Business object recommendation method and device, online business system, and computer device

Abstract

This application relates to a method, apparatus, online business system, computer device, storage medium, and computer program product for recommending business objects. The method includes: responding to a user account's query request for a business object; determining the query object corresponding to the user account and the corresponding business requirement, wherein the business requirement is determined based on requirement conditions under multiple business dimensions; obtaining a recommended object corresponding to the query object and its business parameters from business objects at multiple data sources; generating business recommendation data corresponding to the business requirement based on the business requirement and the business parameters of the recommended object; and recommending the corresponding recommended object to the user account based on the business recommendation data. This method can recommend corresponding business objects to users based on their actual business needs, thereby improving the accuracy of business object recommendations.

Description

Technical Field

[0001] This application relates to the field of big data analytics technology, and in particular to a method, apparatus, online business system, computer equipment, storage medium, and computer program product for recommending business objects. Background Technology

[0002] Material procurement is an indispensable part of business operations. With the advancement of digital transformation, optimizing the procurement management process, reducing procurement costs, and improving procurement efficiency are major concerns for many companies. For businesses, the essence of material procurement is the comparison of product information.

[0003] Traditionally, companies can build their own B2B procurement marketplace by connecting to existing e-commerce platforms. This marketplace then recommends products to the company's purchasing staff, allowing them to compare prices and make selections. However, this traditional product recommendation method is prone to recommending products that don't match the company's purchasing needs, resulting in low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, online business system, computer equipment, computer-readable storage medium, and computer program product with high accuracy for recommending business objects, addressing the aforementioned technical problems.

[0005] Firstly, this application provides a method for recommending business objects. The method includes:

[0006] In response to a user account's query request for a business object, determine the query object corresponding to the user account and the business requirement corresponding to the query object. The business requirement is determined based on requirement conditions under multiple business dimensions.

[0007] Obtain the recommendation object corresponding to the query object, and the business parameters of the recommendation object, from business objects at multiple data sources;

[0008] Based on the business requirements corresponding to the query object and the business parameters of the recommendation object, generate business recommendation data corresponding to the business requirements for the recommendation object;

[0009] Based on the business recommendation data, the recommended object is recommended to the user account.

[0010] In one embodiment, generating business recommendation data corresponding to the business requirements of the query object and the business parameters of the recommendation object includes:

[0011] Obtain the requirement weight for each of the business dimensions corresponding to the stated business requirement;

[0012] For the business parameters of the recommended object under each business dimension, the business parameters are processed according to the demand weight to generate dimension recommendation data corresponding to the recommended object and each business dimension;

[0013] Based on the dimension recommendation data under each of the business dimensions, generate the business recommendation data for the recommended object.

[0014] In one embodiment, the business dimension includes a business value dimension and a business evaluation dimension;

[0015] The step of obtaining the requirement weight under each business dimension corresponding to the business requirement includes:

[0016] Obtain the first parameter weight under the business value dimension corresponding to the business requirement, and the second parameter weight under the business evaluation dimension.

[0017] In one embodiment, when the business requirement is a consumable requirement, the weight of the first parameter is greater than the weight of the second parameter.

[0018] When the business requirement is a durability requirement, the weight of the first parameter is less than the weight of the second parameter.

[0019] In one embodiment, obtaining the recommendation object corresponding to the query object and the business parameters of the recommendation object from business objects at multiple data sources includes:

[0020] The query object is matched with business objects from multiple data sources to determine the similarity between the query object and the business objects.

[0021] Based on the similarity, the recommended object corresponding to the query object is determined from business objects at multiple data sources;

[0022] Obtain the business parameters of the recommended object.

[0023] In one embodiment, generating business recommendation data corresponding to the business requirements of the query object and the business parameters of the recommendation object includes:

[0024] The business parameters and the similarity are weighted using the demand weights corresponding to the business requirements to generate the business recommendation data corresponding to the recommended objects and the business requirements.

[0025] In one embodiment, the step of responding to a user account's query request for a business object, and determining the query object corresponding to the user account and the business requirement corresponding to the query object, includes:

[0026] In response to the query request, obtain the query information entered by the user account;

[0027] Modeling is performed based on the query information to generate the query object corresponding to the user account;

[0028] The query object is input into the classification model to obtain the business category corresponding to the query object, which is output by the classification model.

[0029] Based on the requirement conditions under the multiple business dimensions corresponding to the business category, determine the business requirements corresponding to the query object.

[0030] In one embodiment, the method further includes:

[0031] Obtain the transaction request parameters of the user account for the recommended object;

[0032] Based on the transaction request parameters and the business parameters of the recommended object, a transaction score corresponding to the transaction request parameters is generated;

[0033] In response to the comparison result between the transaction score and the preset score threshold, a transaction approval strategy corresponding to the transaction request parameters is determined;

[0034] The transaction request parameters are processed according to the transaction approval strategy to generate the approval result of the transaction request parameters.

[0035] In one embodiment, the method further includes:

[0036] Collect business object data from multiple of the aforementioned data sources;

[0037] Modeling is performed based on the business object data to generate business objects corresponding to the business object data.

[0038] Secondly, this application also provides a business object recommendation device. The device includes:

[0039] The query object determination module is used to respond to a user account's query request for a business object, determine the query object corresponding to the user account, and the business requirement corresponding to the query object. The business requirement is determined based on requirement conditions under multiple business dimensions.

[0040] The recommendation object acquisition module is used to obtain the recommendation object corresponding to the query object and the business parameters of the recommendation object from business objects in multiple data sources;

[0041] The recommendation data generation module is used to generate business recommendation data corresponding to the business needs of the query object and the business parameters of the recommendation object.

[0042] The business object recommendation module is used to recommend the recommended objects to the user account based on the business recommendation data.

[0043] Thirdly, this application also provides an online business system. The online business system includes:

[0044] The information acquisition module is used to acquire business objects from multiple data sources;

[0045] The user response module is used to respond to a user account's query request for a business object, determine the query object corresponding to the user account, and the business requirement corresponding to the query object. The business requirement is determined based on the requirement conditions under multiple business dimensions. The module receives the recommended object corresponding to the query object and the business recommendation data of the recommended object, and recommends the recommended object to the user account based on the business recommendation data.

[0046] The object recommendation module is used to receive the query object and the business requirement, obtain the recommended object corresponding to the query object and the business parameters of the recommended object from business objects in multiple data sources, and generate business recommendation data corresponding to the business requirement based on the business requirement corresponding to the query object and the business parameters of the recommended object.

