the initial customer service user scheduling table.

[0086] Second, in response to detecting an external customer service user information input operation acting on any table row including the total amount of units to be processed, add the input external customer service user information to the any table row. The external customer service user information input operation may be an operation of inputting each external customer service user information for the total amount of units to be processed to make up the total amount of units to be processed. For example, an external customer service user information field may be added to input the external customer service user information in a newly added cell within the any table row. The external customer service user information may be the user information of other consultative type customer service users that can be scheduled, or the user information of customer service users to be newly hired corresponding to the same consultative type as the set of schedulable customer service user information.

[0087] Third, in response to detecting a save operation acting on the initial customer service user scheduling table, determine the initial customer service user scheduling table with the added external customer service user information as the customer service user scheduling table. Thus, the dispatcher can set the customer service users to be scheduled through the visual table by himself.

[0088] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: Through the consultation business volume prediction method of some embodiments of the present disclosure, the comprehensiveness of covered services, the prediction efficiency of consultation business volume, and the accuracy of predicted consultation business volume are improved. Specifically, the reasons for the poor comprehensiveness of covered services, the low prediction efficiency of consultation business volume, and the low accuracy of predicted consultation business volume are as follows: When predicting the consultation business volume for services with a large business volume, the comprehensiveness of covered services is poor, and data needs to be manually summarized and analyzed at regular intervals, resulting in a large manual workload and low prediction efficiency of consultation business volume. In addition, there are many internal data calibers. When the caliber of manually exported data is incorrect, the accuracy of predicted consultation business volume is low. Based on this, in the consultation business volume prediction method of some embodiments of the present disclosure, first, a historical consultation business volume sequence group is obtained. Among them, the historical consultation business volume sequences in the above-mentioned historical consultation business volume sequence group correspond to consultation categories. Thus, the time series of historical consultation business volumes of different consultation categories can be uniformly obtained. Then, it is determined whether the prediction time period meets the service change point detection condition. Thus, it can be pre-confirmed whether it is necessary to perform service change point detection on each historical consultation business volume sequence when predicting the consultation business volume in the future time period. After that, for each historical consultation business volume sequence in the above-mentioned historical consultation business volume sequence group, the following steps are executed: The first step, in response to determining that the above-mentioned prediction time period meets the above-mentioned service change point detection condition, perform service change point processing on the above-mentioned historical consultation business volume sequence to obtain the historical consultation business volume sequence after service change point processing as a sample sequence. Thus, when the prediction time period meets the above-mentioned service change point detection condition, service change point detection can be performed on the historical consultation business volume sequence to generate a sample sequence for predicting the consultation business volume. The second step, determine the consultation category corresponding to the above-mentioned historical consultation business volume sequence as the target consultation category. Thus, the consultation category of the currently polled consultation business volume sequence can be determined. The third step, according to the above-mentioned sample sequence, generate a predicted consultation business volume sequence corresponding to the above-mentioned prediction time period and the above-mentioned target consultation category. Thus, the time series of the consultation business volume of the target consultation category within the prediction time period can be automatically predicted according to the determined sample sequence. Also because the historical consultation business volumes in the historical consultation business volume sequence are uniformly obtained, the caliber error of data acquisition can be avoided, thereby improving the accuracy of predicted consultation business volume. Also because the consultation business volume is predicted by consultation category, the consultation business volume of different consultation categories can be automatically predicted, thereby improving the comprehensiveness of covered services. Also because the acquisition and processing of the historical consultation business volume sequence and the prediction of the consultation business volume are all automatically executed without manual participation, the prediction efficiency of the consultation business volume is improved. Thus, the comprehensiveness of covered services, the prediction efficiency of consultation business volume, and the accuracy of predicted consultation business volume are improved.

[0089] Further referenceFigure 4 , which shows the process 400 of some other embodiments of the consulting business volume prediction method. The process 400 of the consulting business volume prediction method includes the following steps:

[0090] Step 401, obtain the historical consulting business volume sequence group.

[0091] Step 402, determine whether the prediction time period meets the business change point detection condition.

[0092] Step 403, for each historical consulting business volume sequence in the historical consulting business volume sequence group, perform the following steps:

[0093] Step 4031, in response to determining that the prediction time period meets the business change point detection condition, perform business change point processing on the historical consulting business volume sequence to obtain the historical consulting business volume sequence after business change point processing as the sample sequence.

