A method, device, equipment and medium for predicting procurement data

Abstract

The present invention discloses a method, apparatus, device and medium for predicting procurement data. The method includes: constructing historical procurement feature data of the previous procurement cycle according to procurement prediction-related data; the procurement cycle includes at least two years; training decision trees for each year of the procurement cycle according to the historical procurement feature data, and constructing an initial prediction model according to the leaf ratio in each decision tree; performing trend analysis on the historical procurement feature data to obtain expected procurement feature data of the future cycle; adjusting and optimizing the initial prediction model according to the expected procurement feature data to obtain a target prediction model. Embodiments of the present invention can improve the prediction efficiency of procurement data.

Description

Technical Field

[0001] The present invention relates to the technical field of procurement data prediction, and particularly to a method, device, equipment and medium for predicting procurement data. Background Art

[0002] Scientific decision-making is an important responsibility of modern managers. In enterprise supplier procurement management, a common scenario is that some procurement indicators are known, but a considerable part of procurement indicators are unknown. Therefore, it is necessary to infer unknown procurement indicators based on known procurement indicators.

[0003] Currently, the prediction systems for procurement on the market are basically based on one or two procurement indicators, such as procurement based on sales or production. If an enterprise sets many procurement indicators and the existing prediction systems cannot support complex predictions with multiple indicators, manual judgment is required for prediction, resulting in low prediction efficiency. Summary of the Invention

[0004] The present invention provides a method, device, equipment and medium for predicting procurement data to solve the problem of improving the prediction efficiency of procurement.

[0005] According to one aspect of the present invention, a method for predicting procurement data is provided, including:

[0006] Constructing historical procurement feature data of the previous procurement cycle according to procurement prediction-related data; the procurement cycle includes at least two years;

[0007] Training decision trees for each year of the procurement cycle according to the historical procurement feature data, and constructing an initial prediction model according to the leaf proportion in each decision tree;

[0008] Performing trend analysis on the historical procurement feature data to obtain expected procurement feature data for the future cycle;

[0009] Adjusting and optimizing the initial prediction model according to the expected procurement feature data to obtain a target prediction model.

[0010] According to another aspect of the present invention, a device for predicting procurement data is provided, including:

[0011] A data sorting module, configured to construct historical procurement feature data of the previous procurement cycle according to procurement prediction-related data; the procurement cycle includes at least two years;

[0012] A model construction module, configured to train decision trees for each year of the procurement cycle according to the historical procurement feature data, and construct an initial prediction model according to the leaf proportion in each decision tree;

[0013] A trend analysis module for performing trend analysis on the historical purchase feature data to obtain expected purchase feature data for a future period;

[0014] A model optimization module for adjusting and optimizing the initial prediction model according to the expected purchase feature data to obtain a target prediction model.

[0015] According to another aspect of the present invention, there is provided a computer program product including a computer program which, when executed by a processor, implements the prediction method for purchase data according to any embodiment of the present invention.

[0016] According to another aspect of the present invention, there is provided an electronic device including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the prediction method for purchase data according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the prediction method for purchase data according to any embodiment of the present invention when executed.

[0018] In the embodiments of the present invention, a decision tree is trained to generate a prediction model, and the prediction model is used to predict purchase features, reducing the amount of manual participation in prediction and improving the prediction efficiency. And the periodicity and future trends of purchases are considered during the generation process, improving the accuracy of prediction.

[0019] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0021] Figure 1 is a flowchart of a prediction method for purchase data according to an embodiment of the present invention;

[0022] Figure 2It is a flowchart of a method for predicting procurement data provided according to another embodiment of the present invention;

[0023] Figure 3 It is a schematic diagram of data in a prediction model provided according to another embodiment of the present invention;

[0024] Figure 4 It is a schematic structural diagram of a method for predicting procurement data provided according to another embodiment of the present invention;

[0025] Figure 5 It is a schematic structural diagram of an electronic device for implementing the embodiment of the present invention. Detailed implementation manners

[0026] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] Figure 1 It is a flowchart of a method for predicting procurement data provided in an embodiment of the present invention. This embodiment is applicable to generating a prediction model based on historical procurement feature data and determining an unknown procurement index value based on the prediction model and known procurement index values. This method can be executed by a prediction device for procurement data, and this device can be implemented in the form of hardware and / or software. This device can be configured in an electronic device with corresponding data processing capabilities, such as the server of a procurement prediction system. As Figure 1 shown, the method includes:

[0029] S110. Construct historical procurement feature data for the previous procurement cycle according to procurement prediction-related data.

