Material warehouse-in and warehouse-out control method, computing device and readable storage medium

Abstract

The invention discloses a material warehouse-in and warehouse-out control method, computing equipment and a computer readable storage medium, the material warehouse-in and warehouse-out control method comprises the following steps: obtaining at least one target attribute tag for a material; performing material prediction according to a pre-established classification model and the at least one target attribute tag to obtain target material information; and performing warehouse-out recommendation control according to the target material information. Through the technical scheme of the invention, the attribute tag of the material required by the user is imported into the pre-established classification model, the corresponding target material information can be quickly and accurately determined, and warehouse-out recommendation can be carried out to the user, so that the user can select the required material to carry out warehouse-out operation. Therefore, according to the technical scheme of the invention, a quick material searching function can be realized, so that the material selection requirement of a user is met.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a material entry and exit control method, a computing device, and a computer-readable storage medium. Background Art

[0002] In the current warehouse management model, there are three main ways to search for materials: (1) Search by list: Engineers search for materials at the maintenance location based on the bill of materials of the equipment components and search by material number; (2) Search by atlas: Engineers check the materials on the atlas based on the equipment parts and search from the material library based on the material number; (3) Search by name: Search directly in the warehouse management system based on the material name of the maintenance location, check the material properties, and identify the required materials.

[0003] In the current material search method, in the scenario of a large amount of material information, coupled with a large number of material attributes and high similarity of some materials, engineers rely on traditional methods such as lists, atlases and name searches, which are difficult to meet the needs of fast material search. Therefore, how to achieve fast material search to meet the user's material selection needs is a technical problem that technicians in this field urgently need to solve. Summary of the invention

[0004] The purpose of this application is to provide a material in and out of warehouse control method, computing device and computer-readable storage medium, which can realize rapid search of materials to meet the user's material selection needs.

[0005] To achieve the above objectives: In a first aspect, an embodiment of the present application provides a method for controlling the entry and exit of materials, including: obtaining at least one target attribute label for a material; performing material prediction based on a pre-built classification model and at least one target attribute label to obtain target material information; and performing outbound recommendation control based on the target material information.

[0006] In one embodiment, before the step of performing material prediction based on a pre-built classification model and at least one target attribute label to obtain target material information, the step includes: building a material library, the material library includes at least one association relationship information, the association relationship information includes the association relationship between material information and at least one attribute label; selecting at least one association relationship information from the material library to build a data set, the data set includes one or more of a training set, a validation set, and a test set; and training the classification model framework according to the data set to build a classification model.

[0007] In one embodiment, the classification model framework is trained according to the data set before the step of building the classification model, including: determining multiple decision tree models, wherein each decision tree model corresponds to a different attribute label dimension; and combining the multiple decision tree models to obtain the classification model framework.

[0008] In one embodiment, after the step of training the classification model framework according to the data set to build the classification model, the following steps are included: obtaining newly added material information; performing label prediction according to the classification model and the newly added material information to obtain recommended attribute labels; obtaining newly added association relationship information based on the newly added material information and the recommended attribute labels, and storing the newly added association relationship information in the material library.

[0009] In one implementation, the attribute labels in the association relationship information stored in the material library are labels represented by binary codes obtained by binary transcoding character labels.

[0010] In one embodiment, the step of obtaining at least one target attribute tag for a material includes: obtaining at least one query tag; clustering material information in a material library according to the at least one query tag to obtain a clustering result; outputting at least one attribute tag according to the clustering result; and responding to a selection operation on the output attribute tag to use the selected attribute tag as a target attribute tag.

[0011] In one embodiment, after the step of performing outbound recommendation control based on the target material information, it includes: in response to the outbound confirmation operation of the target material information, associating the target material information with at least one target attribute tag to obtain target association information; importing the target association information into the classification model for regression optimization training to update the classification model.

[0012] In a second aspect, an embodiment of the present application provides a method for controlling the entry and exit of materials, including: obtaining newly added material information; performing label prediction based on a pre-built classification model and the newly added material information to obtain recommended attribute labels; based on the newly added material information and the recommended attribute labels, obtaining newly added association relationship information, and storing the newly added association relationship information in the material library.

[0013] In a third aspect, an embodiment of the present application provides a device, comprising: a processor and a memory storing a computer program, and when the processor runs the computer program, the steps of the material entry and exit control method described in any one of the above items are implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of the material warehousing and in-and-out control method described in any one of the above items are implemented.

