Task volume prediction method, device, electronic device and storage medium

Abstract

The present disclosure provides a method, device, electronic device and storage medium for predicting task volume. The method includes: obtaining historical task data generated by a task object during historical operation, and obtaining attribute data of the associated object according to the identification information of the associated object in the historical task data; determining the task volume generated by the task object in the historical time according to the task information in the historical task data, and using the task volume as a category label for model training; generating a sample data set according to the historical task data, attribute data and category label, and training a classification model using the sample data set to obtain a trained classification model; obtaining feature information of the predicted object, inputting the feature information into the trained classification model, and using the trained classification model to predict the task volume of the predicted object within a preset time in the future. The present disclosure can achieve accurate prediction of task volume, improve the accuracy of prediction results, and enhance customer experience.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, electronic device and storage medium for predicting task volume. Background Art

[0002] With the rapid development of robots and artificial intelligence, robots have been widely used in various fields to replace humans in completing some tasks. For example, in offline hotel business, robots installed in hotels are used to perform tasks such as delivering goods, welcoming guests, and patrolling. Since the services provided by robots and their use effects are closely related to the amount of tasks of customers (such as hotel customers), how to accurately predict the amount of tasks that customers will have in the future is of great significance to the evaluation of the use effect of robots.

[0003] Taking the use of robots in the hotel industry as an example, in the traditional way of predicting the hotel's task volume, the overall task volume of all hotels in a period of time is first manually counted, and then the overall task volume is divided by the number of hotels to get the total task volume of each hotel in a few days, and then the task volume of the hotel in each day is further calculated based on the total task volume. However, this simple method of averaging the total task volume cannot accurately predict the task volume that the hotel may generate in the future, and there is a large gap between the calculated result and the actual task volume, which reduces the customer experience and affects the development of the enterprise. Summary of the invention

[0004] In view of this, the embodiments of the present disclosure provide a method, device, electronic device and storage medium for predicting task volume to solve the problem that the prior art cannot accurately predict the customer's task volume, and there is a large gap between the calculated result and the actual task volume, resulting in a reduced customer experience.

[0005] In a first aspect of an embodiment of the present disclosure, a method for predicting a task volume is provided, including: obtaining historical task data generated by a task object during a historical operation process, and obtaining attribute data corresponding to the associated object based on identification information of the associated object contained in the historical task data; determining the task volume generated by the task object within a historical time based on task information in the historical task data, and using the task volume within the historical time as a category label for model training; generating a sample data set based on the historical task data, attribute data, and category labels, and training a preset classification model using the sample data set to obtain a trained classification model; determining a prediction object for task volume prediction, obtaining feature information of the prediction object, inputting the feature information into the trained classification model, and using the trained classification model to predict the task volume of the prediction object within a preset future time.

[0006] According to a second aspect of an embodiment of the present disclosure, a task quantity prediction device is provided, including: an acquisition module, configured to acquire historical task data generated by a task object during a historical operation process, and acquire attribute data corresponding to the associated object based on identification information of the associated object contained in the historical task data; a determination module, configured to determine the task quantity generated by the task object within a historical time based on task information in the historical task data, and use the task quantity within the historical time as a category label for model training; a training module, configured to generate a sample data set based on the historical task data, attribute data and category labels, and use the sample data set to train a preset classification model to obtain a trained classification model; a prediction module, configured to determine a prediction object for task quantity prediction, acquire feature information of the prediction object, input the feature information into the trained classification model, and use the trained classification model to predict the task quantity of the prediction object within a preset future time.

[0007] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the above method are implemented when the processor executes the program.

[0008] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0009] At least one of the above technical solutions adopted in the embodiments of the present disclosure can achieve the following beneficial effects:

[0010] By obtaining the historical task data generated by the task object during the historical operation process, according to the identification information of the associated object contained in the historical task data, the attribute data corresponding to the associated object is obtained; according to the task information in the historical task data, the task volume generated by the task object in the historical time is determined, and the task volume in the historical time is used as the category label for model training; a sample data set is generated according to the historical task data, the attribute data and the category label, and the preset classification model is trained using the sample data set to obtain the trained classification model; the prediction object for task volume prediction is determined, the feature information of the prediction object is obtained, the feature information is input into the trained classification model, and the task volume of the prediction object in the future preset time is predicted using the trained classification model. The present disclosure can accurately estimate the task volume, reduce the gap between the prediction result and the actual task volume, improve the accuracy of the prediction result, and thus enhance the customer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 is a flowchart of a method for predicting task volume provided by an embodiment of the present disclosure;

