Method and device for processing service data

Abstract

The invention discloses a business data processing method and device, and relates to the technical field of intelligent logistics. A specific embodiment of the method comprises the following steps: screening out a plurality of service providers matched with demand data according to the demand data published by a service demand side; based on the extracted demand features of the business demander and the service features of the business provider; predicting an intention probability that a service provider provides a service for the service demander; determining candidate service providers from the plurality of service providers for the service demander in combination with the demand constraint of the service demander and the intention probability of the service providers; according to the embodiment of the invention, the accuracy of bilateral matching of the service demander and the service provider is improved, and the efficiency of determining the service provider by the service demander is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart logistics technology, and in particular to a method and device for processing business data. Background Art

[0002] In various business processing scenarios, multi-party cooperation is usually adopted to complete the business; in the cooperation process, the business demander usually publishes the demand information (such as publishing the subject of the tender), and the business provider determines whether to provide the service (such as participating in the bidding) based on the demand raised by the business demander.

[0003] Currently, the demand information of business demanders and the service information provided by service providers are usually obtained through business personnel. The existing methods have the problems of poor real-time information acquisition, limited amount of information obtained and low efficiency, which makes it more difficult for business demanders to determine matching service providers and may increase their business costs. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and device for processing business data, which can screen out multiple business providers that match the demand data based on the demand data released by the business demand party; and based on the extracted demand characteristics of the business demand party and the service characteristics of the business provider; predict the probability of the business provider's intention to provide services to the business demand party; combine the demand constraints of the business demand party and the intention probability of the business provider to determine candidate business providers from multiple business providers for the business demand party; the embodiment of the present invention improves the accuracy of bilateral matching between business demand parties and business providers, and improves the efficiency of business demand parties in determining business providers.

[0005] To achieve the above-mentioned purpose, according to one aspect of an embodiment of the present invention, a method for processing business data is provided, which is characterized in that it includes: based on the demand data published by the business demander, screening out multiple business providers that match the demand data from the business provider data source; and obtaining the service data of the business provider; extracting demand characteristics from the demand data of the business demander; extracting service characteristics from the service data of multiple business providers; based on the demand characteristics and service characteristics, predicting the intention probability of each business provider to provide services to the business demander; combining the demand constraints of the business demander and the intention probability of the business provider, determining candidate business providers from multiple business providers; and sending the demand data of the business demander to the candidate business providers.

[0006] Optionally, extracting demand features from demand data corresponding to the business demand side includes: parsing demand keywords indicating different demand types from the demand data using a text processing model; and extracting feature data corresponding to each demand keyword as the demand feature of the business demand side.

[0007] Optionally, service characteristics are extracted from the service data corresponding to each business provider, including: for each business provider, performing the following operations: obtaining historical service data of services provided by the business provider, and historical demand data of the business demander associated with the historical service data from the business provider data source; extracting historical demand characteristics from the historical demand data, and extracting historical service characteristics from the historical service data; determining the cost data of the business provider and the risk probability of not providing the service based on the historical demand characteristics and the historical service characteristics; and using the cost data and risk probability of the business provider as service characteristics.

[0008] Optionally, based on the demand characteristics and service characteristics, predict the probability of the service provider's intention to provide services to the service demander; including: using multiple graph convolution layers contained in a preset prediction model, representing the demand characteristics and service characteristics of the input graph convolution layer as active demand characteristics, passive demand characteristics, active service characteristics and passive service characteristics; and using multiple graph convolution layers to perform feature interaction convolution processing on the active demand characteristics, passive demand characteristics, active service characteristics and passive service characteristics, and output active demand nodes, passive demand nodes, active service nodes and passive service nodes; inputting the active demand nodes, passive demand nodes, active service nodes and passive service nodes into the prediction layer contained in the preset prediction model, and using the prediction layer to predict the probability of the service provider's intention to provide services to the service demander.

[0009] Optionally, the prediction layer is used to predict the probability of the service provider's intention to provide services to the service demander, including: using the prediction layer in combination with active demand nodes and passive service nodes to calculate a first probability that the service demander actively queries the service provider; combining the passive demand nodes and active service nodes to calculate a second probability that the service provider actively provides services to the service provider; and using a weighted result obtained by calculating the weighted values ​​of the first probability and the second probability as the probability of the service provider's intention to provide services to the service demander.