[0015] The embodiments of the present application provide a method for controlling the entry and exit of materials, a computing device, and a computer-readable storage medium. The method for controlling the entry and exit of materials includes: obtaining at least one target attribute tag for a material; predicting materials based on a pre-built classification model and at least one target attribute tag to obtain target material information; and performing outbound recommendation control based on the target material information. Through the technical solution of the present application, the attribute tags of the materials required by the user are imported into the pre-built classification model, so that the target material information can be quickly and accurately determined and outbound recommendations can be made to the user, so that the user can select the required materials for outbound operations. In this way, the technical solution of the present application can realize a quick search function for materials to meet the user's material selection needs.

[0016] In addition, the present application also provides a material in-and-out warehouse control method, including: obtaining newly added material information; performing label prediction based on a pre-built classification model and the newly added material information to obtain a recommended attribute label; based on the newly added material information and the recommended attribute label, obtaining newly added association relationship information, and storing the newly added association relationship information in the material library. In this way, the technical solution of this embodiment can label the characteristic attributes of the material through the pre-built classification model, and associate the attribute label with the material information of the material and store it in the material library, so as to realize the rapid determination and selection of the target material information from the material library through the attribute label of the required material. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application. In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor.

[0018] Figure 1 A schematic flow chart of a material entry and exit control method provided in the first embodiment of the present application.

[0019] Figure 2 This is a flow chart of the combined classifier model used as an example in this application.

[0020] Figure 3 It is a flow chart of the material entry and exit control method provided in the second embodiment of the present application.

[0021] Figure 4 It is a schematic diagram of the structure of the computing device provided in an embodiment of the present application.

[0022] The realization of the purpose, functional features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. The above-mentioned drawings have shown clear embodiments of this application, which will be described in more detail later. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0023] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0024] It should be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined by their explanation in the specific embodiment or further combined with the context of the specific embodiment.

[0025] It should be understood that, although the terms first, second, third, etc. may be used to describe various information in this article, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this article, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination". Furthermore, as used in this article, the singular forms "one", "one" and "the" are intended to also include plural forms, unless there is an opposite indication in the context. It should be further understood that the terms "comprising", "including" indicate that there are described features, steps, operations, elements, components, projects, kinds, and / or groups, but do not exclude the existence, occurrence or addition of one or more other features, steps, operations, elements, components, projects, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, “A, B, or C” or “A, B and / or C” means “any of the following: A; B; C; A and B; A and C; B and C; A, B, and C.” An exception to this definition will occur only when a combination of elements, functions, steps, or operations are inherently mutually exclusive in some manner.

[0026] It should be understood that, although the various steps in the flowchart in the embodiment of the present application are displayed in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and it can be performed in other orders. Moreover, at least a portion of the steps in the figure may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and their execution order is not necessarily performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0027] It should be noted that, in this article, step codes such as S11, S12, etc. are used for the purpose of expressing the corresponding content more clearly and concisely, and do not constitute a substantial limitation on the sequence. When implementing it specifically, those skilled in the art may execute S12 first and then S11, etc., but these should all be within the scope of protection of this application.

[0028] It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0029] In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present application, and have no specific meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0030] First embodiment See also Figure 1 , is a material in-and-out storage control method provided in the first embodiment of the present application, which can be executed by a computing device provided in the embodiment of the present application, and the computing device can be implemented in software and / or hardware. Optionally, the computing device is, for example, a terminal device, a server, etc.

[0031] The material in-and-out storage control method provided in this embodiment includes the following steps (for example, steps S11 to S13): Step S11: Obtain at least one target attribute tag for the material.

[0032] In one embodiment, the target attribute tag of the material may represent the characteristic attributes of the material during the production process, storage process or use process.

[0033] In one embodiment, each target attribute tag corresponds to a characteristic attribute, such as a specific material identification (such as name, code, etc.), a specific material price, a specific material price effective time period, a specific material sales organization, a specific material unit, a specific material belonging group, a specific material type, a specific material distribution channel, a specific material substitute, a specific material use, etc., or a combination of one or more of the above.

[0034] In one implementation, the specific material price may represent the current selling price or cost price of the material.

[0035] In one implementation, the specific material price effective time period may represent the validity period of the material price, which usually corresponds to a promotional activity or a price change trend.

[0036] In one implementation, a specific material sales organization may represent an organization or department that is responsible for selling the material.

