CN117522256A - Purchasing screening method, system, storage medium and terminal based on supplier portrait - Google Patents
Purchasing screening method, system, storage medium and terminal based on supplier portrait Download PDFInfo
- Publication number
- CN117522256A CN117522256A CN202310973180.0A CN202310973180A CN117522256A CN 117522256 A CN117522256 A CN 117522256A CN 202310973180 A CN202310973180 A CN 202310973180A CN 117522256 A CN117522256 A CN 117522256A
- Authority
- CN
- China
- Prior art keywords
- supplier
- purchasing
- rating
- evaluation index
- provider
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012216 screening Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000011156 evaluation Methods 0.000 claims abstract description 194
- 239000011159 matrix material Substances 0.000 claims description 39
- 238000013507 mapping Methods 0.000 claims description 34
- 238000013139 quantization Methods 0.000 claims description 20
- 230000006870 function Effects 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 abstract description 10
- 238000012545 processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000007475 c-index Methods 0.000 description 2
- 238000009472 formulation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003638 chemical reducing agent Substances 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a purchasing screening method, a purchasing screening system, a purchasing screening storage medium and a purchasing screening terminal based on provider portrait, wherein the purchasing screening method comprises the following steps: when receiving a to-be-processed purchasing demand, acquiring a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier; inputting the evaluation index parameters of the suppliers and the index score data of the evaluation index parameters into a pre-trained supplier rating model, and outputting rating results of each evaluation index; according to the rating results of the evaluation indexes, making a supplier purchase portrait of each supplier; and screening target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers. According to the method and the system, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a purchasing screening method, a purchasing screening system, a storage medium and a purchasing screening terminal based on provider portraits.
Background
Along with the increasing number of suppliers on the upstream of a supply chain of a host factory, the strategic partnership between the host factory and a large number of suppliers is established, and purchasing information between the two parties is controlled and optimized through information means, so that the purchasing cost of the host factory and the delay cost for clients can be further reduced, the value of the clients is increased, and the profit margin of the host factory is improved.
At present, in the production scheduling process of the speed reducer, the main machine factory mostly relies on the oral promise of a plurality of suppliers to make a purchase plan, for example, a purchasing person can make and adjust the purchase plan according to the purchase order exchange period and the oral promise. However, uncertainty of the supplier performance can cause deviation to the establishment and execution of the purchase plan, so that problems of delay of product delay or increase of production cost are caused in the host factory, and the purchase accuracy of the host factory is reduced.
Disclosure of Invention
The embodiment of the application provides a purchase screening method, a system, a storage medium and a terminal based on provider portrait. The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an extensive overview and is intended to neither identify key/critical elements nor delineate the scope of such embodiments. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented later.
In a first aspect, an embodiment of the present application provides a purchase screening method based on a vendor portrait, where the method includes:
when receiving a to-be-processed purchasing demand, acquiring a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier in a supplier list; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
inputting the evaluation index parameters of the suppliers and the index score data of the evaluation index parameters into a pre-trained supplier rating model, and outputting the rating results of the evaluation indexes corresponding to each supplier;
according to the rating results of the evaluation indexes corresponding to each provider, making a provider purchase portrait of each provider;
and screening out target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers.
Optionally, before receiving the purchase demand to be processed, the method further includes:
Acquiring a historical purchase order of each provider in a provider list;
matching purchase data in the historical purchase orders for each preset evaluation index to obtain supplier evaluation index parameters corresponding to the historical purchase orders of suppliers; the supplier evaluation index parameters comprise quality index parameters, cost index parameters, delivery index parameters, service index parameters, technical index parameters, asset index parameters and employee flow index parameters;
establishing an order comment template corresponding to the historical purchase order of each provider according to the provider evaluation index parameter, and pushing the order comment template to a client for evaluation;
collecting comment data sets commented on historical purchase orders of each supplier in a preset time period through a big data technology, calculating index score data of each evaluation index according to the comment data sets, and generating index score data corresponding to the historical purchase orders of each supplier;
and establishing an association relationship between the historical purchase orders of each provider and the corresponding provider evaluation index parameters and index score data.
Optionally, the step of formulating the purchase portraits of the suppliers according to the rating results of the respective evaluation indexes corresponding to the suppliers includes:
Acquiring a plurality of preset purchasing strategies, wherein the purchasing strategies at least comprise a first purchasing strategy with good order performance, a second purchasing strategy with short delivery period and high quality and a third purchasing strategy with long delivery period and low cost;
determining a plurality of first indexes required by a first purchasing strategy, a plurality of second indexes required by a second purchasing strategy and a plurality of third indexes required by a third purchasing strategy in the evaluation index parameters of the suppliers;
matching the rating results corresponding to each first index with the rating results corresponding to each second index and matching the rating results corresponding to each third index in the rating results of each evaluation index corresponding to each supplier;
according to the rating result corresponding to each first index, calculating first similarity between each provider and the first purchasing strategy, determining first providers with the first similarity larger than a preset threshold, and taking the first purchasing strategy as a provider purchasing portrait of each first provider;
calculating a second similarity between each supplier and a second purchasing strategy according to the rating result corresponding to each second index, determining a second supplier with the second similarity larger than a preset threshold value, and taking the second purchasing strategy as a supplier purchasing portrait of each second supplier;
And calculating third similarity between each supplier and the third purchasing strategy according to the rating result corresponding to each third index, determining the third supplier with the third similarity larger than a preset threshold, and taking the third purchasing strategy as a supplier purchasing portrait of each third supplier.