[0094] Step 4032, determine the consulting category corresponding to the historical consulting business volume sequence as the target consulting category.

[0095] In some embodiments, the specific implementation of steps 401-4032 and the technical effects brought can refer to Figure 3 Steps 301-3032 in the corresponding embodiments, which will not be elaborated here.

[0096] Step 4033, according to the sample sequence, adjust each first model parameter of the initial first consulting business volume prediction model to train the first consulting business volume prediction model corresponding to the prediction time period and the target consulting category.

[0097] In some embodiments, the execution subject of the consulting business volume prediction method (such as Figure 1 or Figure 2 the computing device shown) can adjust each first model parameter of the initial first consulting business volume prediction model according to the above sample sequence to train the first consulting business volume prediction model corresponding to the above prediction time period and the above target consulting category.

[0098] In some optional implementation manners of some embodiments, the above execution subject can adjust each first model parameter of the initial first consulting business volume prediction model according to the above sample sequence to train the first consulting business volume prediction model corresponding to the above prediction time period and the above target consulting category through the following steps:

[0099] Based on the initial first consulting business volume prediction model, perform the following parameter optimization steps:

[0100] First step, based on the above sample sequence and the initial first consulting business volume prediction model, generate a set of parameter verification information corresponding to the first parameter space. Among them, the parameter verification information in the above set of parameter verification information includes accuracy and uncertainty information. The above first parameter space corresponds to each of the above first model parameters. When the parameter optimization step is executed for the first time, the first parameter space can be the value range of each initialized first model parameter. When the parameter optimization step is executed subsequently, the first parameter space can be the value range of each first model parameter adjusted during the previous execution of the parameter optimization step. Each of the first model parameters of the initial first consulting business volume prediction model can be preset. The model type of the initial first consulting business volume prediction model can be prophet. Each of the first model parameters can include, but is not limited to: the adjustment intensity of the seasonal model, the adjustment intensity of the holiday component model, and the flexibility adjustment parameter of the automatic potential change point. The greater the adjustment intensity of the seasonal model, the more adaptable it is to larger seasonal fluctuations. The smaller the adjustment intensity of the seasonal model, the more it suppresses seasonality. The larger the flexibility adjustment parameter, the more potential change points will be allowed. The smaller the flexibility adjustment parameter, the fewer potential change points will be allowed. Uncertainty information can characterize the uncertainty of the predicted consulting business volume. For example, the uncertainty information can be used as the variance of a Gaussian distribution.

[0101] In practice, the above execution entity can generate all the parameter verification information in the above first parameter space as a set of parameter verification information through Gaussian process regression. Each parameter verification information can correspond to a set of first model parameters.

[0102] Second step, based on the set of parameter verification information, determine each first model adjustment parameter corresponding to each of the above first model parameters. In practice, the above execution entity can solve for an optimal set of first model parameters as each first model adjustment parameter through the set of parameter verification information, the sets of first model parameters corresponding to the set of parameter verification information, and an acquisition function. The acquisition function can be UCB or EI. In practice, the above execution entity can solve for an optimal set of first model parameters as each first model adjustment parameter through a solver.

[0103] Third step, update the initial first consulting business volume prediction model according to each first model adjustment parameter. In practice, the above execution entity can set each of the first model parameters of the initial first consulting business volume prediction model to each first model adjustment parameter respectively to update the initial first consulting business volume prediction model.

[0104] Fourthly, according to the updated initial first consultation business volume prediction model and the above sample sequence, generate model verification information. In practice, the above-mentioned execution entity can use the verification set included in the above sample sequence to perform model verification on the updated initial first consultation business volume prediction model to obtain model verification information. The model verification information may include accuracy rate.

[0105] Fifthly, in response to determining that the model verification information meets the preset convergence condition, determine the updated initial first consultation business volume prediction model as the first consultation business volume prediction model. Among them, the preset convergence condition may be that the accuracy rate is greater than the accuracy rate threshold.