[0037] In one implementation, the specific material unit may represent a measurement unit of the material, such as piece, kilogram, meter, etc.

[0038] In one embodiment, a specific material belonging group may represent the classification or group to which the material belongs, which is helpful for organization and retrieval.

[0039] In one embodiment, the specific material type may represent the classification of the material, such as raw materials, semi-finished products, finished products, etc.

[0040] In one implementation, a specific material distribution channel may represent a material distribution method, such as direct sales, distributors, e-commerce, etc.

[0041] In one embodiment, a specific material substitute may represent other materials that can replace the current material.

[0042] In one embodiment, the specific material usage may characterize the specific function and effect of the material.

[0043] In one embodiment, step S11: obtaining at least one target attribute tag for a material may include: obtaining at least one query tag; clustering the material information in the material library according to at least one query tag to obtain a clustering result; outputting at least one attribute tag according to the clustering result; and responding to a selection operation on the output attribute tag to use the selected attribute tag as the target attribute tag. In this way, the technical solution of this embodiment can cluster the material information in the material library based on the query tag, so as to cluster the material information with similar attribute tag associations based on the query tag, so as to expand the attribute tags used by the user to search for the target material information according to the attribute tag associations corresponding to these material clusters, so as to facilitate the user to determine more and more appropriate target attribute tags to characterize the user's material needs, so that the determined multiple target attribute tags and the pre-built classification model can be used later to achieve faster and more accurate material prediction, so that the prediction results are more in line with the user's actual material needs, and thus the user can quickly complete the material selection. Therefore, the technical solution of this embodiment can use the query tag as a guide to derive more attribute tags for users to select as target attribute tags, thereby facilitating subsequent accurate material prediction based on the selected multiple target attribute tags, obtaining target material information that meets the actual needs of users, and then realizing a quick search function for materials to meet the user's material selection needs.

[0044] In one embodiment, the step of outputting at least one attribute label according to the clustering results may include: for each material cluster in the clustering results, obtaining a first attribute label set respectively associated with each material cluster; extracting attribute labels that meet preset extraction conditions from each first attribute label set to obtain and output a second attribute label set. Among them, the attribute labels of the preset extraction conditions can represent representative attribute labels in each material cluster. In this way, the technical solution of this embodiment can control the number of attribute labels output, reduce interference items when the user selects the target attribute label, and facilitate the user to quickly determine the target attribute label.

[0045] In one implementation, the clustering process may be implemented based on a K-Means clustering algorithm.

[0046] Step S12: Material prediction is performed based on a pre-built classification model and at least one target attribute label to obtain target material information.

[0047] In one embodiment, the machine learning model (or classification model framework) corresponding to the classification model is, for example, a single vector machine (SVM), a random forest, a gradient boosted decision tree (GBDT), etc., or a combined model framework formed by multiple identical or different single machine learning models.

[0048] In one embodiment, the classification model may be obtained by importing one or more association relationship information into a machine learning model for model training.

[0049] In one embodiment, the association relationship information includes an association relationship between material information and at least one attribute tag.

[0050] In one embodiment, step S12: before performing material prediction based on a pre-built classification model and at least one target attribute label to obtain target material information, it may include: building a material library, the material library includes at least one association relationship information; selecting at least one association relationship information from the material library to build a data set, the data set includes one or more of a training set, a validation set, and a test set; and training the classification model framework according to the data set to build a classification model. In this way, the technical solution of this embodiment can train the classification model framework through the data in the material library to build a classification model, so that the classification model can predict the target material information in the material library that meets the user's material needs according to the target attribute label, and realize the rapid search function of the material information in the material library to meet the user's material selection needs.

[0051] In one embodiment, the step of building a material library may include: determining a label framework for attribute labels; loading material information, and entering attribute labels according to the material information and the label framework to form association relationship information; and storing the association relationship information in the material library.

[0052] In one embodiment, the label frameworks of different material information may be the same, but the specific attribute characteristics represented by the attribute labels under the label framework may be different. For example, the label framework includes material price, material sales organization, material unit, and material use. The multiple attribute labels of material information X under the label framework include material price of 80 yuan / piece, material sales organization of Company B, material unit of piece, and material use of Use 1. The attribute labels of material information Y under the label framework include material price of 10 yuan / meter, material sales organization of Company C, material unit of meter, and material use of Use 2.