Optionally, screening the target suppliers according with the purchasing requirement to be processed based on the purchasing portraits of the suppliers includes:
analyzing a target purchasing strategy of purchasing requirements to be processed;
when the target purchasing strategy is a first purchasing strategy with good order performance, calculating an intersection between a purchasing portrait of a supplier of each supplier and the first purchasing strategy, and determining suppliers with non-empty intersections; or,
when the target purchasing strategy is a second purchasing strategy with short period and high quality, calculating an intersection between the purchasing portraits of the suppliers and the second purchasing strategy, and determining suppliers with non-empty intersection; or,
when the target purchasing strategy is a third purchasing strategy with long period and low cost, calculating an intersection between the purchasing portraits of the suppliers and the third purchasing strategy, and determining suppliers with non-empty intersection;
When the provider with the intersection not being empty is one, determining the provider with the intersection not being empty as a target provider of the purchasing demand to be processed; or when the number of suppliers with non-empty intersections is a plurality of suppliers, calculating the purchasing cost of each supplier according to the purchasing demand to be processed, and taking the supplier with the lowest purchasing cost as the target supplier of the purchasing demand to be processed.
Optionally, generating the pre-trained vendor rating model comprises:
a plurality of historical evaluation index parameters of the historical real orders of each provider are formulated;
establishing a scoring true value of each historical evaluation index parameter;
marking a preset rating label on the scoring true value of each historical evaluation index parameter to obtain each historical evaluation index parameter of each provider and rating label mapping data thereof;
establishing a parameter matrix of each provider according to each historical evaluation index parameter and the rating label mapping data thereof;
establishing a provider rating model by adopting a neural network, inputting a parameter matrix of each provider into the provider rating model, and outputting a model loss value;
when the model loss value reaches the minimum, generating a pre-trained provider rating model; alternatively, when the model loss value does not reach the minimum, the model loss value is back-propagated to update the model parameters of the provider rating model, and the step of inputting the parameter matrix of each provider into the provider rating model is continued until the model loss value reaches the minimum.
Optionally, establishing a parameter matrix of each provider according to each historical evaluation index parameter and the rating label mapping data thereof, including:
inputting each historical evaluation index parameter and rating label mapping data into an encoder, and quantizing each historical evaluation index parameter and rating label mapping data into 11 when the real part and the imaginary part of the encoder at the current time are positive numbers, or quantizing each historical evaluation index parameter and rating label mapping data into 10 when the real part of the encoder at the current time is positive numbers and the imaginary part of the encoder is negative numbers, or quantizing each historical evaluation index parameter and rating label mapping data into 01 when the real part and the imaginary part of the encoder at the current time are negative numbers, or quantizing each historical evaluation index parameter and rating label mapping data into 00 when the real part and the imaginary part of the encoder at the current time are both negative numbers, and outputting a first quantization result of each historical evaluation index parameter and a second quantization result of the rating label thereof after quantization is finished;
carrying out binary conversion on the first quantization result of each historical evaluation index parameter to obtain a plurality of unsigned first integer values;
Carrying out binary conversion on the second quantization result of the rating label of each historical evaluation index parameter to obtain a plurality of unsigned second integer values;
and combining the plurality of unsigned first integer values and the plurality of unsigned second integer values into a feature matrix to obtain a parameter matrix of each provider.
Optionally, the model loss function expression of the vendor rating model is:
wherein lambda is the weight of the similarity matrix loss function, p and q are the vector representations output after the parameter matrix of each provider enters the normalization layer, and i is each element in the parameter matrix;
the normalization layer is defined as:wherein,
vector Z is the vector of the output of the full link layer, q i Representing normalized probability of ith element in parameter matrix, Z i Ith dimension, Z, representing full link layer output vector j The j-th dimension representing the full connection layer output vector, T is a parameter controlling the output probability smoothness.
In a second aspect, embodiments of the present application provide a purchase screening system based on vendor representation, the system comprising:
the data acquisition module is used for acquiring the supplier evaluation index parameters and index score data thereof associated with the historical purchase orders of each supplier in the supplier list when receiving the purchase demand to be processed; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
The rating result output module is used for inputting the evaluation index parameters of the suppliers and the index score data thereof into a pre-trained supplier rating model and outputting the rating result of each evaluation index corresponding to each supplier;
the supplier purchase portrait making module is used for making a supplier purchase portrait of each supplier according to the rating results of the evaluation indexes corresponding to each supplier;
and the supplier screening module is used for screening out target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers.
In a third aspect, embodiments of the present application provide a computer storage medium having stored thereon a plurality of instructions adapted to be loaded by a processor and to perform the above-described method steps.
In a fourth aspect, embodiments of the present application provide a terminal, which may include: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method steps described above.
The technical scheme provided by the embodiment of the application can comprise the following beneficial effects:
in the embodiment of the application, when receiving a to-be-processed purchasing demand, a purchasing screening system based on a supplier portrait acquires a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier, inputs the supplier evaluation index parameter and the index score data thereof into a pre-trained supplier rating model, outputs a rating result of each evaluation index, formulates a supplier purchasing portrait of each supplier according to the rating result of each evaluation index, and finally screens a target supplier meeting the to-be-processed purchasing demand based on the supplier purchasing portrait of each supplier. According to the method and the system, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
FIG. 1 is a flow chart of a purchasing screening method based on vendor portraits provided in an embodiment of the present application;
FIG. 2 is a graph of a relationship between a purchase strategy and a rating result provided by an embodiment of the present application;
FIG. 3 is a schematic diagram of a training process for a purchase screening model based on vendor representation provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of a purchasing screening system based on vendor representation according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a terminal according to an embodiment of the present application.
Detailed Description
The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them.
It should be understood that the described embodiments are merely some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of systems and methods that are consistent with aspects of the invention as detailed in the accompanying claims.
In the description of the present invention, it should be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art. Furthermore, in the description of the present invention, unless otherwise indicated, "a plurality" means two or more. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship.
The application provides a purchasing screening method, a purchasing screening system, a purchasing screening storage medium and a purchasing screening terminal based on provider portrait, which are used for solving the problems existing in the related technical problems. In the technical scheme provided by the application, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved, and the following detailed description is given by adopting an exemplary embodiment.