[0010] Optionally, candidate service providers are determined in combination with the demand constraints of the service demander and the intention probability of the service provider, including: constructing a matching objective function for the service demander to determine the candidate service providers; the matching objective function is composed of a first data item constructed according to the demand constraints of the service demander and the intention probability of the service provider, a second data item representing the fairness of the service provider, and the demand constraints; performing multiple iterative calculations on the matching objective function, and determining one or more candidate service providers according to the optimal solution of the calculation results.

[0011] Optionally, multiple iterative calculations are performed on the matching objective function, and one or more candidate service providers are determined based on the calculated optimal solution, including: for each iterative calculation performed on the matching objective function, scoring is performed based on whether the matching objective function and constraint conditions are met; the optimal solution of the matching objective function is determined based on the scoring results, and one or more candidate service providers are determined based on the optimal solution.

[0012] To achieve the above-mentioned purpose, according to a second aspect of an embodiment of the present invention, there is provided a device for processing business data, characterized in that it comprises: a data acquisition module, an intention prediction module and a target determination module; wherein,

[0013] The data acquisition module is used to select multiple business providers that match the demand data from the business provider data source according to the demand data released by the business demander; and obtain the service data of the business provider;

[0014] The prediction intention module is used to extract demand features from the demand data of the business demander; extract service features from the service data of multiple business providers; based on the demand features and service features, predict the probability of each business provider's intention to provide services to the business demander;

[0015] The target determination module is used to determine a candidate business provider from multiple business providers based on the demand constraints of the business demander and the intention probability of the business provider; and send the demand data of the business demander to the candidate business provider.

[0016] Optionally, the device for processing business data is used to extract demand characteristics from the demand data corresponding to the business demand party, including: using a text processing model to parse demand keywords indicating different demand types from the demand data; extracting feature data corresponding to each demand keyword as the demand characteristics of the business demand party.

[0017] Optionally, the device for processing business data is used to extract service characteristics from the service data corresponding to each business provider, including: for each business provider, performing the following operations: obtaining historical service data of services provided by the business provider, and historical demand data of the business demander associated with the historical service data from the business provider data source; extracting historical demand characteristics from the historical demand data, and extracting historical service characteristics from the historical service data; based on the historical demand characteristics and historical service characteristics, determining the cost data of the business provider and the risk probability of not providing the service; and using the cost data and risk probability of the business provider as service characteristics.

[0018] Optionally, a device for processing business data is used to predict the probability of a business provider's intention to provide services to a business demander based on demand characteristics and service characteristics; including: using multiple graph convolution layers contained in a preset prediction model to represent the demand characteristics and service characteristics of the input graph convolution layer as active demand characteristics, passive demand characteristics, active service characteristics and passive service characteristics; and using multiple graph convolution layers to perform feature interaction convolution processing on the active demand characteristics, passive demand characteristics, active service characteristics and passive service characteristics, and output active demand nodes, passive demand nodes, active service nodes and passive service nodes; inputting the active demand nodes, passive demand nodes, active service nodes and passive service nodes into the prediction layer contained in the preset prediction model, and using the prediction layer to predict the probability of the business provider's intention to provide services to the business demander.

[0019] Optionally, the device for processing business data is used to use the prediction layer to predict the probability of the service provider's intention to provide services to the business demander, including: using the prediction layer in combination with the active demand node and the passive service node to calculate a first probability that the business demander actively queries the service provider; combining the passive demand node and the active service node to calculate a second probability that the service provider actively provides services to the service provider; and using the weighted result obtained by calculating the weighted values ​​of the first probability and the second probability as the probability of the service provider's intention to provide services to the business demander.

[0020] Optionally, the device for processing business data is used to determine candidate business providers in combination with the demand constraints of the business demander and the intention probability of the business provider, including: constructing a matching objective function for the business demander to determine the candidate business providers; the matching objective function is composed of a first data item constructed according to the demand constraints of the business demander and the intention probability of the service provider, a second data item representing the fairness of the service provider, and the demand constraints; performing multiple iterative calculations on the matching objective function, and determining one or more candidate business providers based on the optimal solution of the calculation results.