[0106] Optionally, the above parameter optimization step may further include: Sixthly, in response to determining that the model verification information does not meet the above preset convergence condition, according to the updated initial first consultation business volume prediction model and the adjusted first parameter space, execute the above parameter optimization step again. Thus, the parameter optimization step can be iterated to realize the automatic adjustment of the parameters of the first consultation business volume prediction model.

[0107] Step 4034, according to the sample sequence, adjust each second model parameter of the initial second consultation business volume prediction model to train the second consultation business volume prediction model corresponding to the prediction time period and the target consultation category.

[0108] In some embodiments, the above-mentioned execution entity can adjust each second model parameter of the initial second consultation business volume prediction model according to the above sample sequence to train the second consultation business volume prediction model corresponding to the above prediction time period and the above target consultation category.

[0109] In some optional implementation manners of some embodiments, the above-mentioned execution entity can adjust each second model parameter of the initial second consultation business volume prediction model according to the above sample sequence to train the second consultation business volume prediction model corresponding to the above prediction time period and the above target consultation category through the following steps, including:

[0110] Step 1: Construct each second consulting business volume prediction model to be screened through a preset second parameter space corresponding to the above-mentioned second model parameters. The second parameter space can be the value range of each second model parameter set in advance. Each second model parameter of the initial second consulting business volume prediction model can be set in advance. The model type of the initial second consulting business volume prediction model can be sarima (Seasonal Autoregressive Integrated Moving Average). Each second model parameter can include but is not limited to: non-seasonal autoregressive order, non-seasonal differencing order, non-seasonal moving average order, seasonal autoregressive order, seasonal differencing order, and seasonal moving average order. In practice, the above-mentioned execution entity can set each group of second model parameters for the initial second consulting business volume prediction model within the above-mentioned second parameter space to obtain each initial second consulting business volume prediction model with the second model parameters set as each second consulting business volume prediction model to be screened.

[0111] Step 2: For each second consulting business volume prediction model to be screened constructed above, generate model index information corresponding to the second consulting business volume prediction model to be screened according to the above-mentioned sample sequence. In practice, the above-mentioned execution entity can use the validation set in the above-mentioned sample sequence to verify the second consulting business volume prediction model to be screened to obtain the model index information. The model index information can include but is not limited to: AIC value, BIC value.

[0112] Step 3: Select, from the above-mentioned second consulting business volume prediction models to be screened, the second consulting business volume prediction model whose corresponding model index information meets the preset model index conditions as the second consulting business volume prediction model. For example, the preset model index condition can be the minimum AIC value.

[0113] Step 4035: Generate a first predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the first consulting business volume prediction model and the sample sequence.

[0114] In some embodiments, the above-mentioned execution entity can generate a first predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the above-mentioned first consulting business volume prediction model and the above-mentioned sample sequence. In practice, the above-mentioned execution entity can input the above-mentioned sample sequence into the above-mentioned first consulting business volume prediction model to obtain a first predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category.

[0115] Step 4036: Generate a second predicted consulting business volume sequence corresponding to the prediction time period and the target consulting category according to the second consulting business volume prediction model and the sample sequence.

[0116] In some embodiments, the above-mentioned execution entity may generate a second predicted consultation business volume sequence corresponding to the above-mentioned prediction time period and the above-mentioned target consultation category according to the above-mentioned second consultation business volume prediction model and the above-mentioned sample sequence. In practice, the above-mentioned execution entity may input the above-mentioned sample sequence into the above-mentioned second consultation business volume prediction model to obtain a second predicted consultation business volume sequence corresponding to the above-mentioned prediction time period and the above-mentioned target consultation category.

[0117] Step 4037: Generate a predicted consultation business volume sequence corresponding to the prediction time period and the target consultation category according to the first predicted consultation business volume sequence and the second predicted consultation business volume sequence.

[0118] In some embodiments, the above-mentioned execution entity may generate a predicted consultation business volume sequence corresponding to the above-mentioned prediction time period and the above-mentioned target consultation category according to the above-mentioned first predicted consultation business volume sequence and the above-mentioned second predicted consultation business volume sequence.