[0053] In one embodiment, the tag framework is a manifestation of the attribute tag setting rules. The tag framework can be used as a cornerstone for implementing the material search function. Each material information can be set or associated with an attribute tag according to the tag framework.

[0054] In one embodiment, the specific characteristic attribute represented by each attribute tag can be obtained through sophisticated word segmentation, tokenization and serialization processing, so as to provide a data basis for material retrieval.

[0055] In one implementation, the attribute labels in the association relationship information stored in the material library are labels represented by binary codes obtained by binary transcoding character labels.

[0056] In one implementation, the association relationship information stored in the material library may be an efficient binary transcoding process of the associated attribute tags of the material information, and the tags may be stored in binary form by establishing a bitmap.

[0057] In one embodiment, the steps and methods for processing attribute tags in a database include but are not limited to: (1) Construct a unique identifier: Assign a unique identifier to each attribute tag, which can be a number or a combination of numbers and letters; (2) Binary encoding: Convert the identifier of each attribute tag into binary form. For example, if the identifier of "color" is 1, then its binary encoding is 001; if the identifier of "size" is 2, its binary encoding is 010.

[0058] (3) Bitmap storage: Use bitmap to store these binary codes. Bitmap is a data structure that uses bits to represent data. Each bit can only store two states: 0 or 1. It is very suitable for storing binary codes. For example, if there are 8 different attribute labels, then one byte (8 bits) can be used to store these attribute labels.

[0059] (4) Rule formulation: Formulate a set of rules to define how to map attribute labels to bits in the bitmap. For example, you can stipulate that the first bit in the bitmap represents "color", the second bit represents "size", and so on.

[0060] (5) Encoding mapping: Create a mapping table to map attribute labels to specific locations in the bitmap. In this way, when you need to store or retrieve material attributes, you can directly operate the bitmap through the mapping table.

[0061] In the technical solution of this embodiment, the attribute tags in the material library are stored in binary encoding and bitmap mode, which can significantly reduce the storage space, reduce the storage space, and improve the data processing efficiency and retrieval speed. For example, if there are 1,000 materials, each with 10 attribute tags, the conventional method may need to store 10,000 attribute tags, while the bitmap method may only need to store 1,000 bytes. Therefore, the technical solution of this embodiment can significantly reduce the storage space through the bitmap storage method, and can achieve storage optimization. Specifically, since the bitmap supports fast bit operations, materials with specific attribute tags can be quickly retrieved, because the retrieval speed can be improved; for example, to find all materials with a "color" of red, it is only necessary to check whether the corresponding "color" bit in the bitmap is 1.

[0062] In one embodiment, the step of training the classification model framework according to the data set to build a classification model may include: using the association relationship information in the training set as a sample to train the selected classification model framework to obtain a preliminary classification model; using the validation set data to evaluate and tune the preliminary classification model to update the preliminary classification model; using the test set data to perform a final evaluation on the preliminary classification model to obtain a final classification model after the evaluation passes.

[0063] In one embodiment, before the step of training the classification model framework according to the data set to build the classification model, the steps may include: determining multiple decision tree models, wherein each decision tree model corresponds to a different attribute label dimension; and combining the multiple decision tree models to obtain the classification model framework. In this way, the technical solution of this embodiment can select a corresponding number of decision tree models to combine to build the classification model framework according to the number of divided attribute label dimensions, so that the subsequent use of the classification model built based on the classification model framework can predict the corresponding material information according to the attribute labels under one or more attribute label dimensions.

[0064] In one embodiment, the attribute label dimension can represent a collection of similar characteristic attributes used when describing and classifying materials. For example, the price dimension includes material price labels, material price effective time period labels, etc., and when constructing the classification model framework, a decision tree model (referred to as price model) for classifying price-related content can be determined; for example, the unit dimension includes material unit labels, material outbound minimum unit labels, material outbound maximum unit labels, etc., and when constructing the classification model framework, a decision tree model (referred to as unit model) for classifying material unit-related content can be determined; for example, the use dimension includes material use labels, material usage method labels, etc., and when constructing the classification model framework, a decision tree model (referred to as use model) for classifying material use-related content can be determined; for example, the material organization dimension includes production organization labels, warehouse management organization labels, etc., and when constructing the classification model framework, a decision tree model (referred to as material organization model) for classifying material organization-related content can be determined.