[0013] Figure 2 is a schematic diagram of the structure of a task volume prediction device provided by an embodiment of the present disclosure;

[0014] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0015] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0016] The historical data generated during the task operation process is used to predict the task volume in the future. The prediction results of the task volume can reflect the changing trend of the customer's future task volume, which is conducive to the customer's analysis of their own business situation and making decisions for the execution of subsequent tasks. The following takes the scenario of using robots to perform some tasks in the hotel service industry as an example to explain in detail the process of predicting the future demand for robot tasks by hotel customers, which may include the following contents:

[0017] When hotel customers use robots to perform some tasks, they first need to make an overall evaluation of the robot's use effect. By making an overall evaluation of the workload, it is possible to estimate how many people the robot can replace to perform the work, or to enable hotel customers to clearly know the future delivery status of a robot. It can be seen that since the services provided by the robot and its use effect are closely related to the task volume of the hotel customers, how to accurately predict the task volume of the customers in the future is of great significance to the evaluation of the robot's use effect.

[0018] The current existing technical solution is to obtain some information for predicting the task volume by dividing the system's overall task volume by the number of hotels. For example, by dividing the total task volume by the number of hotels, we can get the total task volume of a hotel in a few days, and then calculate the task volume of each hotel per day by the total task volume. However, in actual scenarios, the task volume of each hotel is not uniform, so simply calculating the average task volume by the total amount cannot obtain accurate results, or it is very different from the actual average business volume of the hotel, resulting in a reduced experience for hotel customers, and even complaints and dissatisfaction from hotel customer staff, affecting the development of the company.

[0019] In view of the above problems in the prior art, the present disclosure proposes a new method for predicting future task volume based on machine learning algorithm. In the task volume prediction method of the present disclosure, the historical task data generated by the robot during the task execution process of the hotel and the characteristic attributes of the hotel itself are used to train the machine learning model, and the trained machine learning model is used to predict the task volume of the hotel in the future, thereby generating the task trend of the hotel in the future. Through this trend, it is helpful for customers to understand their own task situation and it is helpful for enterprises to carry out precision marketing. The embodiment of the present disclosure solves the problem of being unable to accurately estimate the task volume of the hotel.

[0020] It should be noted that the task volume prediction system involved in the actual application scenario of the embodiment of the present disclosure includes at least three modules in the task volume prediction system, namely, a storage and recording module, an algorithm training module and a prediction and calculation module. Among them, the storage and recording module is used to record the number of tasks at the location of the robot and maintain the correspondence between tasks and locations; the algorithm training module is used to establish an algorithm model for task training through hotel characteristic attributes and historical task data; the prediction and calculation module is used to use the trained model to take the customer characteristic information that needs to be predicted for the task volume as input, and use the model to output the prediction result of the task volume.

[0021] The following embodiments of the present disclosure take the hotel task scenario as an example, taking the task volume of the robot in the hotel as the prediction target, and training the model by using the hotel's attribute data and the robot's historical task data to achieve the purpose of predicting the hotel's future task volume. However, the embodiments of the present disclosure are not limited to the above application scenarios, that is, the embodiments of the present disclosure are not limited to the scenario of predicting the task volume in the hotel, nor are they limited to the prediction scenario when the robot is used as the task execution subject. Any other scenario that requires prediction of future task volume is applicable to this solution, such as the prediction scenario of the store order task volume and the prediction scenario of the company's work task volume. Therefore, the application scenarios in the following embodiments of the present disclosure do not constitute a limitation on the technical solution of the present disclosure.

[0022] Figure 1 It is a flowchart of the task volume prediction method provided by the embodiment of the present disclosure. Figure 1 The task volume prediction method can be executed by the server. Figure 1 As shown, the task volume prediction method may specifically include:

[0023] S101, acquiring historical task data generated by the task object during its historical operation, and acquiring attribute data corresponding to the associated object according to identification information of the associated object contained in the historical task data;

[0024] S102, determining the task volume generated by the task object in the historical time according to the task information in the historical task data, and using the task volume in the historical time as the category label for model training;

[0025] S103, generating a sample data set according to the historical task data, attribute data and category labels, and using the sample data set to train a preset classification model to obtain a trained classification model;

[0026] S104, determining a prediction object for task volume prediction, obtaining feature information of the prediction object, inputting the feature information into a trained classification model, and using the trained classification model to predict the task volume of the prediction object within a preset future time.