[0119] In some alternative implementation manners of some embodiments, the above-mentioned execution entity may perform a harmonic mean processing on the above-mentioned first predicted consultation business volume sequence and the above-mentioned second predicted consultation business volume sequence to obtain a predicted consultation business volume sequence corresponding to the above-mentioned prediction time period and the above-mentioned target consultation category. In practice, first, for each unit time included in the prediction time period, the above-mentioned execution entity may perform the following steps:

[0120] The first step: Determine the reciprocal of the first predicted consultation business volume corresponding to the above-mentioned unit time in the above-mentioned first predicted consultation business volume sequence as the first value. The unit time may be one day.

[0121] The second step: Determine the reciprocal of the second predicted consultation business volume corresponding to the above-mentioned time point in the above-mentioned second predicted consultation business volume sequence as the second value.

[0122] The third step: Determine the sum of the above-mentioned first value and the above-mentioned second value as the third value.

[0123] The fourth step: Determine the ratio of a preset value to the above-mentioned third value as the predicted consultation business volume.

[0124] Then, the above-mentioned execution entity may determine the determined predicted consultation business volumes as the predicted consultation business volume sequence.

[0125] From Figure 4 it can be seen that compared with the description of some corresponding embodiments Figure 3 Figure 4The process 400 of the consulting business volume prediction method in some corresponding embodiments embodies the steps of expanding the first consulting business volume prediction model and the second consulting business volume prediction model. Thus, the solutions described in these embodiments can automatically tune the parameters of the two consulting business volume prediction models and adopt a model fusion method to predict the consulting business volume, improving the prediction efficiency and accuracy.

[0126] Further referring to Figure 5 , as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a consulting business volume prediction device. These device embodiments correspond to Figure 3 the method embodiments shown, and the device can be specifically applied to various electronic devices.

[0127] As Figure 5 shown, the consulting business volume prediction device 500 in some embodiments includes: an acquisition unit 501, a determination unit 502, and an execution unit 503. Among them, the acquisition unit 501 is configured to acquire a historical consulting business volume sequence group, where the historical consulting business volume sequences in the above historical consulting business volume sequence group correspond to consulting categories; the determination unit 502 is configured to determine whether the prediction time period meets the business change point detection condition; the execution unit 503 is configured to, for each historical consulting business volume sequence in the above historical consulting business volume sequence group, perform the following steps: in response to determining that the above prediction time period meets the above business change point detection condition, perform business change point processing on the above historical consulting business volume sequence to obtain the historical consulting business volume sequence after business change point processing as a sample sequence; determine the consulting category corresponding to the above historical consulting business volume sequence as the target consulting category; generate a predicted consulting business volume sequence corresponding to the above prediction time period and the above target consulting category according to the above sample sequence.

[0128] Optionally, before generating the predicted consulting business volume sequence corresponding to the above prediction time period and the above target consulting category according to the above sample sequence, the consulting business volume prediction device 500 may further include: a sample sequence determination unit (not shown in the figure), configured to, in response to determining that the above prediction time period does not meet the above business change point detection condition, determine the above historical consulting business volume sequence as the sample sequence.

[0129] Optionally, the execution unit 503 may be further configured to: determine a target business change point according to the above historical consulting business volume sequence; determine each historical consulting business volume arranged after the above target business change point in the above historical consulting business volume sequence as the sample sequence.

[0130] Optionally, the execution unit 503 may be further configured to: determine whether there is a historical consultation traffic volume in the above historical consultation traffic volume sequence that meets the abnormal change condition; in response to determining that there is a historical consultation traffic volume in the above historical consultation traffic volume sequence that meets the above abnormal change condition, determine each historical consultation traffic volume that meets the above abnormal change condition as an abnormal consultation traffic volume set; select an abnormal consultation traffic volume that meets the preset time series condition from the above abnormal consultation traffic volume set as the target service change point.

[0131] Optionally, the execution unit 503 may be further configured to: adjust each first model parameter of the initial first consultation traffic volume prediction model according to the above sample sequence to train the first consultation traffic volume prediction model corresponding to the above prediction time period and the above target consultation category; adjust each second model parameter of the initial second consultation traffic volume prediction model according to the above sample sequence to train the second consultation traffic volume prediction model corresponding to the above prediction time period and the above target consultation category; generate a first predicted consultation traffic volume sequence corresponding to the above prediction time period and the above target consultation category according to the above first consultation traffic volume prediction model and the above sample sequence; generate a second predicted consultation traffic volume sequence corresponding to the above prediction time period and the above target consultation category according to the above second consultation traffic volume prediction model and the above sample sequence; generate a predicted consultation traffic volume sequence corresponding to the above prediction time period and the above target consultation category according to the above first predicted consultation traffic volume sequence and the above second predicted consultation traffic volume sequence.