[0030] S120. Train decision trees for each year of the procurement cycle based on the historical procurement feature data, and construct an initial prediction model according to the leaf proportion in each decision tree.

[0031] S130. Conduct trend analysis on the historical procurement feature data to obtain expected procurement feature data for the future cycle.

[0032] S140. Adjust and optimize the initial prediction model according to the expected procurement feature data to obtain a target prediction model.

[0033] Among them, the procurement prediction-related data includes historical procurement data, market trend data, supplier information, and other relevant data. The historical procurement data includes the procurement volume, procurement frequency, procurement time, etc. The market trend data includes market price fluctuations, market demand changes, etc. The supplier information includes supplier reputation, supply capacity, price, etc. The procurement cycle includes at least two years.

[0034] Specifically, the collection of procurement prediction-related data is the crucial first step. The production and procurement activities of an enterprise have a certain periodicity (for example, a cycle of 5 years). Collect procurement prediction-related data for the previous cycle through data sources such as the enterprise's internal database, market research, and public data, and put it into the source data container. Perform various preprocessings on the data in the source data container, including:

[0035] Invalid processing: Check whether there are outliers, missing values, or duplicate values in the data, and delete or fill them.

[0036] Compensation processing: It is mainly divided into two categories of processing, numerical and character. First, perform numerical filling. For data with null values, extract vector values for all model fields through the corresponding data model, and search for this data row in the entire model. Finally, for data rows with a score of 0.9, extract the current null value field for filling. If it is between 0.5 and 0.8, through the data vectors of each column, continuously query the data to find the core data columns of the data records. Save the columns with values greater than 0.5 in each column, and finally obtain the largest two columns as the data benchmark for filling the same data column values; for the remaining data with null values, use methods such as average value, median, and interpolation for filling. Secondly, for character filling, first reason about the field identification and description to find fields that may be related, such as whether there are dictionary values and linear relationships of values. If the fields meet the reasoning relationship, fill them. For unfilled fields, perform linear reasoning on all numerical columns in terms of days, weeks, months, quarters, and years to obtain the fields matching this time node for filling.

[0037] Governance processing: Theoretically, the larger the amount of data, the more accurate the data predictability. For data of different magnitudes, according to the accuracy coefficient of the configured data, standardization processing needs to be performed to make the values of different features on the same scale.

[0038] After completing the preprocessing of the data, perform feature engineering processing (including feature value construction) on the data in the source data container to obtain the historical procurement feature data of the previous cycle, specifically including the historical procurement feature data of each year in the previous cycle.

[0039] Select the target feature from the historical procurement feature data as the root point of the decision tree, and use the other features except the target feature as the leaf nodes of the decision tree to obtain the initial decision tree.

[0040] Use the historical procurement feature data of each year in the procurement cycle to train the initial decision tree respectively to obtain the corresponding number of decision trees, and each decision tree corresponds to a year of the procurement cycle. Determine the underlying relationship between different features in each year of the procurement cycle according to the leaf ratio in each decision tree, and generate the initial prediction model according to this relationship.

[0041] Perform data normalization processing on the historical procurement feature data, use linear regression and logistic regression algorithms to construct the procurement trend for the future cycle (10 years) to obtain the expected procurement feature data. At the same time, use the prediction model to predict the procurement trend for the future cycle to obtain the predicted procurement feature data. Compare the expected procurement feature data with the predicted procurement feature data, and adjust and optimize the data in the prediction model according to the comparison result of the two to obtain the target prediction model.