[0027] Specifically, the task object of the embodiment of the present disclosure refers to the hotel robot, the associated object refers to the associated hotel corresponding to the hotel robot, and the prediction object is the predicted hotel that actually performs the task volume prediction. The following embodiments of the present disclosure take the hotel as the prediction object and the robot set in the hotel as the subject of task execution. By obtaining the historical task data generated by the robot during the historical task execution process and the attribute data of the hotel associated with the robot as the sample data for model training, the preset machine learning algorithm model is trained to obtain a model for predicting the task volume of the subsequent predicted hotel, and it is deployed to the online business system for prediction.

[0028] Furthermore, historical task data refers to some task-related data generated by the robot during the historical execution of tasks. Taking the hotel robot as an example, hotel customers can use the hotel robot to perform tasks such as delivering goods, patrolling, welcoming guests, and order delivery. Therefore, historical task data is data with robot business attributes generated by the robot during the execution of tasks, such as hotel name, task time, task details, etc. Corresponding to the historical task data is the task volume completed by the robot. The task volume can be considered as the total amount of all task types completed by the robot. Each task can correspond to an order. For example, a patrol task can be considered as a task order, so the task volume can be considered as the total amount of task orders.

[0029] Further, the machine learning algorithm adopted in the embodiment of the present disclosure is a decision tree, that is, a decision tree is constructed through historical task data and hotel attribute data, and the constructed decision tree is used to predict the future task volume of the hotel. Here, the decision tree is a tree structure (which can be a binary tree or a non-binary tree), each non-leaf node of which represents a test on a feature attribute, each branch represents the output of this feature attribute on a certain value domain, and each leaf node is used to store a category. However, it should be understood that the machine learning model of the embodiment of the present disclosure is not limited to decision trees, and other classification models are also applicable to this solution, such as multi-label random forests, multi-label gradient boosting and other models.

[0030] According to the technical solution provided by the embodiment of the present disclosure, the present disclosure obtains the historical task data generated by the task object during the historical operation process, and obtains the attribute data corresponding to the associated object according to the identification information of the associated object contained in the historical task data; determines the task volume generated by the task object in the historical time according to the task information in the historical task data, and uses the task volume in the historical time as the category label for model training; generates a sample data set according to the historical task data, the attribute data and the category label, and uses the sample data set to train the preset classification model to obtain the trained classification model; determines the prediction object for task volume prediction, obtains the feature information of the prediction object, inputs the feature information into the trained classification model, and uses the trained classification model to predict the task volume of the prediction object in the future preset time. The present disclosure can accurately estimate the task volume, reduce the gap between the prediction result and the actual task volume, improve the accuracy of the prediction result, and thus enhance the customer experience.

[0031] In some embodiments, the task object includes a hotel robot, and obtaining historical task data generated by the task object during its historical operation includes: obtaining historical task data generated by the hotel robot during its historical operation from a cloud platform database, wherein the historical task data includes identification information of the associated object, task time, task type, and task information.

[0032] Specifically, the disclosed embodiment is described with the task volume prediction under the hotel task scenario as the application scenario, so the task object (i.e., the task execution subject) of the disclosed embodiment refers to the hotel robot. In the actual scenario, the hotel robot will generate a series of historical task data in the process of executing historical tasks, and the hotel robot will report these historical task data to the cloud platform, wherein the historical task data reported by the hotel robot includes but is not limited to the following information: hotel information, task time, task type, task information, etc. The hotel information may include hotel identification information (such as hotel name) and the location information of the hotel, etc., and the task information may include detailed information of the executed tasks. Before model construction, it is necessary to first obtain data related to the historical tasks of the hotel robot from the cloud platform database, and use these data as part of the sample data.

[0033] In some embodiments, the associated object includes an associated hotel corresponding to the hotel robot, and attribute data corresponding to the associated object is obtained according to the identification information of the associated object contained in the historical task data, including: according to the identification information of the associated hotel in the historical task data, the attribute data of the associated hotel corresponding to the identification information is obtained from the cloud platform database, wherein the attribute data includes the name, location information, number of rooms, occupancy rate and average room price of the associated hotel.