[0047] The request approval module is used to obtain the transaction request parameters of the user account for the recommended object; generate a transaction score corresponding to the transaction request parameters based on the transaction request parameters and the business parameters of the recommended object; determine the transaction approval strategy corresponding to the transaction request parameters in response to the comparison result of the transaction score and a preset score threshold; process the transaction request parameters according to the transaction approval strategy, and generate the approval result of the transaction request parameters.

[0048] Fourthly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the recommended method for the business object described in any of the embodiments of the first aspect.

[0049] Fifthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the method for recommending business objects as described in any of the embodiments of the first aspect.

[0050] Sixthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the method for recommending business objects as described in any of the embodiments of the first aspect above.

[0051] The aforementioned methods, devices, online business systems, computer equipment, storage media, and computer program products for recommending business objects, respond to user accounts' query requests for business objects, determine the query object corresponding to the user account, and the corresponding business requirements, which are determined based on requirement conditions under multiple business dimensions; obtain the recommended object corresponding to the query object and the business parameters of the recommended object from business objects from multiple data sources; generate business recommendation data corresponding to the business requirements based on the business requirements of the query object and the business parameters of the recommended object; and recommend the corresponding recommended object to the user account based on the business recommendation data. This approach enables quantitative evaluation of recommended objects based on the user's actual business needs, and recommends objects to the user based on the business recommendation data obtained from the quantitative evaluation of recommended objects, thereby improving both the accuracy and efficiency of business object recommendations.

[0052] Furthermore, since the business object recommendation method provided in this application obtains the recommended object corresponding to the query object from business objects from multiple data sources, it can expand the selection range of recommended objects based on data from multiple data sources, avoid the limitations of a single data source on recommended objects, and thus further improve the recommendation accuracy of business objects. Attached Figure Description

[0053] Figure 1 A flowchart illustrating the recommendation method for business objects in one embodiment. Figure 1 ;

[0054] Figure 2 This is a flowchart illustrating the steps for generating business requirement parameters in one embodiment.

[0055] Figure 3 This is a flowchart illustrating the steps for determining the recommended object in one embodiment;

[0056] Figure 4 This is a flowchart illustrating the steps for determining business requirements in one embodiment;

[0057] Figure 5This is a flowchart illustrating the transaction request approval steps in one embodiment;

[0058] Figure 6 A flowchart illustrating the recommendation method for business objects in another embodiment. Figure 2 ;

[0059] Figure 7 This is a schematic diagram of the architecture of an online business system in one embodiment;

[0060] Figure 8 A structural block diagram of a business object recommendation device in one embodiment;

[0061] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0063] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0064] In one embodiment, such as Figure 1 As shown, a method for recommending business objects is provided. This embodiment uses the application of this method to a server as an example for illustration. It can be understood that this method can also be applied to terminals, and can also be applied to systems including terminals and servers, and implemented through the interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can be smart speakers, smart TVs, smart in-vehicle devices, etc. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. The server can be implemented using a standalone server or a server cluster composed of multiple servers.

[0065] In this embodiment, the method includes the following steps:

[0066] Step S102: In response to a user account's query request for a business object, determine the query object corresponding to the user account and the business requirements corresponding to the query object.

[0067] In this context, a business object can be used to represent objects that a user account can trade. In one example, a business object could be a product sold in an online business system.

[0068] Business requirements can be used to characterize the transaction requirements of business objects determined based on demand conditions under multiple business dimensions. Business requirements can include, but are not limited to, any of the following types of requirements: consumable requirements, durability requirements, and value preservation requirements.

[0069] Business dimensions can include, but are not limited to, at least two of the following dimensions: price, sales volume, channel, brand, quality, and user business evaluation.

[0070] Specifically, the server can pre-store requirement conditions across multiple business dimensions. The server responds to user account query requests for business objects, parsing the requests to obtain the query object corresponding to the user account. The query object is then evaluated using the requirement conditions across multiple business dimensions to determine the matching degree between the requirement conditions and the query object. Based on the matching degree, the requirement conditions matching the query object are determined. Finally, based on the requirement conditions matching the query object, the corresponding business requirements are determined.

[0071] In one example, if the server determines that the demand conditions matching the query object include demand conditions under the price dimension (such as the price needing to be lower than a preset price threshold) and demand conditions under the sales volume dimension (such as the sales volume needing to be higher than a preset sales volume threshold), it can be determined that the business demand corresponding to the query object is a consumable demand.

[0072] In another example, if the server determines that the requirements matching the query object include requirements under the quality dimension (such as a quality lifecycle longer than a preset lifecycle range) and requirements under the user business evaluation dimension (such as a user evaluation score higher than a preset user rating threshold), then the business requirement corresponding to the query object can be determined to be a durability requirement.

[0073] Step S104: Obtain the recommended object corresponding to the query object and the business parameters of the recommended object from the business objects at multiple data sources.

[0074] The data source can include, but is not limited to, various channels such as personal-to-business e-commerce channels, personal-to-personal e-commerce channels, and business-to-business e-commerce channels.

[0075] Business objects can be used to represent goods that are in a tradable state in e-commerce channels.

[0076] Business parameters may include, but are not limited to, one or more of the following: price parameters, sales volume parameters, quality parameters, channel parameters, user business evaluation parameters, brand parameters, etc.

[0077] Specifically, the server can deploy a database that stores business objects collected from multiple data sources. The query object can be compared with the business objects in the database to determine the similarity between them. If the similarity exceeds a preset threshold, the currently compared business object is selected as the recommended object corresponding to the query object. The business parameters of the recommended object are retrieved from the database. There can be one or more recommended objects.

[0078] Step S106: Based on the business requirements corresponding to the query object and the business parameters of the recommended object, generate business recommendation data corresponding to the business requirements of the recommended object.

[0079] Specifically, the server can score the business parameters of the recommended object under the corresponding business dimension based on the business needs corresponding to the query object, and generate business recommendation data corresponding to the recommended object and business needs. In one example, if the business need corresponding to the query object is determined by the server to be a consumable need, the server can score the price parameter under the price dimension and the sales parameter under the sales volume dimension of the recommended object based on the consumable need, and generate business recommendation data corresponding to the consumable need. In another example, if the business need corresponding to the query object is a value preservation need, the server can score the quality parameter under the quality dimension and the brand parameter (such as manufacturer parameters) under the brand dimension of the recommended object based on the value preservation need, and generate business recommendation data corresponding to the value preservation need.

[0080] Step S108: Based on the business recommendation data, recommend the corresponding recommended objects to the user account.