The purchasing screening method based on the provider portrait according to the embodiment of the present application will be described in detail with reference to fig. 1 to fig. 3. The method may be implemented in dependence on a computer program and may be run on a vendor portrayal-based procurement screening system based on von neumann systems. The computer program may be integrated in the application or may run as a stand-alone tool class application.
Referring to fig. 1, a flow chart of a purchasing screening method based on vendor portrait is provided in an embodiment of the present application. As shown in fig. 1, the method of the embodiment of the present application may include the following steps:
s101, when receiving a to-be-processed purchasing demand, acquiring a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier in a supplier list; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
The purchasing requirement to be processed is a part purchasing list formulated by a current host factory or enterprise according to the production requirement of a customer order. The vendor list is a third party partner that provides parts for the current hosting factory or enterprise. The historical purchase order is historical transaction data for a current host factory or business to purchase components in front of a third party partner for a period of time. The supplier evaluation index parameters are determined according to historical transaction data and various preset evaluation indexes.
Typically, each preset rating scale is a vendor rating 7 standard (see link https:// wenku.baidu.com/view/0f921E8E 624 ccbf121bd7375a 417c6E 3.html.
In the embodiment of the present application, before receiving the pending purchase demand, the provider assessment index parameter and the index score data thereof need to be formulated for the historical purchase order of each provider. Firstly, acquiring a historical purchase order of each supplier in a supplier list, and then matching purchase data in the historical purchase order for each preset evaluation index to obtain a supplier evaluation index parameter corresponding to the historical purchase order of the supplier; the provider evaluation index parameters comprise quality index parameters, cost index parameters, delivery index parameters, service index parameters, technical index parameters, asset index parameters and employee flow index parameters, then an order comment template corresponding to the historical purchase orders of each provider is established according to the provider evaluation index parameters and pushed to a client for evaluation, comment data sets which are commented on by the historical purchase orders of each provider in a preset time period are collected through a big data technology, index score data of each evaluation index are calculated according to the comment data sets, index score data corresponding to the historical purchase orders of each provider are generated, and finally the association relation between the historical purchase orders of each provider and the corresponding provider evaluation index parameters and index score data is established.
In one possible implementation, when the host factory or enterprise determines the purchasing requirement to be processed according to the production requirement, the purchasing requirement to be processed is uploaded to a purchasing screening system based on the supplier portrait, and when the purchasing requirement to be processed is received, the system loads a supplier list and acquires supplier evaluation index parameters and index score data thereof associated with the historical purchasing order of each supplier in the supplier list.
S102, inputting the evaluation index parameters of the suppliers and the index score data of the evaluation index parameters into a pre-trained supplier rating model, and outputting the rating results of the evaluation indexes corresponding to each supplier;
wherein the pre-trained vendor rating model is a mathematical model capable of determining a rating result.
In the embodiment of the application, when a pre-trained provider rating model is generated, firstly, a plurality of historical evaluation index parameters of each provider historical real order are formulated, then, a scoring true value of each historical evaluation index parameter is established, then, a preset rating label is marked on the scoring true value of each historical evaluation index parameter to obtain each historical evaluation index parameter of each provider and rating label mapping data thereof, secondly, a parameter matrix of each provider is established according to each historical evaluation index parameter and rating label mapping data thereof, a provider rating model is established by adopting a neural network, the parameter matrix of each provider is input into the provider rating model, a model loss value is output, and finally, when the model loss value reaches the minimum, the pre-trained provider rating model is generated; alternatively, when the model loss value does not reach the minimum, the model loss value is back-propagated to update the model parameters of the provider rating model, and the step of inputting the parameter matrix of each provider into the provider rating model is continued until the model loss value reaches the minimum. The scoring reality value is determined by a user according to the evaluation index parameters during model training.
Specifically, when a parameter matrix of each supplier is established according to each historical evaluation index parameter and rating label mapping data thereof, firstly inputting each historical evaluation index parameter and rating label mapping data into an encoder, and when the real part and the imaginary part of the encoder at the current time are positive numbers, quantizing each historical evaluation index parameter and rating label mapping data into 11, or when the real part of the encoder at the current time is positive numbers and the imaginary part of the encoder is negative numbers, quantizing each historical evaluation index parameter and rating label mapping data into 10, or when the real part of the encoder at the current time is negative numbers and the imaginary part of the encoder is positive numbers, quantizing each historical evaluation index parameter and rating label mapping data into 01, or when the real part and the imaginary part of the encoder at the current time are both negative numbers, quantizing each historical evaluation index parameter and rating label mapping data into 00, and outputting a first quantization result of each historical evaluation index parameter and a second quantization result of the rating label thereof after quantization is finished; then carrying out binary conversion on the first quantization result of each historical evaluation index parameter to obtain a plurality of unsigned first integer values; secondly, carrying out binary conversion on a second quantization result of the rating label of each historical evaluation index parameter to obtain a plurality of unsigned second integer values; and finally, forming a feature matrix by the plurality of unsigned first integer values and the plurality of unsigned second integer values to obtain a parameter matrix of each provider.
Specifically, the model loss function expression of the vendor rating model is:
wherein lambda is the weight of the similarity matrix loss function, p and q are the vector representations output after the parameter matrix of each provider enters the normalization layer, and i is each element in the parameter matrix;
the normalization layer is defined as:wherein,
vector Z is the vector of the output of the full link layer, q i Representing normalized probability of ith element in parameter matrix, Z i Ith dimension, Z, representing full link layer output vector j The j-th dimension representing the full connection layer output vector, T is a parameter controlling the output probability smoothness.
In one possible implementation manner, when the provider evaluation index parameter and the index score data thereof associated with the historical purchase order of each provider are obtained based on step S101, the provider evaluation index parameter and the index score data thereof may be input into a pre-trained provider rating model for processing, and the rating result of each evaluation index is output after the model processing is finished.