[0034] Specifically, the associated object of the disclosed embodiment refers to the associated hotel corresponding to the hotel robot, that is, the hotel where the hotel robot was located when executing the historical task. In the process of executing the historical task, an association relationship between the hotel and the robot will be established, so that the attribute data of the associated hotel can be found according to the robot's historical task data, such as the name of the hotel, the number of rooms in the hotel, the hotel occupancy rate, the average price of the hotel rooms and other dimensions. The attribute data of the hotel can also be considered as the characteristic attributes corresponding to the hotel, that is, the characteristics of the hotel in which dimensions. When the model is constructed later, the attribute data of the associated hotel obtained will be used as part of the sample data when the model is trained.

[0035] In some embodiments, the task volume generated by the task object in the historical time is determined based on the task information in the historical task data, and the task volume in the historical time is used as the category label for model training, including: obtaining the task information in the historical task data in the historical time, counting the task information at a time interval of one day, determining the task volume corresponding to each day of the hotel robot in the historical time based on the statistical results, and using the task volume corresponding to each day as the category label for each sample data, wherein the task volume includes the number of tasks performed by the hotel robot on each day.

[0036] Specifically, after obtaining the historical task data and the hotel's attribute data, it is necessary to further determine the historical task volume corresponding to the hotel on each day, and use the daily historical task volume data as the category label for model training of the sample data. In practical applications, the hotel's daily historical task volume can be obtained by analyzing and counting the task information in the historical task data, and the daily historical task volume is used as the category label corresponding to each sample data.

[0037] Furthermore, since the task information contains detailed information about the robot's historical tasks in the hotel, the task volume of the robot corresponding to each hotel can be counted based on the task information, and the task volume of all robots in the hotel can be added up to get the historical task volume of the hotel corresponding to each day. In actual applications, the same robot may perform multiple different tasks every day, so the task volume corresponding to each robot refers to the sum of various types of tasks performed by the robot every day. For example, if the robot performs 50 patrol tasks and 50 meal delivery tasks on a certain day, then the task volume of the robot on that day is the combination of the number of patrol tasks and the number of meal delivery tasks, that is, the task volume of the robot is 100.

[0038] In some embodiments, a sample data set is generated based on historical task data, attribute data, and category labels, including: based on the task time in the historical task data, historical task data and attribute data whose task time belongs to the same day are spliced ​​to obtain sample data corresponding to each day in the historical time, and the task amount of each day is used as the category label of the sample data, and the sample data and the category label are combined to obtain a sample data set.

[0039] Specifically, after obtaining the robot's historical task data, the hotel's attribute data, and the historical task volume of each day, these data are spliced ​​with each day as the time dimension to obtain a sample data set. In practical applications, each piece of data in the sample data set can be composed of the following types of data, namely, the hotel's feature dimension (i.e., attribute data) + historical task data + time dimension (i.e., the date corresponding to each day) + the robot's task dimension (i.e., the historical task volume). Therefore, in the sample data set, each day corresponds to a piece of sample data, and the sample data set contains sample data within several days of history.

[0040] In some embodiments, the classification model adopts a decision tree algorithm and uses a sample data set to train the preset classification model to obtain a trained classification model, including: according to the preset decision tree algorithm, selecting at least one characteristic attribute from the attribute data as a root node, and forming non-leaf nodes in sequence along the root node, and setting corresponding classification conditions and classification thresholds for each non-leaf node until all leaf nodes are generated, so as to obtain a decision tree for predicting the task volume, wherein each non-leaf node corresponds to a characteristic attribute, and each leaf node corresponds to a classification result.

[0041] Specifically, decision tree is a machine learning method. The generation algorithms of decision tree include ID3, C4.5, C5.0, etc. Decision tree is a tree structure, in which each internal node represents a judgment on an attribute, each branch represents the output of a judgment result, and finally each leaf node represents a classification result. The decision tree model has a tree structure. In the classification problem, it represents the process of classifying instances based on features. Its main advantages are that the model is readable and the classification speed is fast. During training, the sample data set is used to establish a decision tree model based on the principle of minimizing the loss function; during prediction, the decision tree model is used to classify new attribute data.