[0065] In one embodiment, the step of combining multiple decision tree models to obtain a classification model framework may include one or more of the following combinations: (1) Data preparation: Divide the training set into multiple complementary subsets, each of which includes attribute labels of a material label dimension; (2) Build a single decision tree model: A decision tree model is independently constructed for each attribute label dimension. Algorithms such as ID3, C4.5, and CART can be used to construct decision trees.

[0066] (3) Model training: Each decision tree model is trained using different combinations of complementary subsets. During the training process, model parameters such as tree depth, splitting criteria, etc. can be adjusted; After training, the remaining complementary subsets were used and a combination of validation methods of different models were used for rigorous validation to ensure the accuracy and reliability of each decision tree model.

[0067] (4) Model evaluation: Evaluate each decision tree model and use indicators such as accuracy, recall, and F1 score to evaluate model performance.

[0068] (5) Determine the combination strategy: Determine how to combine multiple decision tree models into a classification model framework. Common combination strategies include at least one of the following: Voting method: All decision tree models vote on the prediction results of the same sample in the training set, and the final prediction is determined by the category with the most votes; Stacking: Use a meta-learner (such as logistic regression, support vector machine, etc.) to combine the prediction results of various decision tree models.

[0069] (6) Model combination: According to the determined combination strategy, multiple decision tree models are integrated into a whole to obtain the final classification model framework; In one embodiment, the classification model framework obtained above is trained by extracting new samples (i.e., association relationship information) from the data set through repeated regression to promote the classification model framework to approach the optimal solution. Specifically, the difference between the model prediction results and the actual attribute labels is analyzed during the training process, and based on the user's material selection needs, the weight of the key label is increased or the weight of the secondary label is reduced, so that the classification model framework is more suitable for the business and the final classification model is obtained.

[0070] In one embodiment, after obtaining the classification model, an overall model evaluation can be performed, and the combined classification model framework can be evaluated as a whole to ensure the generalization ability of the classification model. In addition, according to the evaluation results, the model parameters are adjusted to optimize the performance of the classification model. Finally, the trained classification model is deployed to the warehouse management system to predict the target material information according to the target attribute label.

[0071] In one embodiment, in the process of adjusting the classification model framework to obtain the final classification model, the BP (backward error propagation) algorithm can also be used to calculate the residual between the predicted value and the label value, and adjust the model parameters in the gradient direction to minimize the prediction error, and through continuous iterative training, gradually approach the global optimal model, thereby obtaining a classification model with better performance (or combined classifier model).

[0072] In one embodiment, the classification model may have forward prediction capability and reverse prediction capability. The forward prediction capability may represent the prediction of matching target material information based on target attribute labels. The reverse prediction capability may represent the prediction of matching attribute labels based on material information.

[0073] In one embodiment, after the step of training the classification model framework according to the data set to build the classification model, the following steps are included: obtaining newly added material information; performing label prediction according to the classification model and the newly added material information to obtain recommended attribute labels; obtaining newly added association relationship information based on the newly added material information and the recommended attribute labels, and storing the newly added association relationship information in the material library. In this way, the technical solution of this embodiment can label the characteristic attributes of the material through the pre-built classification model to assign the material to the attribute label with the highest labeling probability or the attribute label that meets the prediction threshold, and associate the obtained attribute label with the material information of the material and store it in the material library, which is more convenient for quickly realizing the warehousing management of the newly added materials.

[0074] In one embodiment, the step of storing the newly added association relationship information in the material library may include: binary transcoding the attribute tags in the form of character tags in the association relationship information to obtain attribute tags represented by binary codes, and storing the attribute tags represented by binary codes in the material library in the form of bitmaps.

[0075] Step S13: Perform outbound recommendation control based on target material information.

[0076] In one embodiment, step S13: performing outbound recommendation control according to target material information includes: displaying the target material information through a display interface for user selection and confirmation. Optionally, after obtaining one or more target material information, material prediction is performed according to a pre-built classification model and at least one target attribute tag, and one or more target material information can be displayed through a display interface.

[0077] In one embodiment, after the step of performing outbound recommendation control according to the target material information, the method includes: in response to the outbound confirmation operation of the target material information, associating the target material information with at least one target attribute tag to obtain target association information; importing the target association information into the classification model for regression optimization training to update the classification model. In this way, the technical solution of this embodiment can perform a collection process based on the user's selection confirmation result of the target material information and the target attribute tag representing the user's material demand to obtain new samples for the data set, and perform regression training on the classification model through the new samples to further improve the accuracy and applicability of the classification model.