[0132] Optionally, the execution unit 503 may be further configured to: based on the initial first consultation traffic volume prediction model, perform the following parameter optimization steps: generate a parameter verification information set corresponding to the first parameter space through the above sample sequence and the initial first consultation traffic volume prediction model, where the parameter verification information in the above parameter verification information set includes accuracy and uncertainty information, and the above first parameter space corresponds to each of the above first model parameters; determine each first model adjustment parameter corresponding to each of the above first model parameters according to the parameter verification information set; update the initial first consultation traffic volume prediction model according to each first model adjustment parameter; generate model verification information according to the updated initial first consultation traffic volume prediction model and the above sample sequence; in response to determining that the model verification information meets the preset convergence condition, determine the updated initial first consultation traffic volume prediction model as the first consultation traffic volume prediction model.

[0133] Optionally, the above parameter optimization steps further include: in response to determining that the model verification information does not meet the above preset convergence condition, perform the above parameter optimization steps again according to the updated initial first consultation traffic volume prediction model and the adjusted first parameter space.

[0134] Optionally, the execution unit 503 may be further configured to: construct each second consultation traffic prediction model to be screened through a preset second parameter space corresponding to each of the above-mentioned second model parameters; for each constructed second consultation traffic prediction model to be screened, generate model metric information corresponding to the second consultation traffic prediction model to be screened according to the above-mentioned sample sequence; select, from each of the above-mentioned second consultation traffic prediction models to be screened, the second consultation traffic prediction model whose corresponding model metric information meets the preset model metric conditions as the second consultation traffic prediction model.

[0135] Optionally, the execution unit 503 may be further configured to: perform a harmonic mean process on the above-mentioned first predicted consultation traffic sequence and the above-mentioned second predicted consultation traffic sequence to obtain a predicted consultation traffic sequence corresponding to the above-mentioned prediction time period and the above-mentioned target consultation category.

[0136] Optionally, the consultation traffic prediction device 500 may further include: a step execution unit (not shown in the figure), configured to perform the following steps for each generated predicted consultation traffic sequence: obtain a set of schedulable customer service user information corresponding to the consultation category of the above-mentioned predicted consultation traffic sequence; generate a total business processing volume per unit according to the above-mentioned set of schedulable customer service user information; for each predicted consultation traffic volume in the above-mentioned predicted consultation traffic volume set, perform the following steps: in response to determining that the above-mentioned predicted consultation traffic volume is greater than the above-mentioned total business processing volume per unit, determine the difference between the above-mentioned predicted consultation traffic volume and the above-mentioned total business processing volume per unit as the total volume to be processed per unit; determine the above-mentioned set of schedulable customer service user information and the above-mentioned total volume to be processed per unit as customer service user scheduling information; generate an initial customer service user scheduling table according to the determined customer service user scheduling information.

[0137] Optionally, after generating the initial customer service user scheduling table according to the determined customer service user scheduling information, the step execution unit may further include: a display unit, an adding unit, and a customer service user scheduling table determination unit (not shown in the figure). The display unit is configured to display the above-mentioned initial customer service user scheduling table, where each customer service user scheduling information corresponds to a table row of the above-mentioned initial customer service user scheduling table. The adding unit is configured to, in response to detecting an external customer service user information input operation acting on any table row including the total volume to be processed per unit, add the input external customer service user information to the above-mentioned any table row. The customer service user scheduling table determination unit is configured to, in response to detecting a save operation acting on the above-mentioned initial customer service user scheduling table, determine the initial customer service user scheduling table added with the external customer service user information as the customer service user scheduling table.

[0138] It can be understood that the various units described in the device 500 and the reference Figure 3corresponds to each step in the described method. Thus, the operations, features, and beneficial effects described above for the method also apply to the apparatus 500 and the units included therein, and will not be elaborated herein.