[0042] Furthermore, after obtaining the above sample data set, the sample data set is used as the input of the decision tree model, and the decision tree generation algorithm is used to construct the decision tree model. In practical applications, a characteristic attribute can be selected from the hotel's attribute data as the root node, and non-leaf nodes on each branch can be selected from the root node in turn. Each non-leaf node contains two dimensions of data, namely, the hotel's characteristic attributes (i.e., specific attribute data) and date data. Therefore, each different non-leaf node in the decision tree corresponds to hotel features of different dimensions, that is, each non-leaf node corresponds to a different characteristic value of the hotel. The non-leaf nodes here can also be considered as child nodes in the decision tree.

[0043] Furthermore, when selecting which characteristic attribute in the hotel's attribute data is used as the root node and non-leaf node, it can be judged according to the expert rules. In practical applications, characteristic attributes that have a greater impact on the prediction results can be selected as non-leaf nodes. The final leaf node in the decision tree corresponds to the final classification value, that is, the final label, and each leaf node corresponds to the task volume of a certain prediction date.

[0044] In some embodiments, the prediction object includes a predicted hotel as a task volume prediction object, characteristic information of the prediction object is obtained, the characteristic information is input into a trained classification model, and the trained classification model is used to predict the task volume of the prediction object within a preset future time, including: obtaining characteristic information corresponding to the predicted hotel, and processing the characteristic information to obtain attribute data of the predicted hotel, using each non-leaf node in the decision tree to test the attribute data of the predicted hotel until the test process reaches a leaf node, and using the classification result corresponding to the leaf node as the prediction result of the task volume corresponding to the predicted hotel, wherein the prediction result includes the task volume corresponding to the predicted hotel on each day within the preset future time.

[0045] Specifically, after the decision tree model is trained using the sample data set, the trained decision tree model can be deployed to the online business system. When a customer consults through online services or robot services, the hotel name filled in by the customer is obtained, and the cloud system calls third-party software to search for hotel information. The hotel information is input into the analysis module of the business system, and the analysis module is used to predict the task volume of the robots in the hotel to obtain the possible daily task volume of the robots in the hotel in the future.

[0046] Furthermore, when using the decision tree model to predict the task volume, the feature information corresponding to the input predicted hotel is first processed to obtain the attribute data of the predicted hotel. Starting from the root node in the decision tree, the corresponding feature attributes in the items to be classified are tested in turn along the branches of the decision tree, and the output branch is selected according to the value of each non-leaf node until the leaf node is reached, and the category stored in the leaf node is used as the final prediction result.

[0047] Furthermore, after using the decision tree to obtain the task volume of the hotel in the future, a report is drawn based on the task volume of each day in the future to show the trend of the future task volume change. For example, the task volume corresponding to each day in the next year is drawn using a histogram or a line graph to obtain a report on the trend of the hotel's task volume change in the next year. In the report, the horizontal axis represents the future time and the vertical axis represents the task volume. Through this report, the change of the hotel's task volume in the future can be analyzed, so as to determine the number of robots required by the hotel in the future and the task volume that the robots will undertake in the future.

[0048] According to the technical solution provided by the embodiment of the present disclosure, the present disclosure proposes a new prediction method for future task volume based on decision tree. In the task volume prediction method of the present disclosure, the historical task data generated by the robot during the task execution process of the hotel and the characteristic attributes of the hotel itself are used to build a decision tree model. The decision tree model is used to predict the task volume of the hotel in the future, so as to draw a trend report of the change of the task volume of the hotel in the future. Through the trend report, it is helpful to understand the future task volume of hotel customers, so as to achieve precision marketing for hotel customers.

[0049] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0050] Figure 2 Schematic diagram of the structure of the task volume prediction device provided by the embodiment of the present disclosure. Figure 2 As shown, the task volume prediction device includes:

[0051] The acquisition module 201 is configured to acquire historical task data generated by the task object during the historical operation process, and acquire attribute data corresponding to the associated object according to the identification information of the associated object contained in the historical task data;

[0052] A determination module 202 is configured to determine the amount of tasks generated by the task object in the historical time according to the task information in the historical task data, and use the amount of tasks in the historical time as a category label for model training;

[0053] The training module 203 is configured to generate a sample data set according to the historical task data, attribute data and category labels, and train a preset classification model using the sample data set to obtain a trained classification model;

[0054] The prediction module 204 is configured to determine a prediction object for task volume prediction, obtain feature information of the prediction object, input the feature information into a trained classification model, and use the trained classification model to predict the task volume of the prediction object within a preset time in the future.