[0021] Optionally, the device for processing business data is used to perform multiple iterative calculations for the matching objective function, and determine one or more candidate service providers based on the calculated optimal solution, including: for each iterative calculation performed on the matching objective function, scoring whether the matching objective function and constraints are met; determining the optimal solution of the matching objective function based on the scoring results, and determining one or more candidate service providers based on the optimal solution.

[0022] To achieve the above-mentioned purpose, according to the third aspect of an embodiment of the present invention, there is provided an electronic device for processing business data, characterized in that it includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described in the above-mentioned methods for processing business data.

[0023] To achieve the above-mentioned purpose, according to a fourth aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, characterized in that when the program is executed by a processor, any of the above-mentioned methods for processing business data is implemented.

[0024] To achieve the above-mentioned purpose, according to the fifth aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, characterized in that when the computer program is executed by a processor, it implements any of the methods for processing business data as described above.

[0025] An embodiment of the above invention has the following advantages or beneficial effects: it can screen out multiple business providers that match the demand data published by the business demander; and based on the extracted demand characteristics of the business demander and the service characteristics of the business provider; predict the probability of the business provider's intention to provide services to the business demander; combine the business demander's demand constraints and the business provider's intention probability to determine candidate business providers from multiple business providers for the business demander; the embodiment of the present invention improves the accuracy of bilateral matching between business demanders and business providers, and improves the efficiency of business demanders in determining business providers.

[0026] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0028] Figure 1 It is a flowchart of a method for processing business data provided by an embodiment of the present invention;

[0029] FIG2 is a schematic diagram of a process for processing business data provided by an embodiment of the present invention;

[0030] Figure 3 is a schematic diagram of the structure of a preset prediction model provided by an embodiment of the present invention;

[0031] Figure 4 is a schematic diagram of the structure of a device for processing business data provided by an embodiment of the present invention;

[0032] FIG5 is a diagram of an exemplary system architecture in which an embodiment of the present invention may be applied;

[0033] Figure 6 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0035] It should be noted that in the technical solution of the present invention, the collection, use, storage, sharing and transfer of user personal information involved are in compliance with the provisions of relevant laws and regulations, and it is necessary to inform the user and obtain the user's consent or authorization. When applicable, the user's personal information is de-identified and / or anonymized and / or encrypted.

[0036] like Figure 1 As shown, an embodiment of the present invention provides a method for processing business data, which may include the following steps:

[0037] Step S101: According to the demand data published by the service demander, multiple service providers matching the demand data are screened out from the service provider data source; and service data of the service providers are obtained.

[0038] Specifically, in various businesses, there are usually situations where multiple parties work together to complete the business; for example, for infrastructure business, it can be completed by different types of infrastructure partners based on the needs of the business demander; for example, for transportation business in logistics, the business demander can complete it through bidding by multiple transportation carriers (ie, business providers) based on the specific circumstances of the transportation needs.

[0039] In an embodiment of the present invention, a transportation business scenario is taken as an example, for example, the business demander is a logistics company, and the business provider is a variety of partners that can provide transportation services; the demand data includes specific demand information of the business demander, such as: transportation route, transportation vehicle type, reference cargo volume, timeliness requirements, quotation range, etc.; the demand data can be published to the bidding platform through bids, bidding documents, etc.

[0040] Further, multiple business providers matching the demand data are screened out from the preset business provider data source, and the service data of the business providers are obtained; specifically, for transportation business, the service data of multiple business providers (transportation business providers) are stored in the business provider data source; the data in the business provider data source can be stored based on historical business cooperation data, or can be collected from multiple channels; further, for transportation business, the service data of the service provider includes, for example, qualification information, historical quotations, service details, etc. Among them, the strategy of screening multiple business providers matching the demand data from the preset business provider data source can be: based on the activity of the business provider, the area where it is located, etc.

[0041] Step S102: extracting demand features from demand data of a service demander; extracting service features from service data of multiple service providers.

[0042] Specifically, the embodiments of the present invention characterize the behavior characteristics of the service provider by extracting the demand characteristics of the service demander and the service characteristics of the service provider, thereby performing subsequent predictions based on the demand characteristics and the service characteristics.