[0081] Specifically, the server can compare business recommendation data with dimensional recommendation data thresholds. If the business recommendation data for a recommended object exceeds the dimensional recommendation data threshold, the recommended object is displayed in the display area, and the recommended object corresponding to the currently compared business recommendation data is recommended to the user's account. Alternatively, the server can determine the recommendation priority of recommended objects based on the business recommendation data. According to the recommendation priority, the recommended objects are displayed in the area corresponding to the recommendation priority in the display area and recommended to the user's account.

[0082] In one example, when there are multiple recommended objects, the server can first sort the recommended objects according to their price parameters. Then, it obtains the price parameter difference between each pair of recommended objects after sorting. Based on the price parameter differences and the business recommendation data for each recommended object, the server adjusts the sorting results for each object, and uses the adjusted sorting results as the recommendation priority for each object.

[0083] The aforementioned method for recommending business objects, in response to a user account's query request for a business object, determines the query object corresponding to the user account and the corresponding business requirement. The business requirement is determined based on demand conditions across multiple business dimensions. It retrieves the recommended object corresponding to the query object and its business parameters from business objects stored in multiple data sources in the database. Based on the business requirement corresponding to the query object and the business parameters of the recommended object, it generates business recommendation data corresponding to the business requirement. Based on this business recommendation data, it recommends the corresponding recommended object to the user account. This method allows for quantitative evaluation of recommended objects based on the user's actual business needs and provides recommendations to the user based on the business recommendation data of the recommended object, thus improving both the accuracy and efficiency of business object recommendations. Furthermore, because the method for recommending business objects provided in this application retrieves the recommended object corresponding to the query object from business objects in multiple data sources, it expands the selection range of recommended objects based on data from multiple data sources, avoiding the limitations of a single data source and further improving the accuracy of business object recommendations.

[0084] In one embodiment, such as Figure 2 As shown, step S106 involves generating business recommendation data corresponding to the business requirements of the queried object and the business parameters of the recommended object, including:

[0085] Step S202: Obtain the requirement weight for each business dimension corresponding to the business requirements.

[0086] Step S204: For the business parameters of the recommended object under each business dimension, process the business parameters according to the demand weight to generate the dimension recommendation data corresponding to the recommended object and each business dimension.

[0087] Step S206: Generate business recommendation data for the recommended objects based on the dimension recommendation data under each business dimension.

[0088] Specifically, the server can determine multiple business dimensions corresponding to the business requirements of the query object. It then obtains the requirement weight for each business dimension. For the business parameters of the recommended object under each business dimension, the server processes the business parameters using the requirement weight corresponding to the business dimension, generating dimension recommendation data for the recommended object and each corresponding business dimension. Finally, it performs calculations on the dimension recommendation data for the recommended object and each corresponding business dimension to generate business recommendation data corresponding to the business requirements.

[0089] In this embodiment, by obtaining the demand weight under each business dimension corresponding to the business needs, and calculating the dimension recommendation data of the recommended object under each business dimension according to the demand weight and the business parameters under the corresponding business dimension, the business recommendation data of the recommended object corresponding to the business needs is calculated based on the dimension recommendation data under each business dimension. This enables the business recommendation data of the recommended object to meet the scoring criteria of the business needs under each business dimension, thereby improving the accuracy of the business recommendation data.

[0090] In one embodiment, the business dimension may include a business value dimension and a business evaluation dimension. Step S202, obtaining the requirement weight under each business dimension corresponding to the business requirement, includes: obtaining the first parameter weight under the business value dimension corresponding to the business requirement, and the second parameter weight under the business evaluation dimension.

[0091] Among them, business parameters under the business value dimension can include, but are not limited to, any one or more of various parameters such as price parameters and sales volume parameters.

[0092] Business parameters under the business evaluation dimension can include, but are not limited to, one or more of the following parameters: quality parameters, channel parameters, user business evaluation parameters, brand parameters, etc.

[0093] Specifically, the server can obtain the first parameter weight under the business value dimension corresponding to the business needs of the queried object, and the second parameter weight under the business evaluation dimension.

[0094] When the business requirement is a consumable requirement, since the price and sales parameters need to meet preset requirements, the weight of the first parameter under the business value dimension is set to be greater than the weight of the second parameter under the business evaluation dimension. This makes the parameter weight under the business value dimension in the business requirement parameters higher and more in line with the consumable requirements of the query object.

[0095] When the business requirement is to preserve value, this requires brand, quality, and channel parameters to meet preset conditions. Therefore, the weight of the first parameter under the business value dimension is set lower than the weight of the second parameter under the business evaluation dimension. This ensures that the parameters under the business evaluation dimension have a higher weight in the business requirement parameters, better aligning with the value preservation requirements of the query object.

[0096] In this embodiment, by obtaining the first parameter weight corresponding to the business demand and business value dimensions, and the second parameter weight corresponding to the business evaluation dimension, the recommended object can be scored from the business value dimension and the business evaluation dimension, respectively. This makes the business recommendation data obtained based on the parameter weights more suitable for the business demand scoring scenario, thereby improving the accuracy of the subsequently generated business recommendation data.

[0097] In one embodiment, such as Figure 3 As shown, step S104 involves obtaining the recommendation object corresponding to the query object and the business parameters of the recommendation object from business objects at multiple data sources, including:

[0098] Step S302: Match the query object with business objects from multiple data sources to determine the similarity between the query object and the business objects.

[0099] Step S304: Based on similarity, determine the recommended object corresponding to the query object from business objects from multiple data sources.

[0100] Step S306: Obtain the business parameters of the recommended object.

[0101] Specifically, the server can pre-store similarity algorithms. The query object is matched with business objects from multiple data sources in the database, and the similarity algorithm is used to calculate the similarity between the query object and the business objects. The similarity algorithm can include, but is not limited to, one or more of the following algorithms: Pearson correlation coefficient algorithm, Jaccard similarity coefficient algorithm, cosine similarity algorithm, etc.

[0102] After determining the similarity, each business object can be sorted from highest to lowest based on its similarity to the query object. The top n business objects in the sorted list are then selected as recommended objects corresponding to the query object. Alternatively, the similarity can be compared to a preset threshold; if the similarity exceeds the threshold, the currently compared business object is selected as the recommended object corresponding to the query object. The business parameters of the recommended objects are then retrieved from the database.

[0103] In one example, the similarity between the query object and the business object can be determined using the following formula:

[0104]

[0105] Where, r AB It can be used to characterize the similarity between query objects and business objects. A can be used to characterize the query object, B can be used to characterize the business object, conv(A,B) can be used to characterize the covariance between the query object and the business object, σ A It can be used to characterize the standard deviation of the query object, σ B It can be used to characterize the standard deviation of business objects.