[0055] In some embodiments, the task object includes a hotel robot, Figure 2 The acquisition module 201 acquires historical task data generated by the hotel robot during historical operation from the cloud platform database, wherein the historical task data includes identification information of the associated object, task time, task type, and task information.

[0056] In some embodiments, the associated object includes an associated hotel corresponding to the hotel robot. Figure 2The acquisition module 201 acquires the attribute data of the associated hotel corresponding to the identification information from the cloud platform database according to the identification information of the associated hotel in the historical task data, wherein the attribute data includes the name, location information, number of rooms, occupancy rate and average room price of the associated hotel.

[0057] In some embodiments, Figure 2 The determination module 202 obtains the task information in the historical task data within the historical time, counts the task information with one day as the time interval, determines the task volume corresponding to each day of the hotel robot within the historical time according to the statistical results, and uses the task volume corresponding to each day as the category label of each sample data, wherein the task volume includes the number of tasks performed by the hotel robot on each day.

[0058] In some embodiments, Figure 2 The training module 203 splices the historical task data and attribute data belonging to the same day according to the task time in the historical task data, obtains the sample data corresponding to each day in the historical time, and uses the task amount of each day as the category label of the sample data, and combines the sample data and the category label to obtain the sample data set.

[0059] In some embodiments, the classification model uses a decision tree algorithm. Figure 2 The training module 203 selects at least one characteristic attribute from the attribute data as a root node according to a preset decision tree algorithm, and forms non-leaf nodes in sequence along the root node, and sets corresponding classification conditions and classification thresholds for each non-leaf node until all leaf nodes are generated, so as to obtain a decision tree for predicting the task volume, wherein each non-leaf node corresponds to a characteristic attribute, and each leaf node corresponds to a classification result.

[0060] In some embodiments, the forecast object includes a forecast hotel as a task volume forecast object. Figure 2 The prediction module 204 obtains the feature information corresponding to the predicted hotel, and processes the feature information to obtain the attribute data of the predicted hotel, and uses each non-leaf node in the decision tree to test the attribute data of the predicted hotel until the test process reaches the leaf node, and uses the classification result corresponding to the leaf node as the prediction result of the task volume corresponding to the predicted hotel, wherein the prediction result includes the task volume corresponding to the predicted hotel on each day within a preset time in the future.

[0061] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.

[0062] Figure 3Schematic diagram of the structure of the electronic device 3 provided in the embodiment of the present disclosure. Figure 3 As shown, the electronic device 3 of this embodiment includes: a processor 301, a memory 302, and a computer program 303 stored in the memory 302 and executable on the processor 301. When the processor 301 executes the computer program 303, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 301 executes the computer program 303, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0063] Exemplarily, the computer program 303 may be divided into one or more modules / units, which are stored in the memory 302 and executed by the processor 301 to complete the present disclosure. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program 303 in the electronic device 3.

[0064] The electronic device 3 may be a desktop computer, a notebook, a PDA, a cloud server, or other electronic device. The electronic device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that Figure 3 It is only an example of the electronic device 3 and does not constitute a limitation of the electronic device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0065] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0066] The memory 302 may be an internal storage unit of the electronic device 3, for example, a hard disk or memory of the electronic device 3. The memory 302 may also be an external storage device of the electronic device 3, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 3. Further, the memory 302 may also include both an internal storage unit and an external storage device of the electronic device 3. The memory 302 is used to store computer programs and other programs and data required by the electronic device. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0067] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0068] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0069] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0070] In the embodiments provided in the present disclosure, it should be understood that the disclosed apparatus / computer equipment and methods can be implemented in other ways. For example, the apparatus / computer equipment embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. Multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of the apparatus or unit, which may be electrical, mechanical or other forms.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0072] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0073] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present disclosure implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. The computer program may include computer program code, and the computer program code may be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.