S103, according to the rating results of the evaluation indexes corresponding to each provider, making a provider purchase portrait of each provider;
in the embodiment of the application, when a provider purchase portrait of each provider is formulated according to the rating results of each evaluation index corresponding to each provider, a plurality of pre-formulated purchase strategies are firstly obtained, wherein the purchase strategies at least comprise a first purchase strategy with good order performance, a second purchase strategy with short delivery period and high quality and a third purchase strategy with long delivery period and low cost; then, determining a plurality of first indexes required by a first purchasing strategy, a plurality of second indexes required by a second purchasing strategy and a plurality of third indexes required by a third purchasing strategy in the supplier evaluating index parameters; secondly, matching the rating result corresponding to each first index with the rating result corresponding to each second index and matching the rating result corresponding to each third index in the rating results of each evaluation index corresponding to each supplier; calculating first similarity between each supplier and the first purchasing strategy according to the rating result corresponding to each first index, determining first suppliers with the first similarity larger than a preset threshold, and taking the first purchasing strategy as a supplier purchasing portrait of each first supplier; calculating a second similarity between each supplier and a second purchasing strategy according to the rating result corresponding to each second index, determining a second supplier with the second similarity larger than a preset threshold value, and taking the second purchasing strategy as a supplier purchasing portrait of each second supplier; and finally, calculating a third similarity between each supplier and the third purchasing strategy according to the rating result corresponding to each third index, determining a third supplier with the third similarity larger than a preset threshold, and taking the third purchasing strategy as a supplier purchasing portrait of each third supplier.
Specifically, the good order performance is that the ratio of the transaction performance times to the total transaction times calculated according to the index parameters is in a preset first range, the delivery period is short, the quality is high, for example, the duration between the actual delivery date and the contract delivery date calculated according to the index parameters is longer than the preset duration, the yield of delivery is greater than the preset threshold, the delivery period is long, the cost is low, for example, the duration of the actual delivery date calculated according to the index parameters exceeds the contract delivery date is longer than the preset threshold, and the total price of the order is smaller than the preset amount.
For example, as shown in fig. 2, fig. 2 is a relationship diagram of matching between a purchasing strategy and rating results provided in the present application, first determining corresponding index parameters for different purchasing strategies, then matching corresponding rating results for the index parameters in the rating results of each evaluation index corresponding to each supplier, and if some suppliers in the matched rating results tend to be the first purchasing strategy, using the first purchasing strategy as a purchasing portrait of the supplier; if some suppliers in the matched rating results are prone to the second purchasing strategy, the second purchasing strategy is used as a purchasing portrait of the supplier; if some suppliers in the matched rating results tend to be in the third purchasing strategy, the third purchasing strategy is taken as the purchasing portrait of the supplier.
For example, the rating result of each evaluation index corresponding to the Y supplier is that the q index level is a, the C index level is B, and the q index and the C index belong to the second purchasing policy, and the level a and the level B are larger than the preset level C, so that it can be determined that the Y supplier is inclined to the second purchasing policy, and at this time, it can be determined whether the supplier purchasing image of the Y supplier is the second purchasing policy by calculating the similarity.
S104, screening out target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers.
In the embodiment of the application, when screening out target suppliers meeting the purchasing demand to be processed based on the purchasing portraits of suppliers, firstly analyzing the target purchasing strategy of the purchasing demand to be processed; then when the target purchasing strategy is a first purchasing strategy with good order performance, calculating an intersection between the purchasing portraits of the suppliers and the first purchasing strategy, and determining suppliers with non-empty intersections; or when the target purchasing strategy is a second purchasing strategy with short period and high quality, calculating the intersection between the purchasing portraits of the suppliers and the second purchasing strategy, and determining suppliers with non-empty intersection; or when the target purchasing strategy is a third purchasing strategy with long period and low cost, calculating the intersection between the purchasing portraits of the suppliers and the third purchasing strategy, and determining suppliers with non-empty intersection; when the provider with the intersection not being empty is one, determining the provider with the intersection not being empty as a target provider of the purchasing demand to be processed; or when the number of suppliers with non-empty intersections is a plurality of suppliers, calculating the purchasing cost of each supplier according to the purchasing demand to be processed, and taking the supplier with the lowest purchasing cost as the target supplier of the purchasing demand to be processed.
For example, a provider with a high q-index, d-index, and c-index will be assigned a low-grade, and therefore a small-lot purchase order with a high quality requirement and a short delivery period. Suppliers with low q-index and d-index levels will have high c-index levels assigned to their large volume orders.
Further, it is also possible to make adjustments to purchase catalogs, supplier catalogs, purchase strategies, action plans for each supplier based on the supplier purchase portraits of each supplier, or to continuously improve bad suppliers, eliminate failed suppliers, and develop new suppliers.
In the embodiment of the application, when receiving a to-be-processed purchasing demand, a purchasing screening system based on a supplier portrait acquires a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier, inputs the supplier evaluation index parameter and the index score data thereof into a pre-trained supplier rating model, outputs a rating result of each evaluation index, formulates a supplier purchasing portrait of each supplier according to the rating result of each evaluation index, and finally screens a target supplier meeting the to-be-processed purchasing demand based on the supplier purchasing portrait of each supplier. According to the method and the system, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved.