[0078] The material in-and-out control method provided in this embodiment includes: step S11: obtaining at least one target attribute tag for the material; step S12: performing material prediction based on a pre-built classification model and at least one target attribute tag to obtain target material information; step S13: performing outbound recommendation control based on the target material information. Through the technical solution of this embodiment, the attribute tags of the user's required materials are imported into the pre-built classification model, and the target material information can be quickly and accurately determined and recommended to the user, so that the user can select the required materials for outbound operations. In this way, the technical solution of this embodiment can realize the rapid search function of materials to meet the user's material selection needs.

[0079] Based on the same inventive concept as the above-mentioned embodiment, the following example illustrates a scenario in which an engineer searches for repair materials after a device fails and is reported for repair, so as to exemplarily illustrate the material entry and exit control method of the above-mentioned embodiment: In the case where equipment failures are diverse and complex, and the materials required for repairs are delicate and numerous, engineers rely on traditional methods such as lists, atlases, and name searches, which are difficult to meet the needs of quickly searching for materials. Due to the large amount of material data, the simple search process is slow. In addition, there are many material attributes and some materials are highly similar. Engineers need to rely on rich experience and high proficiency to improve the speed and accuracy of finding materials, which significantly increases the difficulty of on-site operations. Once the material selection is wrong due to the engineer's personal experience, it is necessary to go through the process of rejection and reselection, which may cause the equipment failure to take too long to repair, affecting customer satisfaction evaluation. Therefore, when the technical solution of the aforementioned embodiment is applied to the scenario where engineers search for repair materials after equipment failure and repair, the aforementioned shortcomings can be overcome, and the rapid and accurate selection of repair materials can be achieved, so that the repair materials can be obtained in a timely and accurate manner to complete the replacement of the faulty parts of the equipment, thereby improving customer satisfaction. The details are as follows: 1. The equipment involved in the technical solution of this example includes but is not limited to: (1) Relational database: used for data persistence; (2) Non-relational database: used for persisting material attribute tags; (3) Application server PC: used to access relational and non-relational databases and deploy a maintenance platform for material data.

[0080] 2. The workflow of the material data maintenance platform includes: (1) Confirm the label framework of material data, such as material price, material price effective time period, material sales organization, material unit, material belonging group, material type, material distribution channel, etc.

[0081] (2) Loading material data (i.e., the aforementioned material information), and after preprocessing the material data according to the label framework, entering attribute tags associated with the material data; these attribute tags can be character attribute tags obtained after word segmentation, tokenization, and serialization of the material data, and binary transcoding of the character attribute tags is performed to obtain attribute tags represented by binary codes; (3) Based on the decision tree model, a price model, a material organization model, a material use model, a unit model, etc. are constructed, and a combination model method is used for training to obtain a combination classifier model (i.e., the aforementioned classification model). Specifically, multiple iterations of training are performed with the help of the results of the price model, the material organization model, the material use model, the unit model, etc., and the combination classifier model is obtained by adjusting the model weights and optimizing and reducing the loss function in the gradient direction.

[0082] See also Figure 2The combined classifier model is a fusion of multiple independent decision tree models (price model, material organization model, material usage model, unit model) into one. With the training results of these decision tree models, new samples are extracted from the data set through repeated regression to push the combined classifier model closer to a better solution. During the training process, the difference between the model prediction results and the actual labels is analyzed. Based on the needs of maintenance material selection, the weight of key labels is increased or the weight of secondary labels is reduced, making the combined classifier model more suitable for the business. In addition, the combined classifier model also uses the BP (backward error propagation) algorithm to minimize the prediction error by calculating the residual between the predicted value and the label value and adjusting the model parameters in the gradient direction. Through continuous iterative training, the global optimal model is gradually approached, thus obtaining a combined classifier model with better performance.

[0083] (4) Material label prediction processing: Use the trained combined classifier model to predict the newly added material data to obtain the material prediction results, and assign the material data to the attribute label with the highest probability or the attribute label determined by the threshold according to the material prediction results, and convert the assigned attribute label into binary code for mapping to realize the warehousing management of the newly added material data.