[0106] The similarity score generated by the server ranges between -1 and 1. A similarity greater than zero indicates a positive correlation between the query object and the business object. A similarity less than zero indicates a negative correlation. A similarity of zero indicates no correlation between the query object and the business object. The closer the absolute value of the similarity score is to 1, the higher the correlation between the query object and the business object. A similarity of 1 indicates that the query object and the business object can be well described by a linear equation, with all data points lying on a straight line, and the data of the business object increasing as the data of the query object increases. A similarity of -1 indicates that all data points lie on a straight line, and the data of the business object decreasing as the data of the query object increases.

[0107] In this embodiment, by using a similarity algorithm to determine the similarity between the query object and the business object, and then determining the recommended object from the business object based on the similarity, the accuracy of the recommended object can be improved.

[0108] In one embodiment, step S106, generating business recommendation data corresponding to the business requirements of the queried object and the business parameters of the recommended object, includes: weighting the business parameters and similarity with the requirement weight corresponding to the business requirements to generate business recommendation data of the recommended object.

[0109] Specifically, the server can obtain the demand weights corresponding to business needs. These demand weights can include parameter demand weights corresponding to business parameters and similarity demand weights corresponding to similarity. For the business parameters under each business dimension of the recommended object, the parameter demand weights corresponding to the business dimension can be used to process the business parameters, resulting in weighted business parameters, i.e., dimension recommendation data. For similarity, the similarity demand weights corresponding to the similarity can be used to process the similarity, resulting in weighted similarity. The weighted business parameters and weighted similarity are then processed to generate the business recommendation data for the recommended object.

[0110] In one example, the similarity requirement weight can also be calculated by the server. For instance, the server can use the same demand conditions across multiple business dimensions as the query object to judge the recommended object and determine its business requirements. Based on the business requirements of the recommended object and the query object, a similarity coefficient can be determined between the business requirements of the recommended object and the query object, and this similarity coefficient is used as the similarity requirement weight. A higher similarity coefficient indicates a higher consistency between the business requirements of the recommended object and the query object.

[0111] In another example, the server can generate business recommendation data using the following formula:

[0112] T i =(1-α)(P i +S i )+α(B i +R i )+γr i i = 1, 2, ..., n

[0113] Among them, T i It can be used to characterize the business recommendation data for recommendation object i. α can be used to characterize the parameter requirement weights. P i It can be used to characterize the price parameter of the recommended object i. i This can be used to characterize the sales volume of recommended object i. B i This can be used to characterize the channel parameters (e.g., transaction channel) of the recommended object i. i γ can be used to characterize user rating parameters for recommended object i. γ can be used to characterize the similarity weight of business needs between recommended object i and the query object. i It can be used to characterize the similarity between the recommended object i and the query object.

[0114] Preferably, the server can determine the price coefficient of the recommended item based on its price parameter, and use this price coefficient as the parameter demand weight. A higher price coefficient indicates a higher probability that the recommended item is a valuable item. In this case, the parameter demand weight corresponding to the user review parameters and channel parameters under the business evaluation dimension is greater than the parameter demand weight corresponding to the price parameters and sales parameters under the business value dimension.

[0115] In this embodiment, by determining the parameter requirement weights corresponding to the business parameters and the similarity requirement weights corresponding to the similarity based on business needs, the business parameters are weighted using the parameter requirement weights, and the similarity is weighted using the similarity requirement weights. Based on the data obtained after weighting, business recommendation data for recommended objects is calculated, which can improve the accuracy of business recommendation data. Furthermore, it makes the recommended objects recommended based on the business recommendation data more in line with the business needs of the user account.

[0116] In one embodiment, such as Figure 4 As shown, step S102, in response to a user account's query request for a business object, determines the query object corresponding to the user account and the business requirements corresponding to the query object, including:

[0117] Step S402: In response to the query request, obtain the query information entered by the user account.

[0118] Step S404: Model based on the query information to generate a query object corresponding to the user account.

[0119] Specifically, the server can respond to a user account's query request for a business object, parse the query request to obtain the query information entered by the user account, such as the business object name, transaction channel, and brand. Based on the query information, it retrieves the corresponding business object data from the database. Using the retrieved business object data, it models and generates a query object corresponding to the user account.

[0120] Step S406: Input the query object into the classification model to obtain the business category corresponding to the query object output by the classification model.

[0121] Step S408: Determine the business requirements corresponding to the query object based on the requirement conditions under multiple business dimensions corresponding to the business category.

[0122] Among them, business category can be used to represent the category of business object. For example, business category can include, but is not limited to, any of the following categories: consumables, durable goods, valuables, etc.

[0123] Specifically, a classification model can be pre-deployed on the server. The query object is input into the classification model, which extracts the vector features of the query object. These vector features are then matched with the vector features corresponding to preset business categories to determine the preset business category that matches the query object's vector features. The business category corresponding to the query object is then output. The requirement conditions under multiple business dimensions corresponding to the business category are obtained to determine the business requirements corresponding to the query object.

[0124] In one example, when the business category is consumables, consumables are characterized by low prices and high transaction volumes. Therefore, if multiple recommended objects have the same similarity to the query object, they are evaluated primarily based on price and sales volume. In this case, the demand conditions corresponding to the business category obtained by the server can include demand conditions based on both price and sales volume. The server can then use these price and sales volume demand conditions to evaluate the query object and determine that the business demand for the query object is a consumable demand.

[0125] In another example, when the business category is valuable goods, the prices of valuable goods are relatively high, and users pay more attention to the brand, quality, and transaction channel (such as the purchasing platform) when purchasing. In this case, the demand conditions obtained by the server corresponding to the business category can include demand conditions based on brand, quality, and channel. The server can use these demand conditions to evaluate the query object and determine that the business demand for the query object is a value preservation demand.

[0126] In this embodiment, by simulating the business objects required by the user account, the query information is modeled to generate query objects corresponding to the user account. The business category of the query object is determined using a classification model, and the business requirements of the query object are determined based on the business category. This not only improves the accuracy of business requirements, but also makes the business requirements determination method applicable to business requirements determination scenarios under multiple business categories, thereby improving the flexibility of the business requirements determination method.

[0127] In one embodiment, such as Figure 5 As shown, a flowchart illustrating the transaction approval process is provided, including:

[0128] Step S502: Obtain the transaction request parameters of the user account for the recommended object.

[0129] Step S504: Generate a transaction score corresponding to the transaction request parameters based on the transaction request parameters and the business parameters of the recommended object.