Referring to fig. 3, a flow chart of a model training method is provided in an embodiment of the present application. As shown in fig. 3, the method of the embodiment of the present application may include the following steps:
s201, formulating a plurality of historical evaluation index parameters of each provider historical real order;
s202, establishing a scoring true value of each historical evaluation index parameter;
s203, marking a preset rating label on the scoring true value of each historical evaluation index parameter to obtain each historical evaluation index parameter of each provider and rating label mapping data thereof;
s204, establishing a parameter matrix of each provider according to each historical evaluation index parameter and the rating label mapping data thereof;
s205, a provider rating model is established by adopting a neural network, a parameter matrix of each provider is input into the provider rating model, and a model loss value is output;
s206, when the model loss value reaches the minimum, generating a pre-trained provider rating model; alternatively, when the model loss value does not reach the minimum, the model loss value is back-propagated to update the model parameters of the provider rating model, and the step of inputting the parameter matrix of each provider into the provider rating model is continued until the model loss value reaches the minimum.
In the embodiment of the application, when receiving a to-be-processed purchasing demand, a purchasing screening system based on a supplier portrait acquires a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier, inputs the supplier evaluation index parameter and the index score data thereof into a pre-trained supplier rating model, outputs a rating result of each evaluation index, formulates a supplier purchasing portrait of each supplier according to the rating result of each evaluation index, and finally screens a target supplier meeting the to-be-processed purchasing demand based on the supplier purchasing portrait of each supplier. According to the method and the system, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved.
The following are system embodiments of the present invention that may be used to perform method embodiments of the present invention. For details not disclosed in the system embodiments of the present invention, please refer to the method embodiments of the present invention.
Referring to FIG. 4, a schematic diagram of a purchasing screening system based on vendor representation according to an exemplary embodiment of the present invention is shown. The provider representation-based purchase screening system may be implemented as all or part of the terminal by software, hardware, or a combination of both. The system 1 comprises a data acquisition module 10, a rating result output module 20, a supplier purchase portrayal formulation module 30 and a supplier screening module 40.
The data acquisition module 10 is configured to acquire, when receiving a purchase demand to be processed, a supplier evaluation index parameter and index score data thereof associated with a historical purchase order of each supplier in the supplier list; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
the rating result output module 20 is configured to input the provider rating index parameter and the index score data thereof into a pre-trained provider rating model, and output a rating result of each rating index corresponding to each provider;
A supplier purchase portrait making module 30, configured to make a supplier purchase portrait of each supplier according to the rating results of the respective evaluation indexes corresponding to each supplier;
and the supplier screening module 40 is used for screening out target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers.
It should be noted that, in the purchase screening system based on the vendor portrait provided in the foregoing embodiment, when the purchase screening method based on the vendor portrait is executed, only the division of the functional modules is used for illustrating, in practical application, the functional allocation may be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the purchasing screening system based on the provider portrait provided in the above embodiment belongs to the same concept as the purchasing screening method based on the provider portrait, which shows a detailed implementation process in the method embodiment, and is not described herein.
The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.
In the embodiment of the application, when receiving a to-be-processed purchasing demand, a purchasing screening system based on a supplier portrait acquires a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier, inputs the supplier evaluation index parameter and the index score data thereof into a pre-trained supplier rating model, outputs a rating result of each evaluation index, formulates a supplier purchasing portrait of each supplier according to the rating result of each evaluation index, and finally screens a target supplier meeting the to-be-processed purchasing demand based on the supplier purchasing portrait of each supplier. According to the method and the system, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved.
The invention also provides a computer readable medium, on which program instructions are stored, which when executed by a processor implement the purchasing screening method based on vendor portraits provided by the above-mentioned method embodiments.
The invention also provides a computer program product containing instructions which, when run on a computer, cause the computer to perform the supplier portrait based purchase screening method of the various method embodiments described above.
Referring to fig. 5, a schematic structural diagram of a terminal is provided in an embodiment of the present application. As shown in fig. 5, terminal 1000 can include: at least one processor 1001, at least one network interface 1004, a user interface 1003, a memory 1005, at least one communication bus 1002.
Wherein the communication bus 1002 is used to enable connected communication between these components.
The user interface 1003 may include a Display screen (Display) and a Camera (Camera), and the optional user interface 1003 may further include a standard wired interface and a wireless interface.
The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), among others.
Wherein the processor 1001 may include one or more processing cores. The processor 1001 connects various parts within the overall electronic device 1000 using various interfaces and lines, performs various functions of the electronic device 1000 and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 1005, and invoking data stored in the memory 1005. Alternatively, the processor 1001 may be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array (Programmable Logic Array, PLA). The processor 1001 may integrate one or a combination of several of a central processing unit (Central Processing Unit, CPU), an image processor (Graphics Processing Unit, GPU), and a modem, etc. The CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 1001 and may be implemented by a single chip.
The Memory 1005 may include a random access Memory (Random Access Memory, RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer readable medium (non-transitory computer-readable storage medium). The memory 1005 may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory 1005 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the above-described respective method embodiments, etc.; the storage data area may store data or the like referred to in the above respective method embodiments. The memory 1005 may also optionally be at least one storage system located remotely from the processor 1001. As shown in FIG. 5, an operating system, a network communication module, a user interface module, and a purchase screening application based on vendor portraits may be included in memory 1005, which is a type of computer storage medium.
In terminal 1000 shown in fig. 5, user interface 1003 is mainly used for providing an input interface for a user, and acquiring data input by the user; and processor 1001 may be configured to invoke the vendor portrayal-based procurement screening application stored in memory 1005 and specifically:
When receiving a to-be-processed purchasing demand, acquiring a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier in a supplier list; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
inputting the evaluation index parameters of the suppliers and the index score data of the evaluation index parameters into a pre-trained supplier rating model, and outputting the rating results of the evaluation indexes corresponding to each supplier;
according to the rating results of the evaluation indexes corresponding to each provider, making a provider purchase portrait of each provider;
and screening out target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers.