[0042] In the embodiment of the present invention, a prediction model is generated by training a decision tree, and the prediction model is used to predict the procurement features, reducing the amount of manual participation in prediction and improving the prediction efficiency. And considering the periodicity and future trend of procurement during the generation process, the accuracy of prediction is improved.

[0043] Based on the above embodiment, the target features include market trend and annual demand, and the other features include procurement volume, procurement frequency, production volume, sales volume, inventory volume, and bill of materials.

[0044] Specifically, as shown in Table 1, distinguish the business characteristics of procurement, determine one or more features under each business characteristic, and pre-specify the data extraction logic for each feature.

[0045] Table 1 Features under some business characteristics

[0046]

[0047] Figure 2The flowchart of a method for predicting procurement data provided by another embodiment of the present invention is an optimized improvement based on the above embodiment. As Figure 2 shown, the method includes:

[0048] S210. Construct historical procurement feature data of the previous procurement cycle according to procurement prediction-related data.

[0049] S220. Select a target feature as the root node from the historical procurement feature data, and use other features except the target feature as leaf nodes to obtain an initial decision tree; for each year of the procurement cycle, train the initial decision tree according to the historical procurement feature data of the current year to obtain the decision tree of the current year.

[0050] Among them, the numerical ranges of other features in different decision trees are different from each other.

[0051] Specifically, use the target feature as the root node of the decision tree, and at the same time use other features as leaf nodes to complete the construction of the initial decision tree. For each year of the procurement cycle, evenly distribute according to 100% according to the number of leaf nodes, and calculate whether the current actual procurement amount in the historical procurement feature data is the same as the procurement amount output by the initial decision tree, and continuously adjust the ratio of each leaf node accordingly to obtain the decision tree corresponding to each year.

[0052] S230. Construct a prediction model according to the proportion and numerical range of leaf nodes in each decision tree.

[0053] Specifically, as Figure 3 shown, record the proportional relationship between leaf nodes in the decision trees of each year, as well as the features and feature value ranges corresponding to the leaf nodes. Organize this data to obtain a mathematical model, and this data model is the prediction model.

[0054] S240. Perform trend analysis on the historical procurement feature data based on different growth rates respectively to obtain at least two candidate procurement feature data; select the expected procurement feature data for the future cycle from the candidate procurement features according to the information gain rate of each candidate procurement feature data.

[0055] Specifically, different growth rates (such as 10%, 20%, 30%) are preset, and trend analysis is performed on the historical procurement feature data according to different growth rates respectively to obtain multiple candidate procurement feature data, and the growth rates adopted by each candidate procurement feature data are different. The information gain rate of each candidate procurement feature data is calculated through a preset algorithm. The larger the information gain rate, the higher the reference value of the candidate procurement feature data. The candidate procurement feature data with an information gain rate greater than the threshold will be determined as the expected procurement feature data for the future period. That is, there can be multiple pieces of expected procurement feature data, and each piece of expected procurement feature data corresponds to a growth rate. Using different pieces of expected procurement feature data to adjust and optimize the initial prediction model respectively, prediction models applicable to different growth rates can be obtained. During subsequent prediction, according to the target growth rate, a prediction model matching the target growth rate can be selected from multiple prediction models.

[0056] S250. Adjust and optimize the initial prediction model according to the expected procurement feature data to obtain a target prediction model.

[0057] S260. Based on the target prediction model, determine the hit leaf node and the hit decision tree where the hit leaf node is located according to the numerical interval where the known procurement feature value is located.

[0058] S270. Substitute the known procurement feature value and the proportion of the hit leaf node into the hit decision tree, and retrain the hit decision tree to obtain the target numerical interval where the unknown procurement feature value is located; determine the target numerical interval as the prediction result of the unknown procurement feature.