[0130] The transaction request parameters may include, but are not limited to, any one or more of the following: transaction time, transaction channel, transaction quantity, etc.

[0131] Specifically, after the server recommends a corresponding item to a user account, it can also respond to the user account's transaction request for the recommended item, obtaining the transaction request parameters, such as the transaction quantity. The transaction request parameters and the business parameters of the recommended item are then processed to generate a transaction score corresponding to the transaction request parameters.

[0132] In one example, the server can generate a transaction score using the following formula:

[0133] y i =w0+w1A i1 +w2A i2 +…+w m A im +∈ i i = 1, 2, ..., n

[0134] Among them, y i This can be used to characterize the random dependent variable, namely the transaction score of the recommendation object i. w0,…,w m It can be used to characterize regression coefficients. A i1 A i2 ;...A im The feature variables can be used to characterize non-random recommendation objects i. These feature variables can be determined based on business parameters such as the business category, user rating parameters, and price parameters of recommendation object i, as well as the number of transactions in the transaction request parameters. The variable ∈ can be used to characterize the random error term, capable of capturing any pair of variables other than the feature variables y. iThe impact.

[0135] Step S506: In response to the comparison result between the transaction score and the preset score threshold, determine the transaction approval strategy corresponding to the transaction request parameters.

[0136] Step S508: Process the transaction request parameters according to the transaction approval strategy and generate the approval result of the transaction request parameters.

[0137] Specifically, the server can compare the transaction score of the recommended object with a preset score threshold (e.g., 0.8, which can be adjusted according to the actual application scenario).

[0138] If the transaction score exceeds a preset threshold, the server can determine, based on the comparison between the transaction score and the preset threshold, that the transaction approval strategy corresponding to the transaction request parameters can be an automatic approval strategy. The automatic approval strategy is then used to process the transaction request parameters, generating an approval result for the parameters corresponding to the current transaction request. Subsequently, the server can send the approval result back to the user's account, allowing the user to execute the corresponding transaction operation in a third-party channel based on the approval result. For example, if the approval result is approved, the company's purchasing personnel can purchase the recommended object from the corresponding transaction channel according to the transaction request parameters.

[0139] If the transaction score is less than a preset score threshold, the server can respond to the comparison result between the transaction score and the preset score threshold, and determine that the transaction approval strategy corresponding to the transaction request parameters can be a manual approval strategy. Based on the manual approval strategy, the approval process for the transaction request parameters is sent to the review account, prompting the review account to approve the transaction request parameters corresponding to the current transaction request, and generating the approval result for the transaction request parameters.

[0140] In one example, when the recommended item is a low-priced, high-volume consumable product, manually approving the transaction request parameters would require significant manpower. Therefore, using an automated approval strategy with an approval model to make approval decisions on the transaction request parameters can improve the efficiency of the approval process and reduce manpower costs and processing time.

[0141] In another example, when the recommended items are high-priced, low-volume value-preserving goods, due to the importance of the goods and the procurement costs, it is necessary to increase the proportion of manual review and be more cautious in reviewing each procurement expenditure. Therefore, in this case, a manual review strategy is adopted to approve the transaction request parameters.

[0142] In this embodiment, by obtaining the transaction request parameters of the user account for the recommended object and combining them with the business parameters of the recommended object, a transaction score for the transaction request parameters is generated. Based on the transaction score, a corresponding transaction approval strategy is determined, and the transaction request parameters are approved using the corresponding transaction approval strategy. This can reduce the approval cost of the transaction request parameters and improve the approval efficiency of the transaction request parameters.

[0143] In one embodiment, the method for recommending business objects may further include: collecting business object data from multiple data sources, modeling based on the business object data, generating business objects corresponding to the business object data, and storing the business objects in a database.

[0144] The business object data may include, but is not limited to, any one or more of the following: text description data, image data, link data, price data, associated object data, channel data, after-sales data, user review data, etc.

[0145] Specifically, the server can collect business object data from multiple data sources at preset intervals. The collected business object data undergoes data preprocessing, employing one or more of the following methods: sampling analysis, scale analysis, missing value handling, outlier handling, data transformation, and data filtering. Based on the preprocessed business object data, a model is created, generating business objects corresponding to the data, and these business objects are stored in the database.

[0146] In one example, the server can match multiple business objects in the database, determine the similarity between them, and establish associations between them based on the similarity. When a user account performs a query, the server can return the business object currently being queried by the user account, along with the business objects associated with it, to the user account for comparison and selection.

[0147] In this embodiment, by collecting business object data from multiple data sources, modeling and generating corresponding business objects, and storing the business objects in the database, data from multiple data sources can be integrated, the amount of data in the database can be expanded, the recommended objects can be enriched, and the robustness of the business object recommendation method can be improved.

[0148] In one embodiment, such as Figure 6 As shown, a method for recommending business objects is provided, including:

[0149] Step S602: In response to a user account's query request for a business object, obtain the query information entered by the user account, and model the query object based on the query information.

[0150] Specifically, the server can respond to a user account's query request for a business object, parse the query request to obtain the query information entered by the user account, perform data preprocessing on the query information, and use the preprocessed data to model and generate a query object corresponding to the user account's query information.

[0151] Step S604: Input the query object into the classification model to obtain the business category corresponding to the query object output by the classification model, and determine the business requirements based on the requirement conditions corresponding to the business category.

[0152] Specifically, the server can pre-deploy a decision tree classification model. The query object is input into the decision tree classification model, and the model outputs the business category corresponding to the query object. The query object is then processed using requirement conditions across multiple business dimensions corresponding to the business category to determine its business requirements. The specific business requirement determination operation can be implemented using the business requirement determination method provided in the above embodiments, and will not be elaborated upon here.

[0153] Step S606: Match the query object with the business object in the database, determine the similarity between the query object and the business object, and determine the recommended object based on the similarity.

[0154] Specifically, the server can match the query object with business objects collected from multiple data sources in the database to determine the similarity between the query object and the business objects. In response to comparison results where the similarity exceeds a preset threshold, the currently compared business object is used as the recommended object corresponding to the query object.

[0155] Step S608: The business parameters and similarity of the recommended objects are weighted according to the demand weights corresponding to the business requirements to generate business recommendation data for the recommended objects.

[0156] Step S610: Based on the business recommendation data, recommend the corresponding recommended objects to the user account.