In one embodiment, the processor 1001, prior to receiving the pending procurement requirements, further performs the following:
acquiring a historical purchase order of each provider in a provider list;
Matching purchase data in the historical purchase orders for each preset evaluation index to obtain supplier evaluation index parameters corresponding to the historical purchase orders of suppliers; the supplier evaluation index parameters comprise quality index parameters, cost index parameters, delivery index parameters, service index parameters, technical index parameters, asset index parameters and employee flow index parameters;
establishing an order comment template corresponding to the historical purchase order of each provider according to the provider evaluation index parameter, and pushing the order comment template to a client for evaluation;
collecting comment data sets commented on historical purchase orders of each supplier in a preset time period through a big data technology, calculating index score data of each evaluation index according to the comment data sets, and generating index score data corresponding to the historical purchase orders of each supplier;
and establishing an association relationship between the historical purchase orders of each provider and the corresponding provider evaluation index parameters and index score data.
In one embodiment, the processor 1001, when executing the formulation of the purchase portrayal of the supplier of each supplier according to the rating result of the respective rating index corresponding to each supplier, specifically performs the following operations:
Acquiring a plurality of preset purchasing strategies, wherein the purchasing strategies at least comprise a first purchasing strategy with good order performance, a second purchasing strategy with short delivery period and high quality and a third purchasing strategy with long delivery period and low cost;
determining a plurality of first indexes required by a first purchasing strategy, a plurality of second indexes required by a second purchasing strategy and a plurality of third indexes required by a third purchasing strategy in the evaluation index parameters of the suppliers;
matching the rating results corresponding to each first index with the rating results corresponding to each second index and matching the rating results corresponding to each third index in the rating results of each evaluation index corresponding to each supplier;
according to the rating result corresponding to each first index, calculating first similarity between each provider and the first purchasing strategy, determining first providers with the first similarity larger than a preset threshold, and taking the first purchasing strategy as a provider purchasing portrait of each first provider;
calculating a second similarity between each supplier and a second purchasing strategy according to the rating result corresponding to each second index, determining a second supplier with the second similarity larger than a preset threshold value, and taking the second purchasing strategy as a supplier purchasing portrait of each second supplier;
And calculating third similarity between each supplier and the third purchasing strategy according to the rating result corresponding to each third index, determining the third supplier with the third similarity larger than a preset threshold, and taking the third purchasing strategy as a supplier purchasing portrait of each third supplier.
In one embodiment, the processor 1001, when executing the screening of the target suppliers meeting the purchase requirement to be processed based on the purchase portrayal of each supplier, specifically performs the following operations:
analyzing a target purchasing strategy of purchasing requirements to be processed;
when the target purchasing strategy is a first purchasing strategy with good order performance, calculating an intersection between a purchasing portrait of a supplier of each supplier and the first purchasing strategy, and determining suppliers with non-empty intersections; or,
when the target purchasing strategy is a second purchasing strategy with short period and high quality, calculating an intersection between the purchasing portraits of the suppliers and the second purchasing strategy, and determining suppliers with non-empty intersection; or,
when the target purchasing strategy is a third purchasing strategy with long period and low cost, calculating an intersection between the purchasing portraits of the suppliers and the third purchasing strategy, and determining suppliers with non-empty intersection;
When the provider with the intersection not being empty is one, determining the provider with the intersection not being empty as a target provider of the purchasing demand to be processed; or when the number of suppliers with non-empty intersections is a plurality of suppliers, calculating the purchasing cost of each supplier according to the purchasing demand to be processed, and taking the supplier with the lowest purchasing cost as the target supplier of the purchasing demand to be processed.
In one embodiment, the processor 1001, in generating the pre-trained vendor rating model, specifically performs the following:
a plurality of historical evaluation index parameters of the historical real orders of each provider are formulated;
establishing a scoring true value of each historical evaluation index parameter;
marking a preset rating label on the scoring true value of each historical evaluation index parameter to obtain each historical evaluation index parameter of each provider and rating label mapping data thereof;
establishing a parameter matrix of each provider according to each historical evaluation index parameter and the rating label mapping data thereof;
establishing a provider rating model by adopting a neural network, inputting a parameter matrix of each provider into the provider rating model, and outputting a model loss value;
when the model loss value reaches the minimum, generating a pre-trained provider rating model; alternatively, when the model loss value does not reach the minimum, the model loss value is back-propagated to update the model parameters of the provider rating model, and the step of inputting the parameter matrix of each provider into the provider rating model is continued until the model loss value reaches the minimum.
In one embodiment, the processor 1001, when executing the establishment of the parameter matrix for each vendor according to each historical evaluation index parameter and its rating label mapping data, specifically performs the following operations:
inputting each historical evaluation index parameter and rating label mapping data into an encoder, and quantizing each historical evaluation index parameter and rating label mapping data into 11 when the real part and the imaginary part of the encoder at the current time are positive numbers, or quantizing each historical evaluation index parameter and rating label mapping data into 10 when the real part of the encoder at the current time is positive numbers and the imaginary part of the encoder is negative numbers, or quantizing each historical evaluation index parameter and rating label mapping data into 01 when the real part and the imaginary part of the encoder at the current time are negative numbers, or quantizing each historical evaluation index parameter and rating label mapping data into 00 when the real part and the imaginary part of the encoder at the current time are both negative numbers, and outputting a first quantization result of each historical evaluation index parameter and a second quantization result of the rating label thereof after quantization is finished;
carrying out binary conversion on the first quantization result of each historical evaluation index parameter to obtain a plurality of unsigned first integer values;
Carrying out binary conversion on the second quantization result of the rating label of each historical evaluation index parameter to obtain a plurality of unsigned second integer values;
and combining the plurality of unsigned first integer values and the plurality of unsigned second integer values into a feature matrix to obtain a parameter matrix of each provider.