[0139] Reference is now made to Figure 6 , which shows a schematic structural diagram of an electronic device 600 (e.g., a computing device in Figure 1 ) suitable for use in implementing some embodiments of the present disclosure. Figure 6 The electronic device shown is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0140] As Figure 6 shown, the electronic device 600 may include a processing device 601 (e.g., a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0141] Generally, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or wirelesly to exchange data. Although Figure 6 shows the electronic device 600 having various devices, it should be understood that it is not required to implement or include all the shown devices. Instead, more or fewer devices may be implemented or included. Figure 6 Each block shown in

[0142] In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program includes program code for performing the methods shown in the flowcharts. In such some embodiments, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above functions defined in the methods of some embodiments of the present disclosure are performed.

[0143] It should be noted that the computer-readable medium described in some embodiments of the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can 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 some embodiments of the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0144] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0145] The above computer-readable medium may be included in the above electronic device; or may exist separately without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device is caused to: obtain a historical consultation traffic volume sequence group, wherein the historical consultation traffic volume sequences in the historical consultation traffic volume sequence group correspond to consultation categories; determine whether a prediction time period satisfies a service change point detection condition; for each historical consultation traffic volume sequence in the historical consultation traffic volume sequence group, perform the following steps: in response to determining that the prediction time period satisfies the service change point detection condition, perform service change point processing on the historical consultation traffic volume sequence to obtain a historical consultation traffic volume sequence after service change point processing as a sample sequence; determine the consultation category corresponding to the historical consultation traffic volume sequence as a target consultation category; generate a predicted consultation traffic volume sequence corresponding to the prediction time period and the target consultation category according to the sample sequence.

[0146] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0148] The units described in some embodiments of the present disclosure can be implemented in software or in hardware. The described units can also be provided in a processor. For example, it can be described as: a processor includes an acquisition unit, a determination unit, and an execution unit. Among them, the names of these units do not constitute a limitation on the unit itself in some cases. For example, the acquisition unit can also be described as "the unit for acquiring the historical consultation traffic volume sequence group".

[0149] The functions described above herein can be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Product (ASSP), System on Chip (SOC), Complex Programmable Logic Device (CPLD), and so on.

[0150] Some embodiments of the present disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements any of the above-described consultation traffic volume prediction methods.

[0151] The above description is only some preferred embodiments of the present disclosure and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for predicting the volume of consulting services, comprising: Obtaining a historical consulting service volume sequence group, wherein each historical consulting service volume sequence in the historical consulting service volume sequence group corresponds to a consulting category; Determining whether the prediction time period meets the business change point detection condition; For each historical consulting service volume sequence in the historical consulting service volume sequence group, perform the following steps: In response to determining that the prediction time period meets the business change point detection condition, perform business change point processing on the historical consulting service volume sequence to obtain the historical consulting service volume sequence after business change point processing as a sample sequence; Determining the consulting category corresponding to the historical consulting service volume sequence as the target consulting category; Generating a predicted consulting service volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence.

2. The method according to claim 1, wherein Before generating the predicted consulting service volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence, the method further includes: In response to determining that the prediction time period does not meet the business change point detection condition, determining the historical consulting service volume sequence as the sample sequence.

3. The method according to claim 1, wherein, The performing business change point processing on the historical consulting service volume sequence to obtain the historical consulting service volume sequence after business change point processing as a sample sequence includes: Determining a target business change point according to the historical consulting service volume sequence; Determining each historical consulting service volume arranged after the target business change point in the historical consulting service volume sequence as the sample sequence.

4. The method according to claim 1, wherein The generating a predicted consulting service volume sequence corresponding to the prediction time period and the target consulting category according to the sample sequence includes: Adjusting each first model parameter of the initial first consulting service volume prediction model according to the sample sequence to train the first consulting service volume prediction model corresponding to the prediction time period and the target consulting category; Adjusting each second model parameter of the initial second consulting service volume prediction model according to the sample sequence to train the second consulting service volume prediction model corresponding to the prediction time period and the target consulting category; Generating a first predicted consulting service volume sequence corresponding to the prediction time period and the target consulting category according to the first consulting service volume prediction model and the sample sequence; Generating a second predicted consulting service volume sequence corresponding to the prediction time period and the target consulting category according to the second consulting service volume prediction model and the sample sequence; Generating a predicted consulting service volume sequence corresponding to the prediction time period and the target consulting category according to the first predicted consulting service volume sequence and the second predicted consulting service volume sequence.