[0043] 1) For demand characteristics: extract demand characteristics from the demand data corresponding to the business demand side, including: using a text processing model to parse demand keywords indicating different demand types from the demand data; extracting feature data corresponding to each demand keyword as the demand characteristics of the business demand side. Specifically, demand data (subject, bidding documents, etc.) are usually in text form. The embodiment of the present invention uses a text processing model to parse the demand keywords of the demand data, and extracts feature data corresponding to each demand keyword as the demand characteristics of the business demand side; for example, the demand keyword is a transportation model, etc.

[0044] In an embodiment of the present invention, before extracting demand features from demand data, the demand data (subject) released by the business demand party can be processed in real time by monitoring the publishing platform (such as a bidding platform). The specific processing process includes loading the subject file (demand data), parsing different keywords from the subject using a text processing model (such as a generative text processing model, a text segmentation processing model, etc.), and classifying the feature types of the keywords. It can be understood that different demand keywords can belong to the same demand type. Furthermore, the feature data corresponding to each demand keyword is extracted as the demand feature of the business demand party.

[0045] 2) For service characteristics: extract service characteristics from the service data corresponding to each business provider, including: for each business provider, perform the following operations: obtain the historical service data of the services provided by the business provider from the business provider data source, and the historical demand data of the business demander associated with the historical service data; extract historical demand characteristics from the historical demand data, and extract historical service characteristics from the historical service data; determine the cost data and risk probability of the business provider based on the historical demand characteristics and historical service characteristics; use the cost data and risk probability of the business provider as service characteristics.

[0046] Specifically, taking the transportation business as an example, historical service features are extracted from the historical service data, where historical demand features can be features such as routes, timeliness, reference cargo volume, required vehicle model, quotation range, and business type that represent transportation demand; historical service features can be features such as the quotation frequency, number of fleets, credit qualifications, advantageous routes, and quotation weights of the evaluation service provider.

[0047] For example, formula (1) is constructed based on historical demand characteristics

[0048] (1)

[0049] Among them, f(a i ) represents the demand data of a business demand side, θ represents different feature types, qtys represents the quantity of goods, type represents the vehicle type, line represents the route, etc.

[0050] Formula (2) is constructed based on historical service characteristics

[0051] (2)

[0052] Among them, g(β i ) represents the service data of a certain service provider, σ represents different feature types, cars represents available vehicles, habit represents behavioral features, line represents routes, etc.

[0053] Furthermore, based on the historical demand characteristics and historical service characteristics, the cost data of the service provider and the risk probability of not providing the service are determined; and the cost data and risk probability of the service provider are used as service characteristics.

[0054] For example, the cost data of service provider j for service demand i is calculated according to the following formula (3): (For example, the quoted amount, etc.).

[0055] (3)

[0056] The risk probability of service provider j not providing service is calculated according to the following formula (4): (For example, the probability of abandonment of bid, etc.).

[0057] (4)

[0058] Furthermore, the cost data and risk probability of the service provider are used as service characteristics of the service provider.

[0059] Step S103: Based on the demand characteristics and service characteristics, predict the probability of each service provider's intention to provide services to the service demander.

[0060] Specifically, the demand characteristics and service characteristics are input into a preset prediction model, and the preset prediction model is used to output the intention probability of the service provider to provide services to the service demander, wherein the intention probability is also the probability of winning the bid.

[0061] The specific description of the preset prediction model predicting the probability of each service provider's intention to provide services to the service demander is as follows: Figure 3 The description of the structure of the preset prediction model is consistent and will not be repeated here.

[0062] Step S104: Determine candidate service providers from multiple service providers in combination with the service demander's demand constraints and the service provider's intention probability; and send the service demander's demand data to the candidate service providers.

[0063] Specifically, according to the intention probability of the service provider determined in step S103, a candidate service provider is determined from multiple service providers in combination with the demand constraint of the service demander and the intention probability of the service provider.