In the embodiment of the application, when receiving a to-be-processed purchasing demand, a purchasing screening system based on a supplier portrait acquires a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier, inputs the supplier evaluation index parameter and the index score data thereof into a pre-trained supplier rating model, outputs a rating result of each evaluation index, formulates a supplier purchasing portrait of each supplier according to the rating result of each evaluation index, and finally screens a target supplier meeting the to-be-processed purchasing demand based on the supplier purchasing portrait of each supplier. According to the method and the system, the rating results of the evaluation indexes are automatically determined through the pre-training model, and the purchasing image of the supplier can be accurately established based on the rating results of the evaluation indexes, and the order delivery guarantee and the purchasing plan with reduced order cost can be clearly formulated through the image, so that the purchasing accuracy of a host factory is improved.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by a computer program for instructing associated hardware, and that a program for purchasing screening based on vendor representations may be stored in a computer readable storage medium, which program, when executed, may comprise the steps of the embodiments of the methods described above. The storage medium of the program based on purchase screening of the vendor representation may be a magnetic disk, an optical disk, a read-only memory, a random access memory, or the like.
The foregoing disclosure is only illustrative of the preferred embodiments of the present application and is not intended to limit the scope of the claims herein, as the equivalent of the claims herein shall be construed to fall within the scope of the claims herein.
Claims (10)
1. A method for purchasing screening based on vendor representation, the method comprising:
when receiving a to-be-processed purchasing demand, acquiring a supplier evaluation index parameter and index score data thereof associated with a historical purchasing order of each supplier in a supplier list; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
Inputting the supplier evaluation index parameters and the index score data into a pre-trained supplier rating model, and outputting rating results of each evaluation index corresponding to each supplier;
according to the rating results of the evaluation indexes corresponding to each provider, making a provider purchase portrait of each provider;
and screening out target suppliers meeting the purchasing requirements to be processed based on the purchasing portraits of the suppliers.
2. The method of claim 1, further comprising, prior to receiving the pending purchase demand:
acquiring a historical purchase order of each provider in a provider list;
matching purchase data in the historical purchase orders for each preset evaluation index to obtain supplier evaluation index parameters corresponding to the historical purchase orders of each supplier; the supplier evaluation index parameters comprise quality index parameters, cost index parameters, delivery index parameters, service index parameters, technical index parameters, asset index parameters and employee flow index parameters;
establishing an order comment template corresponding to the historical purchase order of each provider according to the provider evaluation index parameter, and pushing the order comment template to a client for evaluation;
Collecting comment data sets commented on historical purchase orders of each supplier in a preset time period through a big data technology, calculating index score data of each evaluation index according to the comment data sets, and generating index score data corresponding to the historical purchase orders of each supplier;
and establishing an association relationship between the historical purchase orders of each provider and the corresponding provider evaluation index parameters and index score data.
3. The method of claim 1, wherein the formulating the supplier purchase portrayal for each supplier based on the rating results of the respective rating indicators for each supplier comprises:
acquiring a plurality of preset purchasing strategies, wherein the purchasing strategies at least comprise a first purchasing strategy with good order performance, a second purchasing strategy with short delivery period and high quality and a third purchasing strategy with long delivery period and low cost;
determining a plurality of first indexes required by the first purchasing strategy, a plurality of second indexes required by the second purchasing strategy and a plurality of third indexes required by the third purchasing strategy in the supplier evaluation index parameters;
Matching the rating results corresponding to each first index with the rating results corresponding to each second index and matching the rating results corresponding to each third index in the rating results of each evaluation index corresponding to each supplier;
calculating first similarity between each supplier and the first purchasing strategy according to the rating result corresponding to each first index, determining first suppliers with the first similarity larger than a preset threshold, and taking the first purchasing strategy as a supplier purchasing portrait of each first supplier;
calculating a second similarity between each supplier and the second purchasing strategy according to the rating result corresponding to each second index, determining a second supplier with the second similarity larger than a preset threshold value, and taking the second purchasing strategy as a supplier purchasing portrait of each second supplier;
and calculating third similarity between each supplier and the third purchasing strategy according to the rating result corresponding to each third index, determining the third suppliers with the third similarity larger than a preset threshold, and taking the third purchasing strategy as a supplier purchasing portrait of each third supplier.
4. The method of claim 3, wherein screening out target suppliers meeting the pending purchase demand based on the supplier purchase portraits of each supplier comprises:
analyzing a target purchasing strategy of the purchasing demand to be processed;
when the target purchasing strategy is a first purchasing strategy with good order performance, calculating an intersection between a purchasing portrait of a supplier of each supplier and the first purchasing strategy, and determining suppliers with non-empty intersections; or,
when the target purchasing strategy is a second purchasing strategy with short period and high quality, calculating an intersection between the purchasing portraits of the suppliers and the second purchasing strategy, and determining suppliers with non-empty intersection; or,
when the target purchasing strategy is a third purchasing strategy with long period and low cost, calculating an intersection between the purchasing portraits of the suppliers and the third purchasing strategy, and determining suppliers with non-empty intersection;
when the provider with the intersection not being empty is one, determining the provider with the intersection not being empty as the target provider of the purchasing demand to be processed; or when the number of suppliers with non-empty intersections is a plurality of, calculating the purchasing cost of each supplier according to the purchasing demand to be processed, and taking the supplier with the lowest purchasing cost as the target supplier of the purchasing demand to be processed.
5. The method of claim 1, wherein generating the pre-trained vendor rating model comprises:
a plurality of historical evaluation index parameters of the historical real orders of each provider are formulated;
establishing a scoring true value of each historical evaluation index parameter;
marking a preset rating label on the scoring true value of each historical evaluation index parameter to obtain each historical evaluation index parameter of each provider and rating label mapping data thereof;
establishing a parameter matrix of each provider according to each historical evaluation index parameter and the rating label mapping data thereof;
establishing a provider rating model by adopting a neural network, inputting a parameter matrix of each provider into the provider rating model, and outputting a model loss value;
when the model loss value reaches the minimum, generating a pre-trained supplier rating model; or when the model loss value does not reach the minimum, back-propagating the model loss value to update the model parameters of the vendor rating model, and continuing to perform the step of inputting the parameter matrix of each vendor into the vendor rating model until the model loss value reaches the minimum.