5. The method according to claim 4, wherein The adjusting each first model parameter of the initial first consulting service volume prediction model according to the sample sequence to train the first consulting service volume prediction model corresponding to the prediction time period and the target consulting category includes: Based on the initial first consulting service volume prediction model, performing the following parameter optimization steps: Generate a set of parameter verification information corresponding to the first parameter space through the sample sequence and the initial first consulting business volume prediction model, where the parameter verification information in the set of parameter verification information includes accuracy and uncertainty information, and the first parameter space corresponds to each of the first model parameters; Determine each first model adjustment parameter corresponding to each of the first model parameters according to the set of parameter verification information; Update the initial first consulting business volume prediction model according to each first model adjustment parameter; Generate model verification information according to the updated initial first consulting business volume prediction model and the sample sequence; In response to determining that the model verification information meets the preset convergence condition, determine the updated initial first consulting business volume prediction model as the first consulting business volume prediction model.

6. The method according to claim 5, wherein, The parameter optimization step further includes: In response to determining that the model verification information does not meet the preset convergence condition, execute the parameter optimization step again according to the updated initial first consulting business volume prediction model and the adjusted first parameter space.

7. The method according to claim 4, wherein, The adjusting each second model parameter of the initial second consulting business volume prediction model according to the sample sequence to train a second consulting business volume prediction model corresponding to the prediction time period and the target consulting category includes: Construct each to-be-screened second consulting business volume prediction model through a preset second parameter space corresponding to each of the second model parameters; For each constructed to-be-screened second consulting business volume prediction model, generate model index information corresponding to the to-be-screened second consulting business volume prediction model according to the sample sequence; Select, from the various to-be-screened second consulting business volume prediction models, the to-be-screened second consulting business volume prediction model whose corresponding model index information meets the preset model index condition as the second consulting business volume prediction model.

8. The method according to any one of claims 1 to 7, wherein, The method further includes: For each generated predicted consulting business volume sequence, perform the following steps: Obtain a set of schedulable customer service user information corresponding to the consulting category of the predicted consulting business volume sequence; Generate the total unit business processing volume according to the set of schedulable customer service user information; For each predicted consulting business volume in the predicted consulting business volume set, perform the following steps: In response to determining that the predicted consulting business volume is greater than the total unit business processing volume, determine the difference between the predicted consulting business volume and the total unit business processing volume as the total unit to-be-processed volume; Determine the set of schedulable customer service user information and the total unit to-be-processed volume as customer service user scheduling information; Generate an initial customer service user scheduling table according to the determined customer service user scheduling information.

9. The method according to claim 8, wherein, After generating the initial customer service user scheduling table according to the determined customer service user scheduling information, the method further includes: Display the initial customer service user scheduling table, where each customer service user scheduling information corresponds to a table row of the initial customer service user scheduling table; In response to detecting an external customer service user information input operation acting on any table row including the total unit to-be-processed volume, add the input external customer service user information to the any table row; In response to detecting a save operation on the initial customer service user schedule, the initial customer service user schedule added with each external customer service user information is determined as the customer service user schedule.

10. A consultation business volume prediction device, comprising: An acquisition unit configured to acquire a historical consultation business volume sequence group, wherein a historical consultation business volume sequence in the historical consultation business volume sequence group corresponds to a consultation category; A determination unit configured to determine whether a prediction time period meets a service change point detection condition; An execution unit configured to perform the following steps for each historical consultation business volume sequence in the historical consultation business volume sequence group: In response to determining that the prediction time period meets the service change point detection condition, perform service change point processing on the historical consultation business volume sequence to obtain the historical consultation business volume sequence after service change point processing as a sample sequence; Determine the consultation category corresponding to the historical consultation business volume sequence as the target consultation category; Generate a predicted consultation business volume sequence corresponding to the prediction time period and the target consultation category according to the sample sequence.

11. An electronic device, comprising: One or more processors; A storage device having stored thereon 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 1-9.

12. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, the method according to any one of claims 1-9 is implemented.

Images