[0064] Still taking the transportation business as an example, the matching module is used to perform the step of determining candidate business providers from multiple business providers. The matching module can match the target (business demand) with the data source of the business provider in a certain time window, adopt the idea of ​​joint estimation optimization model, and take the lowest expected quotation of all current targets (business demands) as the objective function, as shown in formula (5):

[0065] (5)

[0066] Among them, C represents the data source of the business provider, B represents the target pool (demand data set) within a time window; the decision variable Xij=0 or 1, which represents the business demander's intention to send information about target i to the business provider j (such as a contact email); the quotation intention probability Yij = P(Bj | Ci), which represents the probability of the business provider j's quotation intention for target i; the winning probability Zij = Pz (Bj | Ci), which represents the probability of the business provider j's winning bid for target i; the risk weight Wj, which represents the risk abandonment probability of the business provider j; the expected quotation q ij , represents the administrative cost of the quotation amount of service provider j for target i; administrative cost g ij , represents the fixed administrative costs incurred by service provider j when signing a contract for subject i;

[0067] Furthermore, the demand constraints of the business demand side are as follows:

[0068]

[0069] As shown in formula (5), the formula includes two data items, which are connected by a "+" sign, wherein the first data item is constructed according to the demand constraints of the business demand side and the intention probability of the business provider, and the second data item represents the fairness of the business provider; wherein gini represents the Gini coefficient, which represents the data of each business provider being actively contacted (for example, sending a contact email); the smaller the Gini coefficient, the more fair it is. By combining the objective function of the first data item and the second data item, the beneficial effect of balancing the lowest quotation cost and fair distribution is achieved. That is, combining the demand constraints of the business demand side and the intention probability of the business provider, the candidate business provider is determined, including: constructing a matching objective function for the business demand side to determine the candidate business provider; the matching objective function is composed of the first data item constructed according to the demand constraints of the business demand side and the intention probability of the business provider, the second data item representing the fairness of the business provider, and the demand constraints; performing multiple iterative calculations on the matching objective function, and determining one or more candidate business providers according to the optimal solution of the calculation results.

[0070] Further, multiple iterative calculations are performed for the matching objective function, and one or more candidate service providers are determined based on the optimal solution of the calculation results. Specifically, in solving the objective function shown in formula (5), the objective function can be used as a soft constraint (each time the soft constraint is broken, the corresponding score will be penalized), and the demand constraint condition can be used as a hard constraint (each time the hard constraint is broken, the solution is not a feasible solution). The heuristic algorithm is used for multiple iterations, and the solution with the highest score is calculated as the optimal solution. The optimal solution represents the candidate service provider determined to match the service demand side. That is, multiple iterative calculations are performed for the matching objective function, and one or more candidate service providers are determined based on the calculated optimal solution, including: for each iterative calculation performed for the matching objective function, scoring is performed based on whether the matching objective function and the constraint conditions are met; the optimal solution of the matching objective function is determined based on the scoring result, and one or more candidate service providers are determined based on the optimal solution.

[0071] Furthermore, the demand data of the business demander is sent to the candidate business providers. Specifically, after multiple candidate business providers are determined, the demand data (subject matter, etc.) can be sent to the candidate business providers, so that the candidate business providers can actively contact the business demander according to their own conditions, thereby determining one or more matching business providers for the business demander.

[0072] The embodiments of the present invention predict the probability of a service provider's intention to provide services to a service demander based on demand characteristics and service characteristics through a prediction model; determine candidate service providers from multiple service providers for the service demander in combination with the demand constraints of the service demander and the service provider's intention probability; the embodiments of the present invention improve the accuracy of bilateral matching between the service demander and the service provider, improve the efficiency of the service demander in determining the service provider, and can match the service demander with the most suitable service provider based on the demand data of the service demander.

[0073] Compared with the existing XGBoost benchmark model, the method for obtaining candidate business parties using a preset prediction model in the embodiment of the present invention has greatly improved recall and accuracy; for example: using the XGBoost benchmark model and the present invention to perform bilateral matching on the same offline data (demand data and service data), the recall rate of the top 100 business providers is: 66.0% for the XGBoost benchmark model, and 78.2% for the present invention; the accuracy rate for determining the top 5 business providers is: 22.6% for the XGBoost benchmark model, and 29.5% for the present invention; the improvement in recall and accuracy reduces the business costs and business cooperation risks of the business demand side, and improves the efficiency of the business demand side in determining the business provider.

[0074] like Figure 2As shown, an embodiment of the present invention provides a schematic diagram of a process for processing business data; the method may include the following steps:

[0075] Step S201: Based on the demand data published by the business demander, multiple business providers matching the demand data are screened out from the business provider data source; and service data of the business providers are obtained; demand features are extracted from the demand data of the business demander; and service features are extracted from the service data (service data set) of multiple business providers.