[0059] Specifically, during actual prediction, determine the numerical interval where the known procurement feature value is located, then query the prediction model to determine the leaf node corresponding to this numerical interval, and determine it as the hit leaf node, and determine the decision tree where this hit leaf node is located as the hit decision tree. Substitute the known procurement feature value and the proportion of the hit leaf node into the hit decision tree, and use the pre-fetched feature data to retrain the hit decision tree. After training is completed, the data interval of the leaf node corresponding to the unknown procurement feature in this decision tree is the target numerical interval where the unknown procurement feature value is located. Use the upper limit and lower limit of the target numerical interval as the prediction maximum value and prediction minimum value of the unknown procurement feature value.

[0060] The embodiment of the present invention sets different growth rates to obtain prediction models applicable to different growth rates, improving the applicability of the prediction model and the flexibility of the prediction scheme.

[0061] Figure 4 It is a structural schematic diagram of a prediction device for procurement data provided by another embodiment of the present invention. As Figure 4As shown, the device includes:

[0062] A data sorting module 310, configured to construct historical purchase feature data of the previous purchase cycle according to purchase prediction-related data; the purchase cycle includes at least two years;

[0063] A model construction module 320, configured to train decision trees for each year of the purchase cycle according to the historical purchase feature data, and construct an initial prediction model according to the leaf proportion in each decision tree;

[0064] A trend analysis module 330, configured to perform trend analysis on the historical purchase feature data to obtain expected purchase feature data for the future cycle;

[0065] A model optimization module 340, configured to adjust and optimize the initial prediction model according to the expected purchase feature data to obtain a target prediction model.

[0066] The prediction device for purchase data provided by the embodiments of the present invention can execute the prediction method for purchase data provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0067] Optionally, the model construction module 320 includes:

[0068] A decision tree construction unit, configured to select a target feature from the historical purchase feature data as the root node, and use other features except the target feature as leaf nodes to obtain an initial decision tree;

[0069] A decision tree training unit, configured to, for each year of the purchase cycle, train the initial decision tree according to the historical purchase feature data of the current year to obtain a decision tree for the current year;

[0070] Wherein, the numerical ranges of other features in different decision trees are different, including:

[0071] Optionally, the model construction module 320 includes:

[0072] A model construction unit, configured to construct a prediction model according to the proportion and numerical range of leaf nodes in each decision tree.

[0073] Optionally, the device further includes:

[0074] A hit determination module, configured to, based on the target prediction model, determine the hit leaf node and the hit decision tree where the hit leaf node is located according to the numerical range where the known purchase feature value is located;

[0075] A retraining module, configured to substitute the known purchase feature values and the proportion of the hit leaf nodes into the hit decision tree, and retrain the hit decision tree to obtain a target numerical range where the unknown purchase feature value is located;

[0076] A result determination module, configured to determine the target numerical range as the prediction result of the unknown purchase feature.

[0077] Optionally, the target features include market trends and annual demands, and the other features include purchase volume, purchase frequency, production volume, sales volume, and inventory volume.

[0078] Optionally, the trend analysis module 330 includes:

[0079] A trend analysis unit, configured to perform trend analysis on historical purchase feature data based on different growth rates respectively to obtain at least two candidate purchase feature data;

[0080] A data screening unit, configured to select expected purchase feature data for a future period from the candidate purchase features according to the information gain rate of each candidate purchase feature data.

[0081] Furthermore, the prediction device for purchase data described above can also execute the prediction method for purchase data provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution of the method.

[0082] Figure 5 FIG. shows a schematic structural diagram of an electronic device 40 that can be used to implement an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device (such as a helmet, glasses, a watch, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0083] As Figure 4As shown, the electronic device 40 includes at least one processor 41 and a memory communicatively connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc. The memory stores a computer program executable by the at least one processor. The processor 41 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 into the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0084] Multiple components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a magnetic disk, an optical disc, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0085] The processor 41 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as the prediction method for procurement data.

[0086] In some embodiments, the prediction method for procurement data can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the prediction method for procurement data described above can be executed. Alternatively, in other embodiments, the processor 41 can be configured to execute the prediction method for procurement data by any other appropriate means (e.g., by means of firmware).