[0084] (5) Material search: Analyze the engineer's search intention for maintenance materials, and perform detailed classification and clustering based on the engineer's query tags for maintenance materials to expand and determine the attribute tags. Load the combined classifier model, and recommend the best maintenance materials for engineers to choose based on the residual value calculation results of the determined attribute tags and specific distance rules. (5) Regression training based on the maintenance materials selected by the engineer and the determined attribute labels. New samples are formed based on the maintenance materials selected by the engineer and the determined attribute labels. The new samples are used to perform regression training on the combined classifier model to improve the accuracy and applicability of the combined classifier model.

[0085] The technical solution of this example can obtain a combined classifier model with good prediction performance by labeling the characteristic attributes represented by material data, and through multi-model combination training of the combined classifier model and regression training, so that new maintenance materials can be quickly labeled according to the combined classifier model to achieve rapid warehousing, and efficient search for maintenance materials can be achieved according to the combined classifier model and query labels, thereby achieving rapid and accurate selection of maintenance materials, so as to quickly control the outbound delivery, and realize timely collection of maintenance materials to complete the replacement of faulty parts of equipment, thereby improving the efficiency of fault handling and the satisfaction with fault handling.

[0086] Second embodiment See also Figure 3 , is a material in-and-out storage control method provided in the second embodiment of the present application, which can be executed by a computing device provided in the embodiment of the present application, and the computing device can be implemented in software and / or hardware. Optionally, the computing device is, for example, a terminal device, a server, etc.

[0087] This embodiment provides a material entry and exit control method, including the following steps (for example, steps S21 to S23): Step S21: Obtain the newly added material information.

[0088] In one embodiment, the material information may represent detailed data and characteristics of the material, and includes various characteristic attributes and related information describing the material.

[0089] Step S22: Perform label prediction based on the pre-built classification model and the newly added material information to obtain recommended attribute labels.

[0090] In one implementation, the specific implementation of building the classification model can refer to the content related to the classification model described in the first embodiment, which will not be repeated here.

[0091] Step S23: Based on the newly added material information and the recommended attribute tags, newly added association relationship information is obtained, and the newly added association relationship information is stored in the material library.

[0092] The technical solution of this embodiment and the technical solution of the first embodiment can be combined arbitrarily under the premise that there is no contradiction, and will not be described in detail here.

[0093] The present embodiment provides a material entry and exit control method including: step S21: obtaining newly added material information; step S22: performing label prediction based on a pre-built classification model and newly added material information to obtain a recommended attribute label; step S23: obtaining newly added association relationship information based on the newly added material information and the recommended attribute label, and storing the newly added association relationship information in the material library. In this way, the technical solution of the present embodiment labels the characteristic attributes of the material through a pre-built classification model, and associates the attribute label with the material information of the material and stores it in the material library, so as to achieve rapid entry into the warehouse. Furthermore, the technical solution of the present embodiment facilitates the rapid determination and selection of target material information from the material library through the attribute label of the required material, so as to achieve rapid exit of the material.

[0094] Based on the same inventive concept as the above embodiments, the present application embodiment provides a computing device, such as Figure 4 As shown, the device includes: a processor 310 and a memory 311 storing a computer program; wherein, Figure 4The processor 310 shown in the figure is not used to indicate that the number of the processor 310 is one, but is only used to indicate the positional relationship of the processor 310 relative to other devices. In actual applications, the number of the processor 310 may be one or more; similarly, Figure 4 The memory 311 shown in the figure has the same meaning, that is, it is only used to refer to the position relationship of the memory 311 relative to other devices. In practical applications, the number of memories 311 can be one or more. When the processor 310 runs the computer program, the material in and out control method provided by the first embodiment or the second embodiment applied to the above-mentioned device is implemented, for example Figure 1 Step S11 shown: obtaining at least one target attribute tag for a material; Step S12: performing material prediction based on a pre-built classification model and at least one target attribute tag to obtain target material information; Step S13: performing outbound recommendation control based on the target material information. Or for example Figure 3 Step S21 shown: obtaining newly added material information; step S22: performing label prediction based on the pre-built classification model and the newly added material information to obtain recommended attribute labels; step S23: obtaining newly added association relationship information based on the newly added material information and the recommended attribute labels, and storing the newly added association relationship information in the material library.

[0095] The device may also include: at least one network interface 312. The various components in the device are coupled together via a bus system 313. It is understood that the bus system 313 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 313 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 4 Various buses are labeled as bus system 313.

[0096] The memory 311 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory may be a disk memory or a tape memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAMbus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 311 described in the embodiments of the present application is intended to include but is not limited to these and any other suitable types of memories.