[0076] Step S202: Based on the demand characteristics and service characteristics, a preset prediction model is used to predict the probability of each service provider's intention to provide services to the service demander.

[0077] Step S203: Determine a candidate service provider from multiple service providers in combination with the service demander's demand constraints and the service provider's intention probability; and send the service demander's demand data to the candidate service provider.

[0078] Specifically, the description of step S201 to step S203 is consistent with the description of step S101 to step S103, and will not be repeated here.

[0079] like Figure 3 As shown, an embodiment of the present invention provides a schematic diagram of a process structure of a preset prediction model; Figure 3 As shown, the preset prediction model includes a graph convolution layer 302 and a prediction layer 303, wherein the feature input component 301 is used to process the features required by the preset prediction model.

[0080] Specifically, taking the transportation business as an example, the demand feature is a vector (denoted as b) formed by connecting features such as routing and vehicle type, and the service feature is a vector (denoted as c) formed by connecting features such as the vehicle, historical registration records, historical bid abandonment behavior, qualifications, and credit records corresponding to the service provider. The feature input component 301 is used to input the demand features and service features into the graph convolution layer 302; the graph convolution layer 302 is used to represent the demand features and service features as active demand features, passive demand features, active service features, and passive service features, respectively, and further perform different operations according to different interaction methods to finally iterate out different vectors; the processed active demand features, passive demand features, active service features, and passive service features are respectively denoted as b a , b p , c a , c p ,In the graph convolution layer, according to the interactive quotation situation of the ,service demander and service provider, the active and passive features of the ,demand features have different connections with the active and passive features of the ,service provider, that is, the convolution processing of feature interaction is ,performed.

[0081] Specifically, the steps of performing convolution feature processing using the graph convolution layer are as follows, wherein any node representing a service demander or a service provider is denoted as d; further, the process of iteratively encoding each node using the graph convolution layer is as follows:

[0082] 1. Initialize the node, where f represents the demand feature or service feature, which is b a , b p , c a , c p Any of:

[0083]

[0084] 2. According to the following propagation mechanism, the propagation is performed layer by layer. The value of each layer of nodes is the weighted sum of the adjacent nodes in the previous layer. N d is the set of adjacent nodes of node d:

[0085]

[0086] 3. The final node representation is the mean of the nodes in each layer:

[0087]

[0088] Here n represents any one of an active demand node, a passive demand node, an active service node, and a passive service node. The active demand node and the passive demand node are further represented as , , the active service nodes and passive service nodes are represented as , .

[0089] Furthermore, the prediction layer 303 is used to predict the probability of the service provider's intention to provide services to the service demander.

[0090] That is, based on the demand characteristics and service characteristics, predict the intention probability of the service provider to provide services to the business demander; including: using multiple graph convolution layers contained in a preset prediction model, representing the demand characteristics and service characteristics of the input graph convolution layer as active demand characteristics, passive demand characteristics, active service characteristics and passive service characteristics; and using multiple graph convolution layers to perform feature interaction convolution processing on the active demand characteristics, passive demand characteristics, active service characteristics and passive service characteristics, and output active demand nodes, passive demand nodes, active service nodes and passive service nodes; inputting the active demand nodes, passive demand nodes, active service nodes and passive service nodes into the prediction layer contained in the preset prediction model, and using the prediction layer to predict the intention probability of the service provider to provide services to the business demander.

[0091] Furthermore, the prediction layer 303 comprehensively considers the probability r of the service demander actively contacting the service provider (i.e., the first probability of the service demander actively querying the service provider is calculated by combining the active demand node and the passive service node) and the probability s of the service provider actively bidding (i.e., the second probability of the service provider actively providing services to the service provider is calculated by combining the passive demand node and the active service node), and further linearly weights r and s to obtain the final intention probability y (i.e., the weighted result obtained by calculating the weighted values ​​of the first probability and the second probability is used as the intention probability of the service provider providing services to the service demander). The specific formulas for calculating the first probability r, the second probability s and the intention probability y are as follows:

[0092]

[0093]

[0094]

[0095] What is obtained through the prediction layer 303 is the intention probability (for example, the quotation intention probability) of the service provider under each characteristic to the service demander under each characteristic. Similarly, the winning probability z of the service provider can be determined. That is, using the prediction layer to predict the intention probability of the service provider to provide services to the service demander includes: using the prediction layer to combine the active demand node and the passive service node to calculate the first probability that the service demander actively queries the service provider; combining the passive demand node and the active service node to calculate the second probability that the service provider actively provides services to the service provider; and using the weighted result obtained by calculating the weighted value of the first probability and the second probability as the intention probability of the service provider to provide services to the service demander.