[0087] The various embodiments of the systems and techniques described above in this specification can be implemented in digital electronic circuitry, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that receives data and instructions from, and transmits data and instructions to, a storage system, at least one input device, and at least one output device.

[0088] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The computer programs can be executed entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0089] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0090] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) through which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0091] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0092] A computing system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0093] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0094] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for predicting procurement data, characterized in that: The method comprises: Constructing historical procurement characteristic data of a previous procurement cycle based on procurement forecast related data; the procurement cycle includes at least two years; the procurement forecast related data includes historical procurement data, market trend data and supplier information; Training a decision tree for each year of the procurement cycle according to the historical procurement feature data, and constructing an initial prediction model according to the leaf proportions in each decision tree; Performing trend analysis on the historical purchasing characteristic data to obtain expected purchasing characteristic data for future periods; Adjust and optimize the initial prediction model according to the expected purchase characteristic data to obtain a target prediction model; The step of training a decision tree for each year of the procurement cycle according to the historical procurement feature data includes: Selecting a target feature from the historical purchase feature data as a root node, and taking other features other than the target feature as leaf nodes to obtain an initial decision tree; For each year of the procurement cycle, the initial decision tree is trained according to the historical procurement feature data of the current year to obtain a decision tree for the current year; The value ranges of other features in different decision trees are different; the target features include market trends and annual demand, and the other features include purchase volume, purchase frequency, production volume, sales volume and inventory; The step of constructing an initial prediction model according to the leaf proportions in each decision tree includes: A prediction model is constructed based on the proportion and numerical range of leaf nodes in each decision tree.

2. The method for predicting purchase data according to claim 1, characterized in that: After the initial prediction model is adjusted and optimized according to the expected purchase characteristic data to obtain a target prediction model, the method further includes: Based on the target prediction model, determining a hit leaf node and a hit decision tree where the hit leaf node is located according to a numerical interval where a known purchase feature value is located; Substitute the known procurement feature value and the proportion of the hit leaf node into the hit decision tree, and retrain the hit decision tree to obtain the target numerical range where the unknown procurement feature value is located; determine the target numerical range as the prediction result of the unknown procurement feature.

3. The method for predicting purchase data according to claim 1, characterized in that: The expected purchasing characteristic data for the future period obtained by performing trend analysis on the historical purchasing characteristic data includes: Perform trend analysis on the historical purchase feature data based on different growth rates to obtain at least two candidate purchase feature data; The expected procurement feature data of the future period is selected from the candidate procurement features according to the information gain rate of each candidate procurement feature data.

4. A purchase data prediction device, characterized in that: The device comprises: A data sorting module, used to construct historical procurement characteristic data of a previous procurement cycle based on procurement forecast related data; the procurement cycle includes at least two years; the procurement forecast related data includes historical procurement data, market trend data and supplier information; A model building module, used to train a decision tree for each year of the procurement cycle according to the historical procurement feature data, and to build an initial prediction model according to the leaf proportion in each decision tree; A trend analysis module, used to perform trend analysis on the historical purchase characteristic data to obtain expected purchase characteristic data for future periods; A model optimization module, used to adjust and optimize the initial prediction model according to the expected purchase characteristic data to obtain a target prediction model; Wherein, the model building module includes: A decision tree construction unit, used to select a target feature from the historical procurement feature data as a root node, and use other features other than the target feature as leaf nodes to obtain an initial decision tree; A decision tree training unit, used for training the initial decision tree for each year of the procurement cycle according to the historical procurement feature data of the current year to obtain a decision tree for the current year; The value ranges of other features in different decision trees are different; the target features include market trends and annual demand, and the other features include purchase volume, purchase frequency, production volume, sales volume and inventory; Wherein, the model building module includes: The model building unit is used to build a prediction model according to the proportion and value range of leaf nodes in each decision tree.

5. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for predicting purchase data according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the procurement data prediction method according to any one of claims 1 to 3 when executed.

Images