[0074] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A method for predicting task volume, characterized in that: include: Acquire historical task data generated by the task object during its historical operation, and acquire attribute data corresponding to the associated object according to identification information of the associated object contained in the historical task data; Determine the task volume generated by the task object in the historical time according to the task information in the historical task data, and use the task volume in the historical time as the category label for model training; Generate a sample data set according to the historical task data, the attribute data and the category label, and use the sample data set to train a preset classification model to obtain a trained classification model; Determine a prediction object for task volume prediction, obtain feature information of the prediction object, input the feature information into the trained classification model, and use the trained classification model to predict the task volume of the prediction object within a preset future time; the prediction object includes a predicted hotel as a task volume prediction object; the feature information includes hotel information of the predicted hotel; The task object includes a hotel robot, and the acquisition of historical task data generated by the task object during historical operation includes: Acquire historical task data generated by the hotel robot during historical operation from a cloud platform database, wherein the historical task data includes identification information of an associated object, task time, task type, and task information; The associated object includes an associated hotel corresponding to the hotel robot, and according to the identification information of the associated object contained in the historical task data, the attribute data corresponding to the associated object is obtained, including: According to the identification information of the associated hotel in the historical task data, the attribute data of the associated hotel corresponding to the identification information is obtained from the cloud platform database, wherein the attribute data includes the name, location information, number of rooms, occupancy rate and average room price of the associated hotel.

2. The method according to claim 1, characterized in that The determining the task amount generated by the task object in the historical time according to the task information in the historical task data, and using the task amount in the historical time as a category label for model training, includes: Obtain task information from the historical task data within the historical time, perform statistics on the task information at one day as a time interval, determine the task volume corresponding to each day of the hotel robot within the historical time according to the statistical results, and use the task volume corresponding to each day as the category label of each sample data, wherein the task volume includes the number of tasks performed by the hotel robot on each day.

3. The method according to claim 1, characterized in that The generating a sample data set according to the historical task data, the attribute data and the category label includes: According to the task time in the historical task data, the historical task data with the task time belonging to the same day is spliced ​​with the attribute data to obtain sample data corresponding to each day in the historical time, and the task amount of each day is used as the category label of the sample data, and the sample data and the category label are combined to obtain a sample data set.

4. The method according to claim 1, characterized in that: The classification model adopts a decision tree algorithm, and uses the sample data set to train the preset classification model to obtain a trained classification model, including: According to a preset decision tree algorithm, at least one characteristic attribute is selected from the attribute data as a root node, and non-leaf nodes are formed in sequence along the root node, and corresponding classification conditions and classification thresholds are set for each of the non-leaf nodes until all leaf nodes are generated, so as to obtain a decision tree for predicting the task volume, wherein each of the non-leaf nodes corresponds to a characteristic attribute, and each of the leaf nodes corresponds to a classification result.

5. The method according to claim 4, characterized in that Acquiring feature information of the prediction object, inputting the feature information into the trained classification model, and using the trained classification model to predict the task volume of the prediction object within a preset future time, including: Acquire characteristic information corresponding to the predicted hotel, and process the characteristic information to obtain attribute data of the predicted hotel, use each non-leaf node in the decision tree to test the attribute data of the predicted hotel until the test process reaches a leaf node, and use the classification result corresponding to the leaf node as the prediction result of the task volume corresponding to the predicted hotel, wherein the prediction result includes the task volume corresponding to the predicted hotel on each day within a preset time in the future.

6. A task volume prediction device, characterized in that: include: An acquisition module is configured to acquire historical task data generated by the task object during its historical operation, and acquire attribute data corresponding to the associated object according to identification information of the associated object contained in the historical task data; A determination module is configured to determine the amount of tasks generated by the task object in a historical time according to the task information in the historical task data, and use the amount of tasks in the historical time as a category label for model training; A training module is configured to generate a sample data set according to the historical task data, the attribute data and the category label, and train a preset classification model using the sample data set to obtain a trained classification model; A prediction module is configured to determine a prediction object for task volume prediction, obtain feature information of the prediction object, input the feature information into the trained classification model, and use the trained classification model to predict the task volume of the prediction object within a preset future time; the prediction object includes a predicted hotel as a task volume prediction object; the feature information includes hotel information of the predicted hotel; The task object includes a hotel robot, and the acquisition module acquires historical task data generated by the hotel robot during historical operation from a cloud platform database, wherein the historical task data includes identification information of an associated object, task time, task type, and task information; The associated object includes an associated hotel corresponding to the hotel robot. The acquisition module acquires attribute data of the associated hotel corresponding to the identification information from the cloud platform database according to the identification information of the associated hotel in the historical task data, wherein the attribute data includes the name, location information, number of rooms, occupancy rate and average room price of the associated hotel.

7. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 5 when executing the program.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images