[0096] Further, in combination with the demand constraints of the service demander and the intention probability of the service provider, a candidate service provider is determined from multiple service providers; and the demand data of the service demander is sent to the candidate service provider. The description of determining the candidate service provider based on the objective function is consistent with the description of step S104 and will not be repeated here.

[0097] like Figure 4 As shown, an embodiment of the present invention provides a device 400 for processing business data, including: a data acquisition module 401, a prediction intention module 402 and a target determination module 403; wherein,

[0098] The data acquisition module 401 is used to filter out multiple business providers that match the demand data from the business provider data source according to the demand data published by the business demander; and to acquire the service data of the business provider;

[0099] Prediction intention module 402, used to extract demand features from demand data of the business demander; extract service features from service data of multiple business providers; based on demand features and service features, predict the probability of each business provider's intention to provide services to the business demander;

[0100] The target determination module 403 is used to determine a candidate service provider from multiple service providers in combination with the service demander's demand constraints and the service provider's intention probability; and to send the service demander's demand data to the candidate service provider.

[0101] An embodiment of the present invention also provides an electronic device for processing business data, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in any of the above embodiments.

[0102] An embodiment of the present invention further provides a computer-readable medium on which a computer program is stored. When the program is executed by a processor, the method provided in any of the above embodiments is implemented.

[0103] An embodiment of the present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, any one of the above methods for processing business data is implemented.

[0104] Figure 5 An exemplary system architecture 500 is shown to which the method for processing business data or the apparatus for processing business data according to the embodiments of the present invention can be applied.

[0105] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0106] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various client applications can be installed on terminal devices 501, 502, 503, such as e-commerce client applications, web browser applications, search applications, instant messaging tools, and email clients.

[0107] The terminal devices 501 , 502 , and 503 may be various electronic devices having a display screen and supporting various client applications, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0108] The server 505 may be a server that provides various services, such as a background management server that provides support for client applications used by users using the terminal devices 501, 502, and 503. The background management server may process the received request to determine a matching service provider for the service demander, and feed back the information of the determined service provider to the terminal device.

[0109] It should be noted that the method for processing business data provided in the embodiment of the present invention is generally executed by the server 505 , and accordingly, the device for processing business data is generally set in the server 505 .

[0110] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0111] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present invention. Figure 6 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0112] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0113] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0114] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present invention are executed.

[0115] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0116] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0117] The modules and / or units involved in the embodiments of the present invention may be implemented in software or hardware. The modules and / or units described may also be provided in a processor. For example, they may be described as follows: a processor includes a data acquisition module, an intention prediction module, and a target determination module. The names of these modules do not, in some cases, constitute limitations on the modules themselves. For example, the data acquisition module may also be described as a "module for filtering out service data of multiple business providers that match the demand data."

[0118] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: based on the demand data published by the business demander, multiple business providers matching the demand data are screened out from the business provider data source; and the service data of the business provider is obtained; demand features are extracted from the demand data of the business demander; service features are extracted from the service data of multiple business providers; based on the demand features and service features, the intention probability of each business provider to provide services to the business demander is predicted; in combination with the demand constraints of the business demander and the intention probability of the business provider, candidate business providers are determined from multiple business providers; and the demand data of the business demander is sent to the candidate business providers.

[0119] The embodiments of the present invention can screen out multiple business providers that match the demand data based on the demand data released by the business demander; and based on the extracted demand characteristics of the business demander and the service characteristics of the business provider; predict the probability of the business provider's intention to provide services to the business demander; and determine candidate business providers from multiple business providers for the business demander in combination with the demand constraints of the business demander and the intention probability of the business provider; the embodiments of the present invention improve the accuracy of bilateral matching between the business demander and the business provider, and improve the efficiency of the business demander in determining the business provider.