[0157] Specifically, the server can obtain parameter requirement weights for multiple business dimensions corresponding to business needs, as well as similarity requirement weights corresponding to similarity. For the business parameters of the recommended object under each business dimension, the business parameters are weighted using the parameter requirement weights corresponding to the business dimension to obtain the weighted business parameters, i.e., dimension recommendation data. Similarity is weighted using similarity requirement weights to obtain the weighted similarity. The weighted business parameters and weighted similarity are then processed, and the sum of the weighted business parameters and weighted similarity is used as the business recommendation data for the recommended object. Based on the business recommendation data, the recommendation priority of the recommended object is determined, and the corresponding recommended object is recommended to the user account according to the recommendation priority.

[0158] In this embodiment, by modeling the query information of user accounts to generate query objects, the accuracy of query objects can be improved, making it easier to identify similar recommended objects from the database. By using a classification model to output the business category corresponding to the query object, and determining the business requirements of the query object based on the business category, the business requirements can be matched with the corresponding business category, thereby simplifying the business requirements determination process and improving the efficiency of business requirements determination. By matching the query object with business objects collected from multiple data sources in the database, the recommended objects corresponding to the query object can be determined, which can integrate business object data from multiple data sources, expand the data coverage of the database, and enhance the diversity of recommended objects. By using demand weights corresponding to business requirements for weighted processing to generate business recommendation data for recommended objects, the recommended objects can be evaluated from different dimensions based on the user's business requirements. By using demand weights of different dimensions, the importance of parameters under different dimensions to the user account can be distinguished, thereby improving the accuracy of business recommendation data. By recommending corresponding recommended objects to the user account based on business recommendation data, the recommended objects can not only be similar to the query object, but also meet the business requirements of the query object, thereby improving the accuracy of recommended objects.

[0159] In one embodiment, the user account can be a corporate user account operated by corporate procurement personnel. Based on the business object recommendation method provided in the above embodiments, this application also provides an online business system applied to corporate procurement management. For example... Figure 7 As shown, an online business system 700 is provided, including: an information collection module 702, an object recommendation module 704, a user response module 706, and a request approval module 708. The online business system 700 can implement the business object recommendation method provided in any of the above method embodiments.

[0160] Specifically, the online business system 700 can connect to multiple data sources through the information collection module 702 to obtain business objects from multiple data sources.

[0161] In one example, when the business object is a product, the data source can be an existing online e-commerce platform. The information collection module 702 can connect to multiple existing online e-commerce platforms, acquiring product information from each platform. This allows the user response module 706 to compare product prices across multiple platforms, selecting the lower-priced product and further reducing the company's procurement costs. Furthermore, by collecting and storing product information from multiple online e-commerce platforms through the information collection module 702, the number of times procurement personnel need to switch between platforms to compare prices can be reduced, thus improving the efficiency of product comparison. The information collection module 702 can preprocess the product information using the business object modeling method provided in the above embodiments, using the preprocessed product information to model and generate corresponding products, which are then stored in a database. Data preprocessing can include, but is not limited to, one or more of the following methods: sampling analysis, scale analysis, missing value handling, outlier handling, data transformation, and data filtering. Product information can include, but is not limited to, one or more of the following data: product description, product images, details links, price, related products, platform source, after-sales service, and user reviews.

[0162] The online business system 700 can respond to query requests from enterprise user accounts regarding business objects through the user response module 706, determine the query object corresponding to the enterprise user account and the corresponding business requirements, and send the query object and its business requirements to the object recommendation module 704. Subsequently, the user response module 706 can receive the recommended object and its business recommendation data returned by the object recommendation module 704, and recommend the current recommended object to the enterprise user account based on the business recommendation data.

[0163] In one example, when the business object is a product, the user response module 706 can respond to product query requests based on multiple business dimensions such as product name, product category, and product transaction channel, and determine the query product corresponding to the enterprise user account. Based on the requirements under multiple business dimensions, it determines the business requirements corresponding to the query product and sends the query product and its business requirements to the object recommendation module 704.

[0164] The object recommendation module 704 receives the query object and its business requirements. It then determines the recommended object corresponding to the query object, along with the recommended object's business parameters, from multiple data sources obtained by the information collection module 702. Based on the business requirements of the query object and the recommended object's business parameters, it generates business recommendation data corresponding to the business requirements.

[0165] In one example, when the business object is a product, the object recommendation module 704 can determine the similarity between the queried product and products in the database. It retrieves the top n products with the highest similarity, from highest to lowest, and uses these top n products as recommended products corresponding to the queried object. Referring to the business requirement parameter generation method provided in the above embodiments, the business parameters and similarity of the recommended object are weighted according to the procurement requirements corresponding to the queried product to generate the business requirement parameters for the recommended products. The top n products are then sorted according to the business requirement parameters to generate a sorted list of recommended products, which is then sent to the user response module 706 for display.

[0166] The user response module 706 responds to the enterprise user account's selection operation on the recommended product list by adding the selected recommended products to the shopping cart or favorites. In response to the enterprise user account's purchase request for the recommended products, the module obtains the purchase quantity for the recommended products, generates a purchase transaction request corresponding to the recommended products, and sends the purchase transaction request to the request approval module 708.

[0167] The request approval module 708 can obtain the transaction request parameters of the enterprise user account for the recommended object; generate a transaction score corresponding to the transaction request parameters based on the transaction request parameters and the business parameters of the recommended object; determine the transaction approval strategy corresponding to the transaction request parameters based on the comparison result of the transaction score and the preset score threshold; process the transaction request parameters according to the transaction approval strategy and generate the approval result of the transaction request parameters.

[0168] In one example, when the business object is a product, the request approval module 708 can refer to the transaction score generation method in the above embodiment to evaluate the product category, product rating, product price, and purchase quantity of the recommended product, and generate a transaction score for the purchase transaction request. Based on the transaction score, a transaction approval strategy for the purchase transaction request is determined, and the purchase transaction request is reviewed using the transaction approval strategy. The review result is then generated and sent to the user response module 706 for display.

[0169] In one example, the online business system 700 can also include a permission management module. The product management module can be used to manage the purchasing permissions of different departments within the enterprise. For instance, enterprise user accounts first select the departments that need to make purchases, and then add the product categories that the department can purchase, or perform refined management based on a product whitelist. This allows for restrictions on the monthly purchase limit for product categories for departments, as well as restrictions on the maximum price, minimum purchase quantity, maximum purchase quantity, and monthly purchase quantity of individual products, enhancing control over the purchasing process and preventing unauthorized purchasing operations.

[0170] In one example, the information collection module 702 can collect product information from multiple data sources at preset intervals. It then matches the currently collected product information with products in the database. If a match is successful, it updates the product data in the database using the currently collected product information and stores the updated product information.

[0171] In this embodiment, by constructing an information collection module, a user response module, an object recommendation module, and a request approval module in the online business system, the information collection module integrates data from multiple data sources, enriching the data sources for purchasing goods. The object recommendation module identifies recommended goods with high similarity and low prices from goods collected from multiple data sources, improving the efficiency of price comparison and further reducing the enterprise's procurement costs. The user response module responds to user account query requests, recommending corresponding objects to the user account based on business recommendation data, improving user interaction efficiency. The request approval module determines the transaction score corresponding to the user account's transaction request and approves the transaction request based on the transaction score using a corresponding approval strategy, improving the efficiency of transaction request approval.

[0172] Furthermore, while traditional B2B trading platforms are suitable for large enterprises, their integration with existing e-commerce platforms allows for end-to-end monitoring of the entire procurement process, from order placement to completion, facilitating the management of procurement activities. However, for SMEs, traditional B2B trading platforms are not suitable due to the high cost of establishing such platforms and the relatively weak bargaining power of SMEs, leading to higher prices on e-commerce platforms.

[0173] The online business system provided in this application uses a technical means to compare products from multiple data sources to determine the products that best meet the company's procurement needs. Compared with the aforementioned traditional B2B trading platforms, this not only reduces the setup cost of the online business system but also further reduces the company's procurement costs.

[0174] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0175] Based on the same inventive concept, this application also provides a business object recommendation apparatus for implementing the business object recommendation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in the one or more business object recommendation apparatus embodiments provided below can be found in the limitations of the business object recommendation method described above, and will not be repeated here.

[0176] In one embodiment, such as Figure 8 As shown, a business object recommendation device 800 is provided, including: a query object determination module 802, a recommendation object acquisition module 804, a recommendation data generation module 806, and a business object recommendation module 808, wherein:

[0177] The query object determination module 802 is used to respond to a user account's query request for a business object, determine the query object corresponding to the user account, and the business requirements corresponding to the query object. The business requirements are determined based on the requirements conditions under multiple business dimensions.

[0178] The recommendation object acquisition module 804 is used to obtain the recommendation object corresponding to the query object and the business parameters of the recommendation object from business objects in multiple data sources.

[0179] The recommendation data generation module 806 is used to generate business recommendation data corresponding to the business requirements of the query object and the business parameters of the recommendation object.

[0180] The business object recommendation module 808 is used to recommend corresponding recommended objects to user accounts based on business recommendation data.

[0181] In one embodiment, the recommendation data generation module 806 includes: a weight acquisition unit, used to acquire the demand weight under each business dimension corresponding to the business requirements; a dimension data generation unit, used to process the business parameters of the recommended object under each business dimension according to the demand weight, generate dimension recommendation data of the recommended object and each business dimension; and generate business recommendation data of the recommended object based on the dimension recommendation data under each business dimension.

[0182] In one embodiment, the business dimension includes a business value dimension and a business evaluation dimension. The weight acquisition unit is further configured to acquire a first parameter weight under the business value dimension corresponding to the business requirement, and a second parameter weight under the business evaluation dimension; wherein, when the business requirement is a consumable requirement, the first parameter weight is greater than the second parameter weight; when the business requirement is a durability requirement, the first parameter weight is less than the second parameter weight.

[0183] In one embodiment, the recommendation object acquisition module 804 includes: a similarity determination unit, used to match the query object with business objects from multiple data sources to determine the similarity between the query object and the business objects; a recommendation object determination unit, used to determine the recommendation object corresponding to the query object from the business objects from multiple data sources based on the similarity; and a business parameter acquisition unit, used to acquire the business parameters of the recommendation object.

[0184] In one embodiment, the recommendation data generation module 806 is further configured to weight the business parameters and similarity using demand weights corresponding to business requirements, and generate business recommendation data corresponding to the business requirements for the recommended objects.

[0185] In one embodiment, the query object determination module 802 includes: an information acquisition unit, used to acquire query information input by a user account in response to a query request; an information modeling unit, used to model based on the query information and generate a query object corresponding to the user account; a category determination unit, used to input the query object into a classification model and obtain the business category corresponding to the query object output by the classification model; and a requirement determination unit, used to determine the business requirements corresponding to the query object based on the requirement conditions under multiple business dimensions corresponding to the business category.

[0186] In one embodiment, the business object recommendation device 800 further includes an approval module.

[0187] The approval module includes: a request parameter acquisition unit, used to acquire transaction request parameters of the user account for the recommended object; a transaction score generation unit, used to generate a transaction score corresponding to the transaction request parameters based on the transaction request parameters and the business parameters of the recommended object; an approval strategy determination unit, used to determine the transaction approval strategy corresponding to the transaction request parameters in response to the comparison result of the transaction score and the preset score threshold; and an approval result generation unit, used to process the transaction request parameters according to the transaction approval strategy and generate the approval result of the transaction request parameters.

[0188] In one embodiment, the business object recommendation device 800 further includes: a data acquisition module for acquiring business object data from multiple data sources; and an object modeling module for modeling based on the business object data to generate business objects corresponding to the business object data.

[0189] Each module in the aforementioned business object recommendation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0190] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores business objects and business parameters. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a recommended method for a business object.

[0191] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0192] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0193] In one embodiment, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps in the above method embodiments.

[0194] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0195] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0196] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0197] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for recommending business objects, characterized in that, The method includes: In response to a user account's query request for a business object, the query object corresponding to the user account and the business requirement corresponding to the query object are determined. The business requirement is determined based on requirement conditions under multiple business dimensions. The user account is an enterprise user account operated by the enterprise's procurement personnel. The business requirement includes consumable requirements or durable requirements. The recommended object corresponding to the query object and the business parameters of the recommended object are obtained from business objects from multiple data sources, where the data source is an e-commerce channel; Based on the business requirements corresponding to the query object and the business parameters of the recommended object, business recommendation data corresponding to the business requirements is generated. This step specifically includes: obtaining the parameter requirement weights under each business dimension corresponding to the business requirements. The business dimensions include business value dimension and business evaluation dimension. Obtaining the parameter requirement weights under each business dimension corresponding to the business requirements includes: obtaining a first parameter weight under the business value dimension and a second parameter weight under the business evaluation dimension. When the business requirement is a consumable requirement, the first parameter weight is greater than the second parameter weight; when the business requirement is a durability requirement, the first parameter weight is less than the second parameter weight. For the business parameters of the recommended object under each business dimension, the business parameters are processed according to the parameter requirement weights to generate dimension recommendation data corresponding to each business dimension. Based on the dimension recommendation data under each business dimension, the business recommendation data is generated. Based on the business recommendation data, the recommended object is recommended to the user account; The step of obtaining the recommendation object corresponding to the query object from business objects at multiple data sources, and the business parameters of the recommendation object, includes: The query object is matched with business objects from multiple data sources to determine the similarity between the query object and the business objects. Based on the similarity, the recommended object corresponding to the query object is determined from business objects at multiple data sources; Obtain the business parameters of the recommended object; The step of generating business recommendation data corresponding to the business requirements of the query object and the business parameters of the recommendation object further includes: The business parameters are processed using parameter demand weights corresponding to the business dimensions to obtain weighted dimension recommendation data. The similarity is processed using similarity demand weights corresponding to the similarity to obtain weighted similarity. The dimension recommendation data and the weighted similarity are then processed to generate business recommendation data corresponding to the business demand.

2. The method of claim 1, wherein, The step of responding to a user account's query request for a business object, and determining the query object corresponding to the user account and the business requirement corresponding to the query object, includes: In response to the query request, obtain the query information entered by the user account; Modeling is performed based on the query information to generate the query object corresponding to the user account; The query object is input into the classification model to obtain the business category corresponding to the query object, which is output by the classification model. Based on the requirement conditions under the multiple business dimensions corresponding to the business category, determine the business requirements corresponding to the query object.

3. The method of any one of claims 1-2, wherein, The method further includes: Obtain the transaction request parameters of the user account for the recommended object; Based on the transaction request parameters and the business parameters of the recommended object, a transaction score corresponding to the transaction request parameters is generated; In response to the comparison result between the transaction score and the preset score threshold, a transaction approval strategy corresponding to the transaction request parameters is determined; The transaction request parameters are processed according to the transaction approval strategy to generate the approval result of the transaction request parameters.

4. The method of any one of claims 1-2, wherein, The method further includes: Collect business object data from multiple of the aforementioned data sources; Modeling is performed based on the business object data to generate a business object corresponding to the business object data.

5. A recommendation apparatus of a business object, characterized by, The device includes: The query object determination module is used to respond to a user account's query request for a business object, determine the query object corresponding to the user account, and the business requirements corresponding to the query object. The business requirements are determined based on the requirements conditions under multiple business dimensions. The user account is an enterprise user account operated by the enterprise's procurement personnel. The business requirements include consumable requirements or durable requirements. The recommendation object acquisition module is used to obtain the recommendation object corresponding to the query object and the business parameters of the recommendation object from business objects in multiple data sources, wherein the data source is an e-commerce channel; The recommendation data generation module is used to generate business recommendation data corresponding to the business needs of the query object and the business parameters of the recommendation object. The business object recommendation module is used to recommend the recommended object to the user account based on the business recommendation data; The recommendation data generation module includes: The weight acquisition unit is used to acquire the parameter requirement weight under each business dimension corresponding to the business requirement. The business dimension includes a business value dimension and a business evaluation dimension. The weight acquisition unit is also used to acquire the first parameter weight under the business value dimension corresponding to the business requirement and the second parameter weight under the business evaluation dimension. When the business requirement is a consumable requirement, the first parameter weight is greater than the second parameter weight. When the business requirement is a durable requirement, the first parameter weight is less than the second parameter weight. The dimensional data generation unit is used to process the business parameters of the recommended object under each business dimension according to the parameter demand weight, and generate dimensional recommendation data corresponding to the recommended object and each business dimension; and generate the business recommendation data based on the dimensional recommendation data under each business dimension. The recommendation object acquisition module includes: a similarity determination unit, used to match the query object with multiple business objects from the data sources to determine the similarity between the query object and the business objects; a recommendation object determination unit, used to determine the recommendation object corresponding to the query object from the business objects from the multiple data sources based on the similarity; and a business parameter acquisition unit, used to acquire the business parameters of the recommendation object. The recommendation data generation module is further configured to process the business parameters using parameter demand weights corresponding to the business dimension to obtain weighted dimension recommendation data, process the similarity using similarity demand weights corresponding to the similarity to obtain weighted similarity, and perform calculations on the dimension recommendation data and the weighted similarity to generate the business recommendation data corresponding to the recommendation object and the business demand.

6. An online business system, characterized by include: The information acquisition module is used to acquire business objects from multiple data sources; The user response module is used to respond to a user account's query request for a business object, determine the query object corresponding to the user account, and the business requirement corresponding to the query object. The business requirement is determined based on requirement conditions under multiple business dimensions. The user account is an enterprise user account operated by the enterprise's procurement personnel. The business requirement includes consumable requirements or durable requirements. The module receives a recommended object corresponding to the query object and the business recommendation data of the recommended object, and recommends the recommended object to the user account based on the business recommendation data. The object recommendation module is used to receive the query object and the business requirement, obtain the recommended object corresponding to the query object from business objects in multiple data sources, and the business parameters of the recommended object. The data source is an e-commerce channel. Based on the business requirement corresponding to the query object and the business parameters of the recommended object, the module generates business recommendation data corresponding to the business requirement. The request approval module is used to obtain the transaction request parameters of the user account for the recommended object; generate a transaction score corresponding to the transaction request parameters based on the transaction request parameters and the business parameters of the recommended object; determine the transaction approval strategy corresponding to the transaction request parameters in response to the comparison result of the transaction score and a preset score threshold; process the transaction request parameters according to the transaction approval strategy to generate the approval result of the transaction request parameters. The object recommendation module is further configured to obtain parameter requirement weights for each business dimension corresponding to the business requirement. The business dimensions include a business value dimension and a business evaluation dimension. Obtaining the parameter requirement weights for each business dimension includes: obtaining a first parameter weight for the business value dimension and a second parameter weight for the business evaluation dimension. When the business requirement is a consumable requirement, the first parameter weight is greater than the second parameter weight; when the business requirement is a durability requirement, the first parameter weight is less than the second parameter weight. For the business parameters of the recommended object in each business dimension, the module processes the business parameters according to the parameter requirement weights to generate dimension recommendation data for the recommended object corresponding to each business dimension. Based on the dimension recommendation data for each business dimension, the module generates the business recommendation data. The object recommendation module is further configured to match the query object with business objects from multiple data sources to determine the similarity between the query object and the business objects; determine the recommended object corresponding to the query object from the business objects from multiple data sources based on the similarity; obtain the business parameters of the recommended object; process the business parameters using parameter demand weights corresponding to the business dimensions to obtain weighted dimension recommendation data; process the similarity using similarity demand weights corresponding to the similarity to obtain weighted similarity; perform calculations on the dimension recommendation data and the weighted similarity to generate the business recommendation data corresponding to the business demand.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

8. A computer readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

9. A computer program product comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images