[0097] The memory 311 in the embodiment of the present application is used to store various types of data to support the operation of the device. Examples of these data include: any computer program used to operate on the device, such as an operating system and an application program. Among them, the operating system includes various system programs, such as a framework layer, a core library layer, a driver layer, etc., which are used to implement various basic services and handle hardware-based tasks. The application program may include various applications, such as a media player (MediaPlayer), a browser (Browser), etc., which are used to implement various application services. Here, the program that implements the method of the first embodiment or the second embodiment of the present application may be included in the application program.

[0098] Based on the same inventive concept as the above-mentioned embodiment, this embodiment further provides a computer-readable storage medium, in which a computer program is stored. The computer-readable storage medium may be a ferromagnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); it may also be various devices including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc. When the computer program stored in the computer-readable storage medium is executed by a processor, the material in and out control method provided by the first embodiment or the second embodiment is implemented. The specific steps implemented when the computer program is executed by the processor are, for example, Figure 1 Step S11 shown: obtaining at least one target attribute tag for a material; Step S12: performing material prediction based on a pre-built classification model and at least one target attribute tag to obtain target material information; Step S13: performing outbound recommendation control based on the target material information. Or for example Figure 3Step S21 shown: obtaining newly added material information; step S22: performing label prediction based on the pre-built classification model and the newly added material information to obtain recommended attribute labels; step S23: obtaining newly added association relationship information based on the newly added material information and the recommended attribute labels, and storing the newly added association relationship information in the material library.

[0099] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0100] In this document, the terms "comprises," "comprising," or any other variations thereof, are intended to cover a non-exclusive inclusion of elements other than those listed and may also include additional elements not expressly listed.

[0101] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A material in and out control method, characterized in that: include: Obtain at least one target attribute tag for the material; Perform material prediction based on a pre-built classification model and at least one of the target attribute labels to obtain target material information; The outbound recommendation control is performed according to the target material information.

2. The method according to claim 1, characterized in that: Before the step of performing material prediction according to the pre-built classification model and at least one of the target attribute labels to obtain target material information, the method includes: Building a material library, wherein the material library includes at least one association relationship information, wherein the association relationship information includes an association relationship between material information and at least one attribute tag; Select at least one of the association relationship information from the material library to construct a data set, wherein the data set includes one or more of a training set, a validation set, and a test set; The classification model framework is trained according to the data set to build the classification model.

3. The method according to claim 2, characterized in that Before the step of training the classification model framework according to the data set to build the classification model, the step includes: Determine a plurality of decision tree models, wherein each of the decision tree models corresponds to a different attribute label dimension; A plurality of the decision tree models are combined to obtain the classification model framework.

4. The method according to claim 2, characterized in that: After the step of training the classification model framework according to the data set to build the classification model, the following steps are included: Get the newly added material information; Performing label prediction based on the classification model and the newly added material information to obtain a recommended attribute label; Based on the newly added material information and the recommended attribute tag, the newly added association relationship information is acquired, and the newly added association relationship information is stored in the material library.

5. The method according to claim 4, characterized in that The attribute labels in the association relationship information stored in the material library are labels represented by binary codes obtained by binary transcoding character labels.

6. The method according to claim 2, characterized in that The step of obtaining at least one target attribute tag for the material includes: Get at least one query tag; performing clustering processing on the material information in the material library according to at least one of the query tags to obtain a clustering result; Outputting at least one attribute label according to the clustering result; In response to a selection operation on the outputted attribute tag, the selected attribute tag is used as the target attribute tag.

7. The method according to claim 1, characterized in that After the step of performing outbound recommendation control according to the target material information, the method further comprises: In response to a delivery confirmation operation on the target material information, associating the target material information with at least one of the target attribute tags to obtain target association information; The target association information is imported into the classification model for regression optimization training to update the classification model.

8. A material in and out control method, characterized in that: include: Get the newly added material information; Perform label prediction based on the pre-built classification model and the newly added material information to obtain recommended attribute labels; Based on the newly added material information and the recommended attribute tag, newly added association relationship information is acquired, and the newly added association relationship information is stored in a material library.

9. A computing device, characterized in that include: A processor and a memory storing a computer program, when the processor runs the computer program, implements the steps of the material entry and exit control method described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the steps of the material entry and exit control method described in any one of claims 1 to 8 are implemented.

Images