[0120] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for processing business data, characterized in that: include: According to the demand data published by the business demander, multiple business providers matching the demand data are screened out from the business provider data source; and obtaining service data of the service provider; Extracting demand features from the demand data of the service demander; extracting service features from the service data of the plurality of service providers; Based on the demand characteristics and the service characteristics, predict the intention probability of each of the service providers to provide services to the service demander; Determine a candidate service provider from the plurality of service providers in combination with the demand constraint of the service demander and the intention probability of the service provider; The demand data of the service demander is sent to the candidate service provider.

2. The method according to claim 1, characterized in that The extracting demand features from the demand data corresponding to the business demander includes: parsing demand keywords indicating different demand types from the demand data using a text processing model; The characteristic data corresponding to each of the demand keywords is extracted as the demand characteristics of the business demander.

3. The method according to claim 1, characterized in that The extracting service features from the service data corresponding to each of the service providers includes: For each of the service providers, perform the following operations: Acquire, from the data source of the service provider, historical service data of services provided by the service provider and historical demand data of the service demander associated with the historical service data; Extracting historical demand features from the historical demand data, and extracting historical service features from the historical service data; Determining cost data of the service provider and a risk probability of not providing service based on the historical demand characteristics and the historical service characteristics; The cost data and the risk probability of the service provider are used as the service characteristics.

4. The method according to claim 1, characterized in that The predicting, based on the demand characteristics and the service characteristics, the probability of the service provider's intention to provide services to the service demander comprises: Using a plurality of graph convolution layers included in a preset prediction model, the demand features and the service features input into the graph convolution layers are represented as active demand features, passive demand features, active service features, and passive service features; and using the plurality of graph convolution layers, convolution processing of feature interactions is performed on the active demand features, passive demand features, active service features, and passive service features, and active demand nodes, passive demand nodes, active service nodes, and passive service nodes are output; The active demand node, the passive demand node, the active service node and the passive service node are input into the prediction layer included in the preset prediction model, and the prediction layer is used to predict the probability of the service provider's intention to provide services to the service demander.

5. The method according to claim 4, characterized in that The using the prediction layer to predict the probability of the service provider's intention to provide services to the service demander includes: Calculate a first probability of the service demander actively querying the service provider by using the prediction layer in combination with the active demand node and the passive service node; Calculate a second probability that the service provider actively provides service to the service provider by combining the passive demand node and the active service node; A weighted result obtained by calculating the weighted values ​​of the first probability and the second probability is used as the intention probability of the service provider to provide service to the service demander.

6. The method according to claim 1, characterized in that The determining of a candidate service provider by combining the demand constraint of the service demander and the intention probability of the service provider includes: Constructing a matching objective function for the service demander to determine the candidate service providers; The matching objective function is composed of a first data item constructed according to the demand constraint of the service demander and the intention probability of the service provider, a second data item representing the fairness of the service provider, and the demand constraint; Perform multiple iterative calculations on the matching objective function, and determine one or more candidate service providers according to the optimal solution of the calculation results.

7. The method according to claim 6, characterized in that The performing multiple iterative calculations on the matching objective function and determining one or more candidate service providers according to the calculated optimal solution includes: For each iterative calculation performed on the matching objective function, scoring is performed based on whether the matching objective function and the constraint condition are satisfied; An optimal solution of the matching objective function is determined according to the scoring result, and one or more candidate service providers are determined according to the optimal solution.

8. A device for processing business data, characterized in that: include: Acquisition module, prediction module and target determination module; among them, The data acquisition module is used to filter out multiple business providers that match the demand data from the business provider data source according to the demand data published by the business demander; and to acquire the service data of the business providers; The prediction intention module is used to extract demand features from the demand data of the business demander; extract service features from the service data of multiple business providers; and predict the probability of each business provider's intention to provide services to the business demander based on the demand features and the service features; The target determination module is used to determine a candidate service provider from a plurality of service providers in combination with the demand constraints of the service demander and the intention probability of the service provider; and send the demand data of the service demander to the candidate service provider.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

11. A computer program product, comprising a computer program, characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images