6. The method of claim 5, wherein the establishing a parameter matrix for each provider based on each historical evaluation index parameter and its rating label mapping data comprises:
inputting each historical evaluation index parameter and rating label mapping data into an encoder, and when the real part and the imaginary part of the encoder at the current time are positive numbers, quantizing each historical evaluation index parameter and rating label mapping data into 11, or when the real part of the encoder at the current time is positive numbers and the imaginary part is negative numbers, quantizing each historical evaluation index parameter and rating label mapping data into 10, or when the real part of the encoder at the current time is negative numbers and the imaginary part of the encoder is positive numbers, quantizing each historical evaluation index parameter and rating label mapping data into 01, or when the real part and the imaginary part of the encoder at the current time are negative numbers, quantizing each historical evaluation index parameter and rating label mapping data into 00, and outputting a first quantization result of each historical evaluation index parameter and a second quantization result of the rating label thereof after quantization;
carrying out binary conversion on the first quantization result of each historical evaluation index parameter to obtain a plurality of unsigned first integer values;
Carrying out binary conversion on the second quantization result of the rating label of each historical evaluation index parameter to obtain a plurality of unsigned second integer values;
and forming a feature matrix by the plurality of unsigned first integer values and the plurality of unsigned second integer values to obtain a parameter matrix of each provider.
7. The method of claim 5, wherein the model loss function expression of the vendor rating model is:
wherein lambda is the weight of the similarity matrix loss function, p and q are the vector representations output after the parameter matrix of each provider enters the normalization layer, and i is each element in the parameter matrix;
the normalization layer is defined as:wherein,
vector Z is the vector of the output of the full link layer, q i Representing normalized probability of ith element in parameter matrix, Z i Ith dimension, Z, representing full link layer output vector j The j-th dimension representing the full connection layer output vector, T is a parameter controlling the output probability smoothness.
8. A purchase screening system based on a vendor representation, the system comprising:
the data acquisition module is used for acquiring the supplier evaluation index parameters and index score data thereof associated with the historical purchase orders of each supplier in the supplier list when receiving the purchase demand to be processed; the provider evaluation index parameters are purchase data matched with each preset evaluation index in the historical purchase order, the index score data are calculated for each preset evaluation index according to a comment data set, and the comment data set is collected for the historical purchase order in a period of time through a big data technology;
The rating result output module is used for inputting the supplier rating index parameters and the index score data thereof into a pre-trained supplier rating model and outputting the rating result of each rating index corresponding to each supplier;
the supplier purchase portrait making module is used for making a supplier purchase portrait of each supplier according to the rating results of the evaluation indexes corresponding to each supplier;
and the supplier screening module is used for screening out target suppliers meeting the purchasing requirements to be processed based on the supplier purchasing portraits of each supplier.
9. A computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the method of any of claims 1-7.
10. A terminal, comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310973180.0A CN117522256A (en) | 2023-08-03 | 2023-08-03 | Purchasing screening method, system, storage medium and terminal based on supplier portrait |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310973180.0A CN117522256A (en) | 2023-08-03 | 2023-08-03 | Purchasing screening method, system, storage medium and terminal based on supplier portrait |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117522256A true CN117522256A (en) | 2024-02-06 |
Family
ID=89744529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310973180.0A Pending CN117522256A (en) | 2023-08-03 | 2023-08-03 | Purchasing screening method, system, storage medium and terminal based on supplier portrait |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117522256A (en) |
-
2023
- 2023-08-03 CN CN202310973180.0A patent/CN117522256A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020143344A1 (en) | Method and device for warehouse receipt pledge financing based on blockchain architecture | |
| CN111861569B (en) | Product information recommendation method and device | |
| WO2020143341A1 (en) | Blockchain architecture-based warehouse receipt pledge financing assessment method and device | |
| WO2019037202A1 (en) | Method and apparatus for recognising target customer, electronic device and medium | |
| CN107330715B (en) | Method and device for selecting picture advertisement material | |
| CN106296669A (en) | A kind of image quality evaluating method and device | |
| CN113159835B (en) | Power generation side electricity price quotation method and device based on artificial intelligence, storage medium and electronic equipment | |
| CN108805689A (en) | A kind of loan risk evaluation control method and device | |
| CN109409739B (en) | Crowdsourcing platform task allocation method based on POMDP model | |
| CN106408325A (en) | User consumption behavior prediction analysis method based on user payment information and system | |
| CN110322093A (en) | Information processing method, information display method, device and calculating equipment | |
| CN114202018A (en) | Modular joint learning method and system | |
| CN117422553A (en) | Transaction processing method, device, equipment, medium and product of blockchain network | |
| CN112860672A (en) | Method and device for determining label weight | |
| CN111353728A (en) | Risk analysis method and system | |
| Liu et al. | A multi-agent-based mould due date setting approach in stochastic production | |
| CN116503158A (en) | Enterprise bankruptcy risk early warning method, system and device based on data driving | |
| CN117522256A (en) | Purchasing screening method, system, storage medium and terminal based on supplier portrait | |
| CN115545088A (en) | Model construction method, classification method and device and electronic equipment | |
| CN108197158A (en) | A kind of information acquisition method and its equipment | |
| CN115185606A (en) | Method, device, equipment and storage medium for obtaining service configuration parameters | |
| CN111080393A (en) | Transaction matching method and device | |
| US20100205030A1 (en) | Systems and methods for factor-based pricing for products and services | |
| CN116542831B (en) | Method and device for processing recruitment data, electronic equipment and storage medium | |
| CN117422191A (en) | Mathematical model solving method and device, computing equipment and computing equipment cluster |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |