WO2021169659A1 - Resource information pushing method and device, server, and storage medium - Google Patents

Resource information pushing method and device, server, and storage medium Download PDF

Info

Publication number
WO2021169659A1
WO2021169659A1 PCT/CN2021/072131 CN2021072131W WO2021169659A1 WO 2021169659 A1 WO2021169659 A1 WO 2021169659A1 CN 2021072131 W CN2021072131 W CN 2021072131W WO 2021169659 A1 WO2021169659 A1 WO 2021169659A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
resource
user
priority
target
Prior art date
Application number
PCT/CN2021/072131
Other languages
French (fr)
Chinese (zh)
Inventor
王天齐
郑景飞
伍平平
马明浩
唐逸立
Original Assignee
中国银联股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国银联股份有限公司 filed Critical 中国银联股份有限公司
Publication of WO2021169659A1 publication Critical patent/WO2021169659A1/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/50Network services
    • H04L67/55Push-based network services
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241Advertisements
    • G06Q30/0251Targeted advertisements
    • G06Q30/0269Targeted advertisements based on user profile or attribute
    • G06Q30/0271Personalized advertisement
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0631Item recommendations

Definitions

  • This application belongs to the field of data processing, and in particular relates to a method, device, server, and storage medium for pushing resource information.
  • the server uniformly recommends resources to a large number of users.
  • the method of uniformly recommending resources is less accurate for users.
  • the embodiments of the present application provide a method, a device, a server, and a storage medium for pushing resource information, which can improve the accuracy of pushing resource information to users.
  • an embodiment of the present application provides a method for pushing resource information, including:
  • an embodiment of the present application provides a resource information pushing device, including:
  • the information acquisition module is used to acquire the user's resource consumption information and resource provider information
  • the priority determination module is used to determine the priority of the user and the priority of the resource information associated with the user based on the resource consumption information and the resource provider information;
  • the target determination module is configured to use a user with a priority higher than the first preset standard priority as the target user, and use resource information associated with the user with a priority higher than the second preset standard priority as the target resource information;
  • the push module is used to push target resource information associated with the target user to the target user.
  • the embodiments of the present application provide a server, including a processor, a memory, and a computer program stored on the memory and capable of running on the processor.
  • the computer program is executed by the processor to implement the technical solutions in the first aspect Resource information push method.
  • the embodiments of the application provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to realize the resource information in the technical solution of the first aspect Push method.
  • the embodiments of the present application provide a method, device, server, and storage medium for pushing resource information. Based on resource consumption information and resource provider information, the priority of a user is determined, and the priority of resource information associated with the user is determined. Push resource information associated with the user whose priority is higher than the second preset standard priority to a user whose priority is higher than the first preset standard priority.
  • the information of the resource provider represents the relevant information of the resource provider and can make up for the missing part of the resource consumption information. Combining resource consumption information and resource provider information, and using priority, the accurate push of resource information to users is realized, and the accuracy of pushing resource information to users is improved.
  • FIG. 1 is a flowchart of an embodiment of a method for pushing resource information provided by this application
  • FIG. 3 is a schematic structural diagram of an embodiment of a resource information pushing device provided by this application.
  • FIG. 4 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application.
  • FIG. 5 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application.
  • FIG. 6 is a schematic structural diagram of still another embodiment of the resource information pushing device provided by this application.
  • FIG. 7 is a schematic structural diagram of an embodiment of a server provided by this application.
  • This application provides a method, a device, a server, and a storage medium for pushing resource information, which can be applied to a scenario of pushing resource information to a user.
  • the scenarios of pushing services to users, pushing products to users, or pushing activities to users are not limited here.
  • the method for pushing resource information can be executed by a service device such as a server.
  • the number of servers is not limited here, and the number of servers can be set according to the number of users and other factors.
  • the method for pushing resource information can be executed by a server cluster.
  • the cross-comparison of resource consumption information and resource provider information can be used according to the comprehensive use of the user’s resource consumption information and resource provider information, and the priority of the user and the priority of the resource information may be determined according to the user’s priority. Users push resource information to improve the accuracy of resource information push.
  • Fig. 1 is a flowchart of an embodiment of a method for pushing resource information provided by this application. As shown in FIG. 1, the method for pushing resource information may include step S101 to step S104.
  • step S101 the user's resource consumption information and resource provider information are acquired.
  • the user's resource consumption information is related to the user, and can characterize the user's own information, and also characterize the user's orientation of resource consumption, such as the user's preference for resource consumption.
  • a user resource consumption database can be established, and the user's resource consumption information can be stored in the user resource consumption database.
  • the user resource consumption database can be set in a single big data host cluster configured with 24C or 256G or 40T, which is not limited here.
  • resource consumption information includes, but is not limited to, one or more of the following: user information, user resource consumption time preference information, user resource consumption preference information, user resource consumption type preference information, user consumption resource The type preference information of the resource provider, the location information of the resource provider where the user consumes resources, and the relevance information of the resource provider where the user consumes resources.
  • the resource provider association information of the resource consumed by the user is used to characterize the association of two or more resource providers from the source of the user's consumption of resources, for example, characterize the similarity of two or more resource providers from the source of the user's consumption of resources.
  • the resource provider may specifically be a merchant.
  • the resource consumption information can be obtained from the historical information of the transaction of the user account.
  • resource consumption information may include user account number, balance in user account, user transaction time preference, user preference for transaction selection, user selection transaction type preference, user selection transaction type preference, user selection transaction merchant location, user The selection of one or more items of information such as the similarity of the merchants for the transaction is not limited here.
  • the storage form of resource consumption information can be as shown in Table 1 below.
  • the resource provider information is related to the resource provider of the resource consumed by the user, and can characterize the relevant information of the resource provider.
  • a resource provider database can be established, and the resource provider information can be stored in the resource provider database.
  • the resource provider database can be set in a single Kafka cluster configured with 24C/256G/40T, which is not limited here.
  • the resource provider information includes one or more of the following: resource provider name information, resource provider location information, resource provider location information, resource provider type information, resource provider evaluation information, Information on the popularity of the resource provider and the relevance information provided by the resource provider’s resources.
  • the resource provision relevance information of the resource provider is used to characterize the relevance of resources provided by two or more resource providers, for example, characterize the similarity of resources provided by two or more resource providers.
  • the resource provider may specifically be a merchant.
  • the resource provider information can be read through a third-party interface, data file import, or manual input collection to obtain public or shared merchant transaction information data.
  • the resource provider information may include one or more of information such as the name of the business, the location of the business, the location of the business district, the type of the business, the evaluation of the business, the popularity of the business, and the similarity of the business consumption, which is not limited here.
  • the storage form of resource provider information can be as shown in Table 2 below.
  • data cleaning and standardization processing can be performed on the acquired resource provider information for use in subsequent processes.
  • the rules formed in the standardization process such as user location extraction rules, and dictionaries such as business name address proper nouns, stop words, etc., are imported into the rule library or thesaurus.
  • the rule database and thesaurus can be used in the process of data cleaning and standardization of new information.
  • step S102 based on the resource consumption information and the resource provider information, the priority of the user is determined, and the priority of the resource information associated with the user is determined.
  • the priority may include the first level to the Nth level, where the priority of the i-1th level is higher than the priority of the ith level, 1 ⁇ i ⁇ N , I and N are positive integers.
  • the priority may include a high-level, a medium-level, and a low-level, where the high-level priority is higher than the medium-level priority, and the medium-level priority is higher than the low-level priority.
  • the server can determine which users to prefer to push resource information according to the user's priority.
  • a user can be associated with multiple resource information, and the resource information associated with different users can be different.
  • the server can also determine which resource information to preferentially push to the user according to the priority of the resource information.
  • the resource information may include, but is not limited to, one or more of resource providers, resource types, and resource provision schemes.
  • the resource provider may be a merchant
  • the resource category may be a commodity category
  • the resource provision plan may be a commodity sales plan such as a preferential plan.
  • step S103 the user whose priority is higher than the first preset standard priority is taken as the target user, and the resource information associated with the user whose priority is higher than the second preset standard priority is taken as the target resource information.
  • the first preset standard priority can be set according to the work scenario and work requirements, which is not limited here, and the first preset standard priority can be changed according to specific conditions.
  • the second preset standard priority can be set according to the work scenario and work requirements, which is not limited here, and the second preset standard priority can be changed according to specific conditions.
  • the first preset standard priority and the second preset standard priority may be independent of each other.
  • the target user is the user who is given priority to push resource information.
  • the target resource information associated with the user is the resource information that is preferentially pushed to the user.
  • the priority of the same resource information associated with different users can be different, so that the target resource information pushed to different users can be different, which can be specifically determined according to the priority of the resource information associated with each user.
  • step S104 the target resource information associated with the target user is pushed to the target user.
  • the server can push the target resource information associated with the target user to the target user through the application program, namely APP, etc.
  • the real-time data stream is used to push the target resource information associated with the target user to the target user, and the real-time data stream may specifically be a Kafka real-time data stream.
  • the server can also push the target resource information associated with the target user to the target user through short messages and other methods.
  • the method of pushing target resource information associated with the target user to the target user is not limited here.
  • the priority of the user is determined, and the priority of the resource information associated with the user is determined.
  • the information of the resource provider represents the relevant information of the resource provider and can make up for the missing part of the resource consumption information. Combining resource consumption information and resource provider information, and using priority, the accurate push of resource information to users is realized, and the accuracy of pushing resource information to users is improved.
  • the above step S102 may be implemented through user activity information and resource consumption potential information. Specifically, based on the resource consumption information and resource provider information, the user's activity information and resource consumption potential information are set. According to the user's activity information and resource consumption potential information, the priority of the user is determined, and the priority of the resource information associated with the user is determined.
  • the user's activity information is used to characterize the user's activity, and may indicate the number of times the user has consumed resources, the amount of user resource consumption, etc., which are not limited here.
  • the user's activity information indicates that the user's activity is higher, which means that the user's transaction volume is greater, and the transaction volume may refer to the number of transactions, transaction costs, and so on.
  • the activity information of the user may be specifically implemented as an activity score or an activity rating, etc., which is not limited here.
  • the user's activity information may be reflected in the user's bank card activity.
  • the activity of a bank card may include active, inactive, recession, arousal, sleep and so on.
  • bank cards can be divided into active cards, inactive cards, declining cards, wake-up cards, sleep cards, etc. according to their activity.
  • the user's resource consumption potential information is used to characterize the user's resource consumption potential, and can indicate the user's total resource consumption potential, the user's potential to consume a certain type of resource, and so on, which is not limited here.
  • the user's resource consumption potential information indicates that the user's resource consumption potential is higher, indicating that the greater the amount of resources that the user can consume is predicted.
  • the amount of resources may refer to the total amount of resources, the amount of certain types of resources, and so on.
  • the user's resource consumption potential information can be specifically realized as a resource consumption potential score or a resource consumption potential classification, etc., which is not certain here.
  • the user's resource consumption potential can include high-level, medium-level, low-level, and so on.
  • combining the user’s resource consumption information and resource provider information, through the user’s transaction information such as the number of transactions and transaction amount in the resource consumption information and resource provider information it can be determined whether the user’s bank card is active or not. Decline, whether to sleep, that is, determine the user's activity.
  • the GBDT algorithm can also be used to determine the user's resource consumption potential through the use of resource consumption information and the user's historical information in the resource provider information, such as the use of e-commerce applications, the history of the business circle where the user is trading, and so on. According to the user's activity information and resource consumption potential information, the priority of the pushed user and the priority of the pushed resource information are determined.
  • combining the user's resource consumption information and resource provider information, the user's industry preferences, merchant consumption history and other resource consumption information, as well as the user's e-commerce application preferences, merchant attention history and other resources are provided Party information, using the Apriori algorithm to determine the merchant in the resource information pushed to the user.
  • the resource information includes resource providers.
  • a set of candidate resource providers can be generated according to the category of each resource provider.
  • the pruning algorithm is used to process the candidate set of resource providers, and the resource providers who provide users with high-frequency resources for consumption are selected as the resource providers in the resource information. The above process can be repeated to obtain multiple resource providers in the resource information.
  • resource provider information due to the participation of resource provider information, for users with low activity levels, such as users whose transaction cards are evaluated as sleeping cards, a certain amount of resource consumption of the user can also be obtained through resource provider information mining. Ability to push appropriate resource information for the user.
  • the target decision model may be used to obtain the priority of the user and/or the priority of the resource information associated with the user. Specifically, for each user, the user's resource consumption information and resource provider information are used as the input of each target decision model, and the output result of each target decision model is obtained. The output result represents the priority of the user and/or the priority of the resource information associated with the user under the target requirement corresponding to the target decision model. Calculate the sum of the product of the output result of each target decision model and the weight of the target decision model. The priority of the user and/or the priority of the resource information associated with the user is determined according to the sum of the products.
  • the target decision model may specifically be a classifier.
  • the number of target decision models can be multiple. If multiple target decision models are selected, different target decision models correspond to different target requirements.
  • the comprehensive priority of the user and/or the priority of the resource information associated with the user can be obtained according to multiple different target requirements.
  • the following is an example for description.
  • the following formula (1) can be used to obtain the priority of the integrated user and/or the priority of the resource information associated with the user.
  • x is the input user’s resource consumption information and resource provider information.
  • the user’s resource consumption information and resource provider information can be expressed as a vector;
  • F(x) is the user’s integrated priority and/or the user’s
  • the priority of the associated resource information specifically, can be divided into three priority levels: medium, high, and low;
  • h m (x) is the output result of the m-th target decision model;
  • ⁇ m is the weight of the target decision model.
  • the resource consumption result information fed back after the target user receives the target resource information can also be collected. Use the resource consumption result information to update the target decision model to improve the accuracy of the output result of the target decision model.
  • the priority of the user determined by the server and the priority of the resource information associated with the user are both predicted by the server.
  • the target decision model can be continuously updated. After the target user receives the target resource information, the target user may consume resources corresponding to the target resource information, or may not consume resources corresponding to the target resource information.
  • the resource consumption result information fed back may characterize the actual resource consumption of the target user after receiving the target resource information, and may include the priority of the actual user and/or the priority of the actual resource information associated with the user.
  • the target decision model will continuously fit the priority of the actual user and/or the priority of the actual target resource information associated with the user, and the priority of the user predicted by the target decision model and/or the target resource information associated with the user.
  • the residual of the priority makes the output result of the target decision model approach the priority of the actual user and/or the priority of the actual target resource information associated with the user.
  • the target decision model can be updated by the following formula (2).
  • F m-1 (x) is the weighted calculation sum of the output results of the previous m-1 target decision model.
  • F m (x) is the weighted calculation sum of the output results of the previous m target decision model.
  • y i is the i-th actual result, that is, the priority of the i-th user and/or the priority of the resource information associated with the i-th user.
  • L(y i , F m-1 (x i )+h(x i )) is the relational expression of y i and F m-1 (x i )+h(x i ).
  • Fig. 2 is a flowchart of another embodiment of a method for pushing resource information provided by this application.
  • the difference between FIG. 2 and FIG. 1 is that the resource information pushing method shown in FIG. 2 may further include step S105, and step S102 shown in FIG. 1 can be refined into step S1021 shown in FIG. 2.
  • step S105 the credibility calculation model is used to calculate the credibility of the resource consumption information and the credibility of the resource provider information.
  • the algorithm used in the credibility calculation model may include a random forest algorithm, or other algorithms may be used, which is not limited here.
  • the credibility of resource consumption information represents the degree to which the resource consumption information can be trusted.
  • the credibility of the resource provider's information represents the degree to which the resource provider's information can be trusted. In some examples, credibility is expressed in numerical terms.
  • a credibility of information of 80 to 100 indicates that the information is accurate; a credibility of information of 60 to 80 indicates that the information is credible and available; a credibility of information of 40 to 60 indicates that the information is partially available; the credibility of information is 0 to 40 indicate that the information needs to be improved.
  • This application can use the integrated learning method to calculate and classify the credibility.
  • Integrated learning solves a single prediction problem by building a combination of several models.
  • the following uses the random forest algorithm as an example to describe the credibility calculation, but it should be noted that in the embodiment of the present application, other algorithms may also be used to perform the credibility calculation.
  • the random forest algorithm is a sub-category of ensemble learning, which relies on the voting selection of the decision tree to determine the final classification result.
  • the following formula (4) can be used to calculate the information entropy.
  • a decision tree is used to classify the reliability.
  • Information entropy is the foundation of decision trees, which can be used to determine the basis of resource consumption information and the selection order of resource provider information during the process of credibility classification of decision trees.
  • I(x) represents the information of random variables
  • p( xi ) refers to the probability when xi occurs
  • xi is the resource consumption information and resource provider information of the i-th user.
  • a decision tree is a tree structure in which each internal node represents a test on an attribute, each branch represents a test output, and each leaf node represents a category.
  • each branch represents an output of credibility calculation, and each leaf node represents a credibility category.
  • multiple credibility calculation models may be used to perform multi-dimensional calculations on the credibility of resource consumption information and the credibility of resource provider information.
  • the credibility can be assessed from both the source of the information and the information itself.
  • the credibility of the information source can be evaluated from the aspects of the person who entered the information, whether the entry has a verification link, whether it has undergone technical processing and manual data cleaning, etc.
  • the credibility of the user behavior preference information in the resource consumption information and the resource provider information is taken as an example to illustrate the credibility evaluation of the information itself.
  • the credibility of user behavior preference information can be obtained through comprehensive evaluation of the credibility of the user's resource consumption, the credibility of the user's resource consumption method, the credibility of the user's resource consumption habits, the credibility of the user's resource consumption type information, and the weight. Spend.
  • the credibility of the information source, the credibility of the information itself, the weight of the credibility of the information source and the weight of the credibility of the information itself can be used to comprehensively obtain the credibility of the user's resource consumption information and the information of the resource provider Credibility.
  • the steps of credibility calculation and classification can be performed by a single host in a big data host cluster configured as 24C/256G/40T, which is not limited here.
  • step S1021 based on the resource consumption information and the resource provider information, according to the credibility of the resource consumption information and the credibility of the resource provider information, the priority of the user is determined, and the priority of the resource information associated with the user is determined .
  • the credibility of the resource consumption information and the credibility of the resource provider information, the resource consumption information and the resource provider information for determining the priority of the user and the priority of the resource information associated with the user can be selected. For example, selecting highly reliable resource consumption information and resource provider information to determine the priority of the user and the priority of the resource information associated with the user, so as to further improve the accuracy and reliability of pushing resource information.
  • the server may correspondingly push resource information to the user according to the user's resource consumption mode.
  • the priority of the resource information includes, but is not limited to, the resource information including one or more of the priority of the resource provider, the priority of the resource category, and the priority of the resource provision plan.
  • the mode in which the user consumes resources is an online mode.
  • the server may obtain the user's location from the user's application.
  • the server matches the information of the resource provider according to the user's transaction merchant preference and transaction merchant type preference in the resource consumption information to obtain the merchant name, merchant type, merchant popularity and other information of the merchant in the area where the user is located.
  • obtain the business name, business type, business popularity and other information of the business in the area where the user location is located and evaluate the resource consumption potential of the user in the area where the user location is located.
  • Combining the resource consumption potential and user activity in the area determine resource information with a priority higher than the second preset standard priority.
  • the resource information with priority higher than the second preset standard priority can be pushed to users through Kafka real-time data stream. For example, push resource information to the user in the user's application.
  • the relevance of users represented by resource consumption information and resource provider information and the relevance of resource providers can be used to cluster users, so that users in the same cluster obtained by clustering can be resourced. Push of information.
  • the users are clustered to obtain at least two clusters. For each cluster, select a user's resource consumption information and resource provider information as the cluster center of the cluster. Calculate the resource consumption information of other users in the cluster except the user corresponding to the cluster center and the distance between the resource provider information and the cluster center. According to the distance, update the cluster and the cluster center of the cluster until the updated cluster center is the same as the cluster center before the update.
  • the initial clustering of users can be randomly clustered or clustered according to a preset rule, which is not limited here.
  • the resource consumption information and resource provider information of each user can be represented in the form of vectors.
  • the cluster center of a cluster should be the resource consumption information and resource provider information of the user who is closest to the resource consumption information and resource provider information of other users in the cluster. Therefore, by calculating the distance between the resource consumption information and resource provider information of other users in the cluster except the user corresponding to the cluster center and the cluster center, and based on the distance, the division of the cluster and the cluster center are updated to ensure the clustering. accuracy.
  • the updated cluster center is the same as the cluster center before the update, which means that the cluster center is stable and the current clustering result can be maintained.
  • the cross-validation method can be used to select the number of clusters k.
  • dist ed (x (i) , x (j) ) is the vector formed by the resource consumption information and resource provider information of the i-th user in the cluster and the resource consumption information and resource provider information of the j-th user
  • the Euclidean distance of the vector n is the total number of users in the cluster;
  • u is the vector dimension.
  • the resource consumption of users in the same cluster has a certain similarity, and at least one item of resource information pushed to users in the same cluster is the same.
  • the resource categories in the resource information pushed to users in the same cluster are all "catering", but the transaction discount algorithm involved in the resource provision scheme in the resource information pushed to users in the same cluster different.
  • the user's resource consumption mode is the online mode.
  • the server corresponding to the online mode may include a single big data host cluster configured as 24C/256G/40T and a single Kafka cluster configured as 24C/256G/40T, which is not limited here.
  • the mode in which the user consumes resources is an offline mode.
  • a user uses a bank card to conduct a transaction, uses the transaction flow in the resource consumption information to determine the user’s business district, and uses a collaborative filtering algorithm to match the business district’s location in the resource provider’s information To determine the target resource information to be pushed, combined with information such as the name of the merchant, the type of the merchant, and the similarity of the merchant’s consumption, combined with the user’s activity.
  • one or more of the priority of the resource provider, the priority of the resource category, and the priority of the resource provision plan in the target resource information can be determined, and the priority is higher than the second preset standard priority resource provider
  • One or more of, resource category, and resource provision plan are pushed to the user as target resource information, that is, the target resource information pushed to the user is determined.
  • the target resource information can be pushed to the user through the Kafka real-time data stream.
  • the user's resource consumption mode is offline.
  • the server corresponding to the offline mode may include a single big data host cluster configured as 24C/256G/40T and a single Kafka cluster configured as 24C/256G/40T, which is not limited here.
  • FIG. 3 is a schematic structural diagram of an embodiment of the resource information pushing device provided by this application.
  • the resource information pushing device 200 may include an information acquiring module 201, a priority determining module 202, a target determining module 203, and a pushing module 204.
  • the information acquisition module 201 is used to acquire user resource consumption information and resource provider information.
  • the resource consumption information includes one or more of the following: user information, user resource consumption time preference information, user resource consumption resource provider preference information, user resource consumption type preference information, user resource consumption resource provision Party type preference information, location information of the resource provider where the user consumes resources, and information about the relevance of the resource provider where the user consumes resources.
  • the resource provider information includes one or more of the following: resource provider name information, resource provider location information, resource provider location information, resource provider type information, resource provider evaluation information, Information on the popularity of the resource provider and the relevance information provided by the resource provider’s resources.
  • the priority determination module 202 is configured to determine the priority of the user based on the resource consumption information and the resource provider information, and determine the priority of the resource information associated with the user.
  • the resource information includes one or more of resource providers, resource types, and resource provision schemes.
  • the priority of the resource information includes one or more of the resource information including the priority of the resource provider, the priority of the resource category, and the priority of the resource providing solution.
  • the target determination module 203 is configured to use a user with a priority higher than a first preset standard priority as a target user, and use resource information associated with the user with a priority higher than a second preset standard priority as the target resource information.
  • the pushing module 204 is used to push target resource information associated with the target user to the target user.
  • the pushing module 204 may be specifically configured to push the target resource information associated with the target user to the target user using real-time data streams.
  • the priority of the user is determined, and the priority of the resource information associated with the user is determined.
  • the resource provider information represents the relevant information of the resource provider and can make up for the missing part of the resource consumption information. Combining resource consumption information and resource provider information, and using priority, the accurate push of resource information to users is realized, and the accuracy of pushing resource information to users is improved.
  • the priority determination module 202 may be specifically configured to: set user activity information and resource consumption potential information based on resource consumption information and resource provider information; determine according to user activity information and resource consumption potential information The priority of the user, and the priority of determining the resource information associated with the user.
  • FIG. 4 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application.
  • the difference between FIG. 4 and FIG. 3 is that the resource information pushing device 200 shown in FIG. 4 may further include a collection module 205 and an update module 206.
  • the priority determining module 202 in the foregoing embodiment can be specifically used to: for each user, take the user's resource consumption information and resource provider information as the input of each target decision model, and obtain the output result of each target decision model; Calculate the sum of the product of the output result of each target decision model and the weight of the target decision model; determine the priority of the user and/or the priority of the resource information associated with the user according to the sum of the products.
  • the output result represents the priority of the user and/or the priority of the resource information associated with the user under the target requirements corresponding to the target decision model
  • the collection module 205 is configured to collect resource consumption result information fed back by the target user after receiving the target resource information.
  • the update module 206 is used to update the target decision model using the resource consumption result information.
  • FIG. 5 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application.
  • the difference between FIG. 5 and FIG. 3 is that the resource information pushing device 200 shown in FIG. 5 may further include a credibility calculation module 207.
  • the credibility calculation module 207 can be used to use the credibility calculation model to calculate the credibility of resource consumption information and the credibility of resource provider information.
  • the priority determining module 202 in the foregoing embodiment may be specifically used to determine the priority of the user based on the resource consumption information and the resource provider information, according to the credibility of the resource consumption information and the credibility of the resource provider information, and Determine the priority of the resource information associated with the user.
  • FIG. 6 is a schematic structural diagram of still another embodiment of the resource information pushing device provided by this application.
  • the difference between FIG. 6 and FIG. 3 is that the resource information pushing device 200 shown in FIG. 6 may further include a clustering module 208.
  • the clustering module 208 can be used to: cluster users to obtain at least two clusters; for each cluster, select the resource consumption information and resource provider information of a user as the cluster center; The resource consumption information of users other than the user and the distance between the resource provider information and the cluster center; according to the distance, the cluster and the cluster center of the cluster are updated until the updated cluster center is the same as the cluster center before the update.
  • the resource information includes one or more of a resource provider, a resource category, and a resource providing solution, and at least one of the resource information pushed to users in the same cluster is the same.
  • FIG. 7 is a schematic structural diagram of an embodiment of a server provided by this application.
  • the server 300 includes a memory 301, a processor 302, and a computer program stored on the memory 301 and running on the processor 302.
  • the aforementioned processor 302 may include a central processing unit (CPU), or a specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
  • CPU central processing unit
  • ASIC specific integrated circuit
  • the memory 301 may include a large-capacity memory for data or instructions.
  • the memory 301 may include an HDD, a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these.
  • the storage 301 may include removable or non-removable (or fixed) media. Under appropriate circumstances, the storage 301 can be opened inside or outside the server 300 in a terminal hotspot.
  • the memory 301 is a non-volatile solid state memory.
  • the memory 301 includes read-only memory (ROM).
  • the ROM can be mask-programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically rewritable ROM (EAROM) or flash memory or A combination of two or more of these.
  • the processor 302 runs the computer program corresponding to the executable program code by reading the executable program code stored in the memory 301, so as to implement the resource information pushing method in the foregoing embodiment.
  • the server 300 may further include a communication interface 303 and a bus 304. Among them, as shown in FIG. 7, the memory 301, the processor 302, and the communication interface 303 are connected through the bus 304 and complete the communication with each other.
  • the communication interface 303 is mainly used to implement communication between various modules, devices, units and/or devices in the embodiments of the present application.
  • the input device and/or output device can also be accessed through the communication interface 303.
  • the bus 304 includes hardware, software, or both, and couples the components of the server 300 to each other.
  • the bus 304 may include an accelerated graphics port (AGP) or other graphics bus, an enhanced industry standard architecture (EISA) bus, a front side bus (FSB), a hypertransport (HT) interconnect, an industry standard architecture (ISA) ) Bus, unlimited bandwidth interconnect, low pin count (LPC) bus, memory bus, microchannel architecture (MCA) bus, peripheral component interconnect (PCI) bus, PCI-Express (PCI-X) bus, serial advanced Technical Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of these.
  • the bus 304 may include one or more buses.
  • An embodiment of the present application also provides a computer-readable storage medium having a computer program stored on the computer-readable storage medium.
  • the computer program When the computer program is executed by a processor, the method for pushing resource information in the above-mentioned embodiment can be implemented, and can achieve The same technical effect, in order to avoid repetition, will not be repeated here.
  • the computer-readable storage medium may include non-transitory computer-readable storage media, such as read-only memory (Read-Only Memory, ROM for short), Random Access Memory (RAM for short), and magnetic disks. Or CD etc.
  • each block in the flowchart and/or block diagram and the combination of each block in the flowchart and/or block diagram can be implemented by a program or instruction.
  • These programs or instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to generate a machine so that these programs or instructions are executed by the processor of the computer or other programmable data processing device Enables the realization of functions/actions specified in one or more blocks of the flowchart and/or block diagram.
  • Such a processor can be, but is not limited to, a general-purpose processor, a dedicated processor, a special application processor, or a field programmable logic circuit. It can also be understood that each block in the block diagram and/or flowchart and the combination of the blocks in the block diagram and/or flowchart can also be implemented by dedicated hardware that performs specified functions or actions, or can be implemented by dedicated hardware and A combination of computer instructions.

Landscapes

  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Computer And Data Communications (AREA)

Abstract

The present application relates to the field of data processing, and provides a resource information pushing method and device, a server, and a storage medium. The method comprises: acquiring resource consumption information and resource provider information of users; determining priority levels of the users on the basis of the resource consumption information and the resource provider information, and determining priority levels of resource information associated with the users; treating a user having a priority level higher than a first preset standard priority level as a target user, and treating the resource information associated with the user and having a priority level higher than a second preset standard priority level as target resource information; and pushing, to the target user, the target resource information associated with the target user. The technical solution of the present application allows resource information to be pushed to users in a more accurate manner.

Description

资源信息推送方法、装置、服务器及存储介质Resource information pushing method, device, server and storage medium
相关申请的交叉引用Cross-references to related applications
本申请要求享有于2020年02月24日提交的名称为“资源信息推送方法、装置、服务器及存储介质”的中国专利申请202010113562.2的优先权,该申请的全部内容通过引用并入本文中。This application claims the priority of the Chinese patent application 202010113562.2 entitled "Resource Information Push Method, Device, Server and Storage Medium" filed on February 24, 2020, and the entire content of this application is incorporated herein by reference.
技术领域Technical field
本申请属于数据处理领域,尤其涉及一种资源信息推送方法、装置、服务器及存储介质。This application belongs to the field of data processing, and in particular relates to a method, device, server, and storage medium for pushing resource information.
背景技术Background technique
近年来,随着电子信息技术的发展,电子信息技术应用于越来越多的领域中,如交易领域、支付领域等等。In recent years, with the development of electronic information technology, electronic information technology has been applied in more and more fields, such as transaction fields, payment fields, and so on.
为了便于用户进行交易,可向用户推送各类资源。目前,服务器向大量用户统一推荐资源。但用户之间具有差异,统一推荐资源的方式对用户来说,资源推送的精确度较低。In order to facilitate users to conduct transactions, various resources can be pushed to users. Currently, the server uniformly recommends resources to a large number of users. However, there are differences between users, and the method of uniformly recommending resources is less accurate for users.
发明内容Summary of the invention
本申请实施例提供了一种资源信息推送方法、装置、服务器及存储介质,能够提高向用户推送资源信息的精确度。The embodiments of the present application provide a method, a device, a server, and a storage medium for pushing resource information, which can improve the accuracy of pushing resource information to users.
第一方面,本申请实施例提供一种资源信息推送方法,包括:In the first aspect, an embodiment of the present application provides a method for pushing resource information, including:
获取用户的资源消耗信息和资源提供方信息;Obtain user resource consumption information and resource provider information;
基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级;Based on resource consumption information and resource provider information, determine the priority of the user, and determine the priority of the resource information associated with the user;
将优先级高于第一预设标准优先级的用户作为目标用户,将与用户关联的优先级高于第二预设标准优先级的资源信息作为目标资源信息;Taking a user whose priority is higher than the first preset standard priority as the target user, and taking the resource information associated with the user whose priority is higher than the second preset standard priority as the target resource information;
向目标用户推送与目标用户关联的目标资源信息。Push the target resource information associated with the target user to the target user.
第二方面,本申请实施例提供一种资源信息推送装置,包括:In the second aspect, an embodiment of the present application provides a resource information pushing device, including:
信息获取模块,用于获取用户的资源消耗信息和资源提供方信息;The information acquisition module is used to acquire the user's resource consumption information and resource provider information;
优先级确定模块,用于基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级;The priority determination module is used to determine the priority of the user and the priority of the resource information associated with the user based on the resource consumption information and the resource provider information;
目标确定模块,用于将优先级高于第一预设标准优先级的用户作为目标用户,将与用户关联的优先级高于第二预设标准优先级的资源信息作为目标资源信息;The target determination module is configured to use a user with a priority higher than the first preset standard priority as the target user, and use resource information associated with the user with a priority higher than the second preset standard priority as the target resource information;
推送模块,用于向目标用户推送与目标用户关联的目标资源信息。The push module is used to push target resource information associated with the target user to the target user.
第三方面,本申请实施例提供一种服务器,包括处理器、存储器及存储在存储器上并可在处理器上运行的计算机程序,计算机程序被处理器执行时实现第一方面的技术方案中的资源信息推送方法。In the third aspect, the embodiments of the present application provide a server, including a processor, a memory, and a computer program stored on the memory and capable of running on the processor. The computer program is executed by the processor to implement the technical solutions in the first aspect Resource information push method.
第四方面,本申请实施例提供本申请实施例提供一种计算机可读存储介质,计算机可读存储介质上存储计算机程序,计算机程序被处理器执行时实现第一方面的技术方案中的资源信息推送方法。In the fourth aspect, the embodiments of the application provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to realize the resource information in the technical solution of the first aspect Push method.
本申请实施例提供一种资源信息推送方法、装置、服务器及存储介质,基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。向优先级高于第一预设标准优先级的用户推送与该用户关联的优先级高于第二预设标准优先级的资源信息。资源提供方信息表征资源提供方的相关信息,可弥补资源消耗信息所缺失的部分。结合资源消耗信息和资源提供方信息,利用优先级,实现了资源信息向用户的精确推送,提高了向用户推送资源信息的精确度。The embodiments of the present application provide a method, device, server, and storage medium for pushing resource information. Based on resource consumption information and resource provider information, the priority of a user is determined, and the priority of resource information associated with the user is determined. Push resource information associated with the user whose priority is higher than the second preset standard priority to a user whose priority is higher than the first preset standard priority. The information of the resource provider represents the relevant information of the resource provider and can make up for the missing part of the resource consumption information. Combining resource consumption information and resource provider information, and using priority, the accurate push of resource information to users is realized, and the accuracy of pushing resource information to users is improved.
附图说明Description of the drawings
从下面结合附图对本申请的具体实施方式的描述中可以更好地理解本申请。其中,相同或相似的附图标记表示相同或相似的特征。The present application can be better understood from the following description of the specific implementation manners of the present application in conjunction with the accompanying drawings. Wherein, the same or similar reference signs indicate the same or similar features.
图1为本申请提供的资源信息推送方法的一实施例的流程图;FIG. 1 is a flowchart of an embodiment of a method for pushing resource information provided by this application;
图2为本申请提供的资源信息推送方法的另一实施例的流程图;2 is a flowchart of another embodiment of the resource information pushing method provided by this application;
图3为本申请提供的资源信息推送装置的一实施例的结构示意图;FIG. 3 is a schematic structural diagram of an embodiment of a resource information pushing device provided by this application;
图4为本申请提供的资源信息推送装置的另一实施例的结构示意图;4 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application;
图5为本申请提供的资源信息推送装置的又一实施例的结构示意图;FIG. 5 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application;
图6为本申请提供的资源信息推送装置的再一实施例的结构示意图;FIG. 6 is a schematic structural diagram of still another embodiment of the resource information pushing device provided by this application;
图7为本申请提供的服务器的一实施例的结构示意图。FIG. 7 is a schematic structural diagram of an embodiment of a server provided by this application.
具体实施方式Detailed ways
下面将详细描述本申请的各个方面的特征和示例性实施例。在下面的详细描述中,提出了许多具体细节,以便提供对本申请的全面理解。但是,对于本领域技术人员来说很明显的是,本申请可以在不需要这些具体细节中的一些细节的情况下实施。下面对实施例的描述仅仅是为了通过示出本申请的示例来提供对本申请的更好的理解。本申请决不限于下面所提出的任何具体配置和算法,而是在不脱离本申请的精神的前提下覆盖了元素、部件和算法的任何修改、替换和改进。在附图和下面的描述中,没有示出公知的结构和技术,以便避免对本申请造成不必要的模糊。The features and exemplary embodiments of various aspects of the present application will be described in detail below. In the following detailed description, many specific details are proposed in order to provide a comprehensive understanding of the application. However, it is obvious to those skilled in the art that this application can be implemented without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by showing examples of the present application. This application is by no means limited to any specific configurations and algorithms proposed below, but covers any modification, replacement and improvement of elements, components and algorithms without departing from the spirit of this application. In the drawings and the following description, well-known structures and technologies are not shown in order to avoid unnecessary obscurity of the application.
本申请提供一种资源信息推送方法、装置、服务器及存储介质,可应用于对用户进行资源信息推送的场景中。例如,向用户推送服务,向用户推送商品,或向用户推送活动的场景,在此并不限定。该资源信息推送方法可由业务设备如服务器等执行,在此并不限定服务器的数目,服务器的数目可根据用户的数目及其他因素设置。例如,该资源信息推送方法可由服务器集群执行。在本申请实施例中,可根据用户的资源消耗信息和资源提供方信息的综合使用,利用资源消耗信息和资源提供方信息的交叉对比,根据确定的用户的优先级以及资源信息的优先级向用户推送资源信息,以提高资源信息的推送的精确度。This application provides a method, a device, a server, and a storage medium for pushing resource information, which can be applied to a scenario of pushing resource information to a user. For example, the scenarios of pushing services to users, pushing products to users, or pushing activities to users are not limited here. The method for pushing resource information can be executed by a service device such as a server. The number of servers is not limited here, and the number of servers can be set according to the number of users and other factors. For example, the method for pushing resource information can be executed by a server cluster. In the embodiments of this application, the cross-comparison of resource consumption information and resource provider information can be used according to the comprehensive use of the user’s resource consumption information and resource provider information, and the priority of the user and the priority of the resource information may be determined according to the user’s priority. Users push resource information to improve the accuracy of resource information push.
图1为本申请提供的资源信息推送方法的一实施例的流程图。如图1所示,该资源信息推送方法可包括步骤S101至步骤S104。Fig. 1 is a flowchart of an embodiment of a method for pushing resource information provided by this application. As shown in FIG. 1, the method for pushing resource information may include step S101 to step S104.
在步骤S101中,获取用户的资源消耗信息和资源提供方信息。In step S101, the user's resource consumption information and resource provider information are acquired.
用户的资源消耗信息与用户相关,可表征用户自身的信息,也可表征用户对资源消耗的指向,例如用户对资源消耗的偏好。与用户的资源消耗信息对应,可建立用户资源消耗数据库,将用户的资源消耗信息存储于该 用户资源消耗数据库内。在一些示例中,用户资源消耗数据库可设置于单台配置为24C或256G或40T的大数据主机集群中,在此并不限定。The user's resource consumption information is related to the user, and can characterize the user's own information, and also characterize the user's orientation of resource consumption, such as the user's preference for resource consumption. Corresponding to the user's resource consumption information, a user resource consumption database can be established, and the user's resource consumption information can be stored in the user resource consumption database. In some examples, the user resource consumption database can be set in a single big data host cluster configured with 24C or 256G or 40T, which is not limited here.
在一些示例中,资源消耗信息包括但不限于以下的一项或多项:用户信息、用户消耗资源时间偏好信息、用户消耗资源的资源提供方偏好信息、用户消耗资源类型偏好信息、用户消耗资源的资源提供方类型偏好信息、用户消耗资源的资源提供方位置信息、用户消耗资源的资源提供方关联性信息。其中,用户消耗资源的资源提供方关联性信息用于表征用户消耗资源来源的两个以上的资源提供方的关联性,例如,表征用户消耗资源来源的两个以上的资源提供方的相似度。In some examples, resource consumption information includes, but is not limited to, one or more of the following: user information, user resource consumption time preference information, user resource consumption preference information, user resource consumption type preference information, user consumption resource The type preference information of the resource provider, the location information of the resource provider where the user consumes resources, and the relevance information of the resource provider where the user consumes resources. Wherein, the resource provider association information of the resource consumed by the user is used to characterize the association of two or more resource providers from the source of the user's consumption of resources, for example, characterize the similarity of two or more resource providers from the source of the user's consumption of resources.
在交易场景中,资源提供方具体可为商户。资源消耗信息可通过用户账号的交易的历史信息得到。例如,资源消耗信息可包括用户账号、用户账号内余额、用户交易时间偏好、用户选择交易的商户偏好、用户选择的交易类型偏好、用户选择交易的商户类型偏好、用户选择交易的商户位置、用户选择交易的商户的相似度等信息中的一项或多项,在此并不限定。In the transaction scenario, the resource provider may specifically be a merchant. The resource consumption information can be obtained from the historical information of the transaction of the user account. For example, resource consumption information may include user account number, balance in user account, user transaction time preference, user preference for transaction selection, user selection transaction type preference, user selection transaction type preference, user selection transaction merchant location, user The selection of one or more items of information such as the similarity of the merchants for the transaction is not limited here.
例如,在交易场景中,资源消耗信息的存储形式可如下表一所示。For example, in a transaction scenario, the storage form of resource consumption information can be as shown in Table 1 below.
表一Table I
Figure PCTCN2021072131-appb-000001
Figure PCTCN2021072131-appb-000001
资源提供方信息与用户消耗的资源的资源提供方相关,可表征资源提供方的相关信息。与用户的资源提供方信息对应,可建立资源提供方数据库,将资源提供方信息存储于该资源提供方数据库内。在一些示例中,资源提供方数据库可设置于单台配置为24C/256G/40T的Kafka集群中,在此并不限定。The resource provider information is related to the resource provider of the resource consumed by the user, and can characterize the relevant information of the resource provider. Corresponding to the user's resource provider information, a resource provider database can be established, and the resource provider information can be stored in the resource provider database. In some examples, the resource provider database can be set in a single Kafka cluster configured with 24C/256G/40T, which is not limited here.
在一些示例中,资源提供方信息包括以下的一项或多项:资源提供方 名称信息、资源提供方位置信息、资源提供方所属区域位置信息、资源提供方类型信息、资源提供方评价信息、资源提供方热度信息、资源提供方的资源提供关联性信息。其中,资源提供方的资源提供关联性信息用于表征两个以上的资源提供方提供资源的关联性,例如,表征两个以上的资源提供方提供资源的相似度。In some examples, the resource provider information includes one or more of the following: resource provider name information, resource provider location information, resource provider location information, resource provider type information, resource provider evaluation information, Information on the popularity of the resource provider and the relevance information provided by the resource provider’s resources. Among them, the resource provision relevance information of the resource provider is used to characterize the relevance of resources provided by two or more resource providers, for example, characterize the similarity of resources provided by two or more resource providers.
在交易场景中,资源提供方具体可为商户。资源提供方信息可通过第三方接口读取、数据文件导入或人工录入采集,以得到公开或共享的商户交易信息数据。例如,资源提供方信息可包括商户名称,商户地点、商圈位置、商户类型、商户评价、商户热度、商户消费相似度等信息中的一项或多项,在此并不限定。In the transaction scenario, the resource provider may specifically be a merchant. The resource provider information can be read through a third-party interface, data file import, or manual input collection to obtain public or shared merchant transaction information data. For example, the resource provider information may include one or more of information such as the name of the business, the location of the business, the location of the business district, the type of the business, the evaluation of the business, the popularity of the business, and the similarity of the business consumption, which is not limited here.
例如,在交易场景中,资源提供方信息的存储形式可如下表二所示。For example, in a transaction scenario, the storage form of resource provider information can be as shown in Table 2 below.
表二Table II
Figure PCTCN2021072131-appb-000002
Figure PCTCN2021072131-appb-000002
为了规范资源提供方信息,以及提高规范资源提供方信息的准确性,可对获取的资源提供方信息进行数据清洗和标准化处理,以便于在后续过程中使用。在数据清洗和标准化处理的过程中,将标准化处理过程中形成的规则如用户位置提取规则等和词典如商户名称地址专有名词、停用词等导入规则库或词库。规则库和词库可以在对新的信息进行数据清洗和标准化处理的过程中使用。In order to standardize the resource provider information and improve the accuracy of the standardized resource provider information, data cleaning and standardization processing can be performed on the acquired resource provider information for use in subsequent processes. In the process of data cleaning and standardization, the rules formed in the standardization process, such as user location extraction rules, and dictionaries such as business name address proper nouns, stop words, etc., are imported into the rule library or thesaurus. The rule database and thesaurus can be used in the process of data cleaning and standardization of new information.
在步骤S102中,基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。In step S102, based on the resource consumption information and the resource provider information, the priority of the user is determined, and the priority of the resource information associated with the user is determined.
结合用户的资源消耗信息和资源提供方信息,利用算法或策略等可确定该用户的资源消耗是否活跃,也可确定该用户的资源消耗潜力,从而估 算用户对资源的需求,进而可确定用户的优先级以及与用户关联的资源的优先级。其中,算法或策略在此并不限定,例如,可采用梯度提升决策树(Gradient Boosting Decision Tree,GBDT)算法、关联规则算法即Apriori算法、协同过滤算法等。Combining the user’s resource consumption information and resource provider information, algorithms or strategies can be used to determine whether the user’s resource consumption is active, and the user’s resource consumption potential can also be determined, so as to estimate the user’s demand for resources, and then determine the user’s resource consumption. Priority and the priority of the resource associated with the user. The algorithm or strategy is not limited here. For example, a gradient boosting decision tree (Gradient Boosting Decision Tree, GBDT) algorithm, an association rule algorithm that is Apriori algorithm, a collaborative filtering algorithm, etc. can be used.
优先级的具体表现方式在此并不限定,例如,优先级可包括第一级至第N级,其中,第i-1级的优先级高于第i级的优先级,1<i≤N,i与N为正整数。又例如,优先级可包括高级、中级和低级,其中,高级的优先级高于中级的优先级,中级的优先级高于低级的优先级。The specific expression of the priority is not limited here. For example, the priority may include the first level to the Nth level, where the priority of the i-1th level is higher than the priority of the ith level, 1<i≤N , I and N are positive integers. For another example, the priority may include a high-level, a medium-level, and a low-level, where the high-level priority is higher than the medium-level priority, and the medium-level priority is higher than the low-level priority.
服务器可根据用户的优先级判断优向哪些用户推送资源信息。一个用户可关联有多个资源信息,不同用户关联的资源信息可不同。服务器也可根据资源信息的优先级判断优先向用户推送哪些资源信息。The server can determine which users to prefer to push resource information according to the user's priority. A user can be associated with multiple resource information, and the resource information associated with different users can be different. The server can also determine which resource information to preferentially push to the user according to the priority of the resource information.
在一些示例中,资源信息可包括但不限于资源提供方、资源类别、资源提供方案中的一项或多项。例如,在交易场景中,资源提供方可为商户,资源类别可为商品类别,资源提供方案可为商品销售方案如优惠方案等。In some examples, the resource information may include, but is not limited to, one or more of resource providers, resource types, and resource provision schemes. For example, in a transaction scenario, the resource provider may be a merchant, the resource category may be a commodity category, and the resource provision plan may be a commodity sales plan such as a preferential plan.
在步骤S103中,将优先级高于第一预设标准优先级的用户作为目标用户,将与用户关联的优先级高于第二预设标准优先级的资源信息作为目标资源信息。In step S103, the user whose priority is higher than the first preset standard priority is taken as the target user, and the resource information associated with the user whose priority is higher than the second preset standard priority is taken as the target resource information.
其中,第一预设标准优先级可根据工作场景和工作需求设定,在此并不限定,第一预设标准优先级可根据具体情况进行更改。第二预设标准优先级可根据工作场景和工作需求设定,在此并不限定,第二预设标准优先级可根据具体情况进行更改。第一预设标准优先级与第二预设标准优先级可相互独立。Among them, the first preset standard priority can be set according to the work scenario and work requirements, which is not limited here, and the first preset standard priority can be changed according to specific conditions. The second preset standard priority can be set according to the work scenario and work requirements, which is not limited here, and the second preset standard priority can be changed according to specific conditions. The first preset standard priority and the second preset standard priority may be independent of each other.
目标用户即为优先被推送资源信息的用户。与用户关联的目标资源信息即为优先向该用户推送的资源信息。不同用户关联的相同的资源信息的优先级可不同,使得向不同用户推送的目标资源信息可不同,具体可根据每个用户关联的资源信息的优先级确定。The target user is the user who is given priority to push resource information. The target resource information associated with the user is the resource information that is preferentially pushed to the user. The priority of the same resource information associated with different users can be different, so that the target resource information pushed to different users can be different, which can be specifically determined according to the priority of the resource information associated with each user.
在步骤S104中,向目标用户推送与目标用户关联的目标资源信息。In step S104, the target resource information associated with the target user is pushed to the target user.
其中,服务器可通过应用程序即APP等方式向目标用户推送与目标用 户关联的目标资源信息。例如,利用实时数据流向目标用户推送与目标用户关联的目标资源信息,实时数据流具体可为Kafka实时数据流。服务器也可通过短信息等方式向目标用户推送与目标用户关联的目标资源信息。在此并不限定向目标用户推送与目标用户关联的目标资源信息的方式。Among them, the server can push the target resource information associated with the target user to the target user through the application program, namely APP, etc. For example, the real-time data stream is used to push the target resource information associated with the target user to the target user, and the real-time data stream may specifically be a Kafka real-time data stream. The server can also push the target resource information associated with the target user to the target user through short messages and other methods. The method of pushing target resource information associated with the target user to the target user is not limited here.
在本申请实施例中,基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。向优先级高于第一预设标准优先级的用户推送与该用户关联的优先级高于第二预设标准优先级的资源信息。资源提供方信息表征资源提供方的相关信息,可弥补资源消耗信息所缺失的部分。结合资源消耗信息和资源提供方信息,利用优先级,实现了资源信息向用户的精确推送,提高了向用户推送资源信息的精确度。In the embodiment of the present application, based on the resource consumption information and the resource provider information, the priority of the user is determined, and the priority of the resource information associated with the user is determined. Push resource information associated with the user whose priority is higher than the second preset standard priority to a user whose priority is higher than the first preset standard priority. The information of the resource provider represents the relevant information of the resource provider and can make up for the missing part of the resource consumption information. Combining resource consumption information and resource provider information, and using priority, the accurate push of resource information to users is realized, and the accuracy of pushing resource information to users is improved.
在一些示例中,上述步骤S102可通过用户的活跃度信息和资源消耗潜力信息实现。具体地,基于资源消耗信息和资源提供方信息,设置用户的活跃度信息和资源消耗潜力信息。根据用户的活跃度信息和资源消耗潜力信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。In some examples, the above step S102 may be implemented through user activity information and resource consumption potential information. Specifically, based on the resource consumption information and resource provider information, the user's activity information and resource consumption potential information are set. According to the user's activity information and resource consumption potential information, the priority of the user is determined, and the priority of the resource information associated with the user is determined.
用户的活跃度信息用于表征用户的活跃度,可指示用户资源消耗次数的多少,用户资源消耗量等,在此并不限定。例如,在交易场景中,用户的活跃度信息表征用户的活跃度越高,表示用户的交易量越大,交易量可指交易次数、交易花费等。用户的活跃度信息具体可实现为活跃度评分或活跃度分级等,在此并不限定。例如,在交易场景中,用户的活跃度信息可体现为用户的银行卡的活跃度。银行卡的活跃度可包括活跃、非活跃、衰退、唤醒、睡眠等。对应地,银行卡可按照活跃度划分为活跃卡、非活跃卡、衰退卡、唤醒卡、睡眠卡等。The user's activity information is used to characterize the user's activity, and may indicate the number of times the user has consumed resources, the amount of user resource consumption, etc., which are not limited here. For example, in a transaction scenario, the user's activity information indicates that the user's activity is higher, which means that the user's transaction volume is greater, and the transaction volume may refer to the number of transactions, transaction costs, and so on. The activity information of the user may be specifically implemented as an activity score or an activity rating, etc., which is not limited here. For example, in a transaction scenario, the user's activity information may be reflected in the user's bank card activity. The activity of a bank card may include active, inactive, recession, arousal, sleep and so on. Correspondingly, bank cards can be divided into active cards, inactive cards, declining cards, wake-up cards, sleep cards, etc. according to their activity.
用户的资源消耗潜力信息用于表征用户对资源的消耗潜力,可指示用户消耗资源总量的潜力,用户消耗某类资源的潜力等,在此并不限定。例如,在交易场景中,用户的资源消耗潜力信息表征用户的资源消耗潜力越高,表示预测该用户能够消耗的资源量越大,资源量可指资源总量、某类资源的量等。用户的资源消耗潜力信息具体可实现为资源消耗潜力评分或资源消耗潜力分级等,在此并不确定。例如,在交易场景中,用户的资源 消耗潜力可包括高级、中级、低级等。The user's resource consumption potential information is used to characterize the user's resource consumption potential, and can indicate the user's total resource consumption potential, the user's potential to consume a certain type of resource, and so on, which is not limited here. For example, in a transaction scenario, the user's resource consumption potential information indicates that the user's resource consumption potential is higher, indicating that the greater the amount of resources that the user can consume is predicted. The amount of resources may refer to the total amount of resources, the amount of certain types of resources, and so on. The user's resource consumption potential information can be specifically realized as a resource consumption potential score or a resource consumption potential classification, etc., which is not certain here. For example, in a transaction scenario, the user's resource consumption potential can include high-level, medium-level, low-level, and so on.
例如,在交易场景中,结合用户的资源消耗信息和资源提供方信息,通过资源消耗信息和资源提供方信息中用户的交易信息如交易次数、交易金额等,可判断用户银行卡是否活跃、是否衰退、是否睡眠,即确定用户的活跃度。还可通过资源消耗信息和资源提供方信息中用户的历史信息如电商应用程序的使用、用户交易所在的商圈历史等,利用GBDT算法确定用户的资源消耗潜力。根据用户的活跃度信息和资源消耗潜力信息,确定被推送用户的优先级,以及推送的资源信息的优先级。For example, in a transaction scenario, combining the user’s resource consumption information and resource provider information, through the user’s transaction information such as the number of transactions and transaction amount in the resource consumption information and resource provider information, it can be determined whether the user’s bank card is active or not. Decline, whether to sleep, that is, determine the user's activity. The GBDT algorithm can also be used to determine the user's resource consumption potential through the use of resource consumption information and the user's historical information in the resource provider information, such as the use of e-commerce applications, the history of the business circle where the user is trading, and so on. According to the user's activity information and resource consumption potential information, the priority of the pushed user and the priority of the pushed resource information are determined.
又例如,在交易场景中,结合用户的资源消耗信息和资源提供方信息,通过用户的行业偏好、商户消费历史等资源消耗信息,以及用户电商应用程序的使用偏好、商户关注历史等资源提供方信息,利用Apriori算法确定向用户推送的资源信息中的商户。For another example, in a transaction scenario, combining the user's resource consumption information and resource provider information, the user's industry preferences, merchant consumption history and other resource consumption information, as well as the user's e-commerce application preferences, merchant attention history and other resources are provided Party information, using the Apriori algorithm to determine the merchant in the resource information pushed to the user.
在一些示例中,资源信息包括资源提供方。可根据各资源提供方的类别,生成资源提供方备选集合。利用剪枝算法对资源提供方备选集合进行处理,选取高频为用户提供可供消耗资源的资源提供方作为资源信息中的资源提供方。可循环进行上述过程,得到资源信息中的多个资源提供方。In some examples, the resource information includes resource providers. A set of candidate resource providers can be generated according to the category of each resource provider. The pruning algorithm is used to process the candidate set of resource providers, and the resource providers who provide users with high-frequency resources for consumption are selected as the resource providers in the resource information. The above process can be repeated to obtain multiple resource providers in the resource information.
在本申请实施例中,由于有资源提供方信息的参与,对于活跃度较低的用户如交易卡评估为睡眠卡的用户来说,也可通过资源提供方信息挖掘得到该用户一定的资源消耗能力,从而为该用户推送合适的资源信息。In the embodiments of this application, due to the participation of resource provider information, for users with low activity levels, such as users whose transaction cards are evaluated as sleeping cards, a certain amount of resource consumption of the user can also be obtained through resource provider information mining. Ability to push appropriate resource information for the user.
在一些示例中,可利用目标决策模型来得到用户的优先级和/或用户关联的资源信息的优先级。具体地,对于每个用户,将用户的资源消耗信息和资源提供方信息作为每个目标决策模型的输入,得到每个目标决策模型的输出结果。输出结果表征在目标决策模型对应的目标要求下,用户的优先级和/或与用户关联的资源信息的优先级。计算每个目标决策模型的输出结果与目标决策模型的权重的乘积之和。根据乘积之和确定用户的优先级和/或与用户关联的资源信息的优先级。In some examples, the target decision model may be used to obtain the priority of the user and/or the priority of the resource information associated with the user. Specifically, for each user, the user's resource consumption information and resource provider information are used as the input of each target decision model, and the output result of each target decision model is obtained. The output result represents the priority of the user and/or the priority of the resource information associated with the user under the target requirement corresponding to the target decision model. Calculate the sum of the product of the output result of each target decision model and the weight of the target decision model. The priority of the user and/or the priority of the resource information associated with the user is determined according to the sum of the products.
目标决策模型具体可为分类器。目标决策模型的数目可以为多个。若选用多个目标决策模型,不同目标决策模型对应的目标要求不同。在本实施例中,可根据多个不同的目标要求,得到用户综合的优先级和/或与用户 关联的资源信息的优先级。The target decision model may specifically be a classifier. The number of target decision models can be multiple. If multiple target decision models are selected, different target decision models correspond to different target requirements. In this embodiment, the comprehensive priority of the user and/or the priority of the resource information associated with the user can be obtained according to multiple different target requirements.
下面以一示例进行说明。可利用以下的算式(1)得到综合的用户的优先级和/或与用户关联的资源信息的优先级。The following is an example for description. The following formula (1) can be used to obtain the priority of the integrated user and/or the priority of the resource information associated with the user.
Figure PCTCN2021072131-appb-000003
Figure PCTCN2021072131-appb-000003
其中,x为输入的用户的资源消耗信息和资源提供方信息,具体地,用户的资源消耗信息和资源提供方信息可表示为向量;F(x)为用户综合的优先级和/或与用户关联的资源信息的优先级,具体地,可分为中、高、低三种优先级;h m(x)为第m个目标决策模型的输出结果;γ m是目标决策模型的权重。 Among them, x is the input user’s resource consumption information and resource provider information. Specifically, the user’s resource consumption information and resource provider information can be expressed as a vector; F(x) is the user’s integrated priority and/or the user’s The priority of the associated resource information, specifically, can be divided into three priority levels: medium, high, and low; h m (x) is the output result of the m-th target decision model; γ m is the weight of the target decision model.
进一步地,还可采集目标用户接收到目标资源信息后反馈的资源消耗结果信息。利用资源消耗结果信息更新目标决策模型,以提高目标决策模型的输出结果的准确性。Further, the resource consumption result information fed back after the target user receives the target resource information can also be collected. Use the resource consumption result information to update the target decision model to improve the accuracy of the output result of the target decision model.
服务器确定的用户的优先级,以及与用户关联的资源信息的优先级,均为服务器的预测结果。为了能够进一步提高服务器预测的精确度,可对目标决策模型不断进行更新。目标用户接收到目标资源信息后,目标用户可能会进行与目标资源信息对应的资源消耗,也可能不会进行与目标资源信息对应的资源消耗。反馈的资源消耗结果信息可表征目标用户接收到目标资源信息后的实际资源消耗,可包括实际的用户的优先级和/或实际的与用户关联的资源信息的优先级。目标决策模型会不断拟合实际的用户的优先级和/或实际的与用户关联的目标资源信息的优先级,与目标决策模型预测的用户的优先级和/或与用户关联的目标资源信息的优先级的残差,使得目标决策模型的输出结果逼近实际的用户的优先级和/或实际的与用户关联的目标资源信息的优先级。The priority of the user determined by the server and the priority of the resource information associated with the user are both predicted by the server. In order to further improve the accuracy of server prediction, the target decision model can be continuously updated. After the target user receives the target resource information, the target user may consume resources corresponding to the target resource information, or may not consume resources corresponding to the target resource information. The resource consumption result information fed back may characterize the actual resource consumption of the target user after receiving the target resource information, and may include the priority of the actual user and/or the priority of the actual resource information associated with the user. The target decision model will continuously fit the priority of the actual user and/or the priority of the actual target resource information associated with the user, and the priority of the user predicted by the target decision model and/or the target resource information associated with the user. The residual of the priority makes the output result of the target decision model approach the priority of the actual user and/or the priority of the actual target resource information associated with the user.
例如,可通过以下算式(2)实现目标决策模型的更新。For example, the target decision model can be updated by the following formula (2).
F m(x)=F m-1(x)+γ mh m(x)      (2) F m (x)=F m-1 (x)+γ m h m (x) (2)
其中,F m-1(x)为前m-1目标决策模型的输出结果的加权计算总和。F m(x)为前m目标决策模型的输出结果的加权计算总和。 Among them, F m-1 (x) is the weighted calculation sum of the output results of the previous m-1 target decision model. F m (x) is the weighted calculation sum of the output results of the previous m target decision model.
还可通过对上述算式(1)和(2)中的h m(x)进行更新,优化目标决 策模型,以减小目标决策模型的输出结果与实际结果之间的误差。例如,可通过以下算式(3)实现对h m(x)的优化。 It is also possible to optimize the target decision model by updating h m (x) in the above equations (1) and (2) to reduce the error between the output result of the target decision model and the actual result. For example, the optimization of h m (x) can be achieved by the following equation (3).
Figure PCTCN2021072131-appb-000004
Figure PCTCN2021072131-appb-000004
其中,y i为第i个实际结果,即第i个用户的优先级和/或与第i个用户关联的资源信息的优先级。L(y i,F m-1(x i)+h(x i))为y i与F m-1(x i)+h(x i)的关系表达式。 Among them, y i is the i-th actual result, that is, the priority of the i-th user and/or the priority of the resource information associated with the i-th user. L(y i , F m-1 (x i )+h(x i )) is the relational expression of y i and F m-1 (x i )+h(x i ).
图2为本申请提供的资源信息推送方法的另一实施例的流程图。图2与图1的不同之处在于,图2所示的资源信息推送方法还可包括步骤S105,图1所示的步骤S102可细化为图2所示的步骤S1021。Fig. 2 is a flowchart of another embodiment of a method for pushing resource information provided by this application. The difference between FIG. 2 and FIG. 1 is that the resource information pushing method shown in FIG. 2 may further include step S105, and step S102 shown in FIG. 1 can be refined into step S1021 shown in FIG. 2.
在步骤S105中,利用可信度计算模型,计算资源消耗信息的可信度和资源提供方信息的可信度。In step S105, the credibility calculation model is used to calculate the credibility of the resource consumption information and the credibility of the resource provider information.
为了进一步提高推送资源信息的精确度,可通过采用可信度高的优先资源消耗信息和资源提供方信息参与确定用户的优先级和与用户关联的资源信息的优先级的步骤。具体地,可信度计算模型所采用的算法可包括随机森林算法,也可采用其他算法,在此并不限定。资源消耗信息的可信度表征对资源消耗信息可以信赖的程度。资源提供方信息的可信度表征对资源提供方信息可以信赖的程度。在一些示例中,可信度以数值的方式表示。例如,信息的可信度为80至100表示信息准确;信息的可信度为60至80表示信息可信可用;信息的可信度为40至60表示信息部分可用;信息的可信度为0至40表示信息有待完善。In order to further improve the accuracy of pushing resource information, it is possible to use highly reliable priority resource consumption information and resource provider information to participate in the step of determining the priority of the user and the priority of the resource information associated with the user. Specifically, the algorithm used in the credibility calculation model may include a random forest algorithm, or other algorithms may be used, which is not limited here. The credibility of resource consumption information represents the degree to which the resource consumption information can be trusted. The credibility of the resource provider's information represents the degree to which the resource provider's information can be trusted. In some examples, credibility is expressed in numerical terms. For example, a credibility of information of 80 to 100 indicates that the information is accurate; a credibility of information of 60 to 80 indicates that the information is credible and available; a credibility of information of 40 to 60 indicates that the information is partially available; the credibility of information is 0 to 40 indicate that the information needs to be improved.
本申请可利用集成学习方式进行可信度的计算及分类。集成学习通过建立几个模型组合来解决单一预测问题。下面以随机森林算法为例对可信度计算进行说明,但需要说明的是,在本申请实施例中,还可用其他算法进行可信度计算。随机森林算法是集成学习的一个子类,它依靠于决策树的投票选择来决定最后的分类结果。具体地,可利用以下的算式(4)计算信息熵。森林随机算法中通过决策树进行可信度的分类。信息熵是决策树的根本,可在决策树进行可信度的分类的过程中确定资源消耗信息和资源提供方信息选取顺序的依据。This application can use the integrated learning method to calculate and classify the credibility. Integrated learning solves a single prediction problem by building a combination of several models. The following uses the random forest algorithm as an example to describe the credibility calculation, but it should be noted that in the embodiment of the present application, other algorithms may also be used to perform the credibility calculation. The random forest algorithm is a sub-category of ensemble learning, which relies on the voting selection of the decision tree to determine the final classification result. Specifically, the following formula (4) can be used to calculate the information entropy. In the forest stochastic algorithm, a decision tree is used to classify the reliability. Information entropy is the foundation of decision trees, which can be used to determine the basis of resource consumption information and the selection order of resource provider information during the process of credibility classification of decision trees.
I(X=xi)=-log 2p(x i)      (4) I(X=xi)=-log 2 p(x i ) (4)
其中,I(x)表示随机变量的信息;p(x i)指x i发生时的概率;x i为第i个用户的资源消耗信息和资源提供方信息。 Among them, I(x) represents the information of random variables; p( xi ) refers to the probability when xi occurs; and xi is the resource consumption information and resource provider information of the i-th user.
决策树是一种树形结构,其中每个内部节点表示一个属性上的测试,每个分支代表一个测试输出,每个叶节点代表一种类别。在本示例中,每个分支表示可信度计算的一路输出,每个叶节点表示一个可信度类别。A decision tree is a tree structure in which each internal node represents a test on an attribute, each branch represents a test output, and each leaf node represents a category. In this example, each branch represents an output of credibility calculation, and each leaf node represents a credibility category.
在一些示例中,可利用多个可信度计算模型来对资源消耗信息的可信度和资源提供方信息的可信度进行多维度的计算。例如,可从信息来源和信息本身两方面评估可信度。具体地,可从信息的录入方、录入有无校验环节、是否经历技术处理和人工数据清洗等方面,评估在信息来源方面的可信度。这里以资源消耗信息与资源提供方信息中的用户行为偏好信息自身的可信度为例说明信息自身的可信度的评估。可通过用户的资源消耗的可信度、用户资源消耗方式的可信度、用户资源消耗习惯的可信度和用户资源消耗类型信息的可信度以及权重综合评估得到用户行为偏好信息的可信度。可利用信息来源的可信度、信息本身的可信度、信息来源的可信度的权重和信息本身的可信度的权重,综合得到用户的资源消耗信息的可信度和资源提供方信息的可信度。In some examples, multiple credibility calculation models may be used to perform multi-dimensional calculations on the credibility of resource consumption information and the credibility of resource provider information. For example, the credibility can be assessed from both the source of the information and the information itself. Specifically, the credibility of the information source can be evaluated from the aspects of the person who entered the information, whether the entry has a verification link, whether it has undergone technical processing and manual data cleaning, etc. Here, the credibility of the user behavior preference information in the resource consumption information and the resource provider information is taken as an example to illustrate the credibility evaluation of the information itself. The credibility of user behavior preference information can be obtained through comprehensive evaluation of the credibility of the user's resource consumption, the credibility of the user's resource consumption method, the credibility of the user's resource consumption habits, the credibility of the user's resource consumption type information, and the weight. Spend. The credibility of the information source, the credibility of the information itself, the weight of the credibility of the information source and the weight of the credibility of the information itself can be used to comprehensively obtain the credibility of the user's resource consumption information and the information of the resource provider Credibility.
可信度计算、分类的步骤可由单台配置为24C/256G/40T的大数据主机集群中的主机执行,在此并不限定。The steps of credibility calculation and classification can be performed by a single host in a big data host cluster configured as 24C/256G/40T, which is not limited here.
在步骤S1021中,基于资源消耗信息和资源提供方信息,根据资源消耗信息的可信度和资源提供方信息的可信度,确定用户的优先级,以及确定与用户关联的资源信息的优先级。In step S1021, based on the resource consumption information and the resource provider information, according to the credibility of the resource consumption information and the credibility of the resource provider information, the priority of the user is determined, and the priority of the resource information associated with the user is determined .
其中,结合资源消耗信息的可信度和资源提供方信息的可信度,可选择进行用户的优先级以及与用户关联的资源信息的优先级的确定的资源消耗信息和资源提供方信息。例如,选择可信度高的资源消耗信息和资源提供方信息进行用户的优先级以及与用户关联的资源信息的优先级的确定,以进一步提高推送资源信息的精确度和可靠性。Among them, combining the credibility of the resource consumption information and the credibility of the resource provider information, the resource consumption information and the resource provider information for determining the priority of the user and the priority of the resource information associated with the user can be selected. For example, selecting highly reliable resource consumption information and resource provider information to determine the priority of the user and the priority of the resource information associated with the user, so as to further improve the accuracy and reliability of pushing resource information.
在本申请实施例中,服务器可根据用户进行资源消耗的模式,对应向用户推送资源信息。其中,资源信息的优先级包括但不限于资源信息包括 资源提供方的优先级、资源类别的优先级、资源提供方案的优先级中的一项或多项。In the embodiment of the present application, the server may correspondingly push resource information to the user according to the user's resource consumption mode. Wherein, the priority of the resource information includes, but is not limited to, the resource information including one or more of the priority of the resource provider, the priority of the resource category, and the priority of the resource provision plan.
在一些实施例中,用户进行资源消耗的模式为线上模式。例如,在交易场景中,服务器可从用户的应用程序获取用户位置。服务器根据资源消耗信息中的用户交易商户偏好、交易商户类型偏好等信息,与资源提供方信息匹配获得用户位置所在区域内的商户的商户名称、商户类型、商户热度等信息。并根据匹配获得用户位置所在区域内的商户的商户名称、商户类型、商户热度等信息,对用户在用户位置所在区域内的资源消耗潜力进行评估。结合用于在该区域内的资源消耗潜力和用户的活跃度,确定优先级高于第二预设标准优先级的资源信息。可将优先级高于第二预设标准优先级的资源信息通过Kafka实时数据流向用户推送。例如,在用户的应用程序中向用户推送资源信息。In some embodiments, the mode in which the user consumes resources is an online mode. For example, in a transaction scenario, the server may obtain the user's location from the user's application. The server matches the information of the resource provider according to the user's transaction merchant preference and transaction merchant type preference in the resource consumption information to obtain the merchant name, merchant type, merchant popularity and other information of the merchant in the area where the user is located. And according to the matching, obtain the business name, business type, business popularity and other information of the business in the area where the user location is located, and evaluate the resource consumption potential of the user in the area where the user location is located. Combining the resource consumption potential and user activity in the area, determine resource information with a priority higher than the second preset standard priority. The resource information with priority higher than the second preset standard priority can be pushed to users through Kafka real-time data stream. For example, push resource information to the user in the user's application.
在一些示例中,可利用资源消耗信息和资源提供方信息所表征的用户的关联性和资源提供方的关联性,对用户进行聚类,以便于对聚类得到的同一簇内的用户进行资源信息的推送。In some examples, the relevance of users represented by resource consumption information and resource provider information and the relevance of resource providers can be used to cluster users, so that users in the same cluster obtained by clustering can be resourced. Push of information.
具体地,对用户进行聚类,得到至少两个簇。对于每个簇,选取一个用户的资源消耗信息和资源提供方信息作为簇的簇中心。计算簇中除簇中心对应的用户外的其他用户的资源消耗信息和资源提供方信息与簇中心的距离。根据距离,更新簇以及簇的簇中心,直至更新的簇中心与更新前的簇中心相同。Specifically, the users are clustered to obtain at least two clusters. For each cluster, select a user's resource consumption information and resource provider information as the cluster center of the cluster. Calculate the resource consumption information of other users in the cluster except the user corresponding to the cluster center and the distance between the resource provider information and the cluster center. According to the distance, update the cluster and the cluster center of the cluster until the updated cluster center is the same as the cluster center before the update.
对用户的初始分簇可随机分簇或按照预设规则分簇,在此并不限定。每个用户的资源消耗信息和资源提供方信息可采用向量的方式表示。一个簇的簇中心应为与簇内其他用户的资源消耗信息和资源提供方信息的距离均最近的用户的资源消耗信息和资源提供方信息。因此通过计算簇中除簇中心对应的用户外的其他用户的资源消耗信息和资源提供方信息与簇中心的距离,并根据该距离,对簇的划分和簇中心进行更新,以保证分簇的准确性。更新的簇中心与更新前的簇中心相同,表示簇中心已经稳定,可保持当前的分簇结果。The initial clustering of users can be randomly clustered or clustered according to a preset rule, which is not limited here. The resource consumption information and resource provider information of each user can be represented in the form of vectors. The cluster center of a cluster should be the resource consumption information and resource provider information of the user who is closest to the resource consumption information and resource provider information of other users in the cluster. Therefore, by calculating the distance between the resource consumption information and resource provider information of other users in the cluster except the user corresponding to the cluster center and the cluster center, and based on the distance, the division of the cluster and the cluster center are updated to ensure the clustering. accuracy. The updated cluster center is the same as the cluster center before the update, which means that the cluster center is stable and the current clustering result can be maintained.
例如,可利用交叉验证法选取分簇数目k。利用以下算式(5)计算簇 中各用户的资源消耗信息和资源提供方信息与簇中心的欧式距离,以便于根据欧式距离,对簇的划分和簇中心进行更新。For example, the cross-validation method can be used to select the number of clusters k. Use the following formula (5) to calculate the Euclidean distance between the resource consumption information and resource provider information of each user in the cluster and the cluster center, so as to update the cluster division and cluster center based on the Euclidean distance.
Figure PCTCN2021072131-appb-000005
Figure PCTCN2021072131-appb-000005
其中,dist ed(x (i),x (j))为簇中第i个用户的资源消耗信息和资源提供方信息形成的向量与第j个用户的资源消耗信息和资源提供方信息形成的向量的欧式距离;n为簇中的用户总量;u为向量维度。 Among them, dist ed (x (i) , x (j) ) is the vector formed by the resource consumption information and resource provider information of the i-th user in the cluster and the resource consumption information and resource provider information of the j-th user The Euclidean distance of the vector; n is the total number of users in the cluster; u is the vector dimension.
同一簇中的用户的资源消耗具有一定相似性,向同一簇中的用户推送的资源信息中至少一项相同。例如,在交易场景中,向同一簇中的用户推送的资源信息中的资源类别均为“餐饮类”,但向同一簇中的用户推送的资源信息中的资源提供方案中涉及的交易优惠算法不同。The resource consumption of users in the same cluster has a certain similarity, and at least one item of resource information pushed to users in the same cluster is the same. For example, in a transaction scenario, the resource categories in the resource information pushed to users in the same cluster are all "catering", but the transaction discount algorithm involved in the resource provision scheme in the resource information pushed to users in the same cluster different.
用户进行资源消耗的模式为线上模式对应的服务器可包括单台配置为24C/256G/40T的大数据主机集群与单台配置为24C/256G/40T的Kafka集群,在此并不限定。The user's resource consumption mode is the online mode. The server corresponding to the online mode may include a single big data host cluster configured as 24C/256G/40T and a single Kafka cluster configured as 24C/256G/40T, which is not limited here.
在另一些实施例中,用户进行资源消耗的模式为线下模式。例如,在交易场景中,用户利用银行卡进行交易,利用资源消耗信息中的交易流水商户位置确定用户的所处的商圈,利用协同过滤算法,匹配资源提供方信息中的该商圈位置中的商户名称、商户类型、商户消费相似度等信息,结合用户的活跃度,确定推送的目标资源信息。例如,可确定目标资源信息中资源提供方的优先级、资源类别的优先级、资源提供方案的优先级中的一项或多项,将优先级高于第二预设标准优先级资源提供方、资源类别、资源提供方案中的一项或多项作为目标资源信息向用户推送,即确定了向用户推送的目标资源信息。在一些示例中,可通过Kafka实时数据流将目标资源信息推送给用户。In other embodiments, the mode in which the user consumes resources is an offline mode. For example, in a transaction scenario, a user uses a bank card to conduct a transaction, uses the transaction flow in the resource consumption information to determine the user’s business district, and uses a collaborative filtering algorithm to match the business district’s location in the resource provider’s information To determine the target resource information to be pushed, combined with information such as the name of the merchant, the type of the merchant, and the similarity of the merchant’s consumption, combined with the user’s activity. For example, one or more of the priority of the resource provider, the priority of the resource category, and the priority of the resource provision plan in the target resource information can be determined, and the priority is higher than the second preset standard priority resource provider One or more of, resource category, and resource provision plan are pushed to the user as target resource information, that is, the target resource information pushed to the user is determined. In some examples, the target resource information can be pushed to the user through the Kafka real-time data stream.
用户进行资源消耗的模式为线下模式对应的服务器可包括单台配置为24C/256G/40T的大数据主机集群与单台配置为24C/256G/40T的Kafka集群,在此并不限定。The user's resource consumption mode is offline. The server corresponding to the offline mode may include a single big data host cluster configured as 24C/256G/40T and a single Kafka cluster configured as 24C/256G/40T, which is not limited here.
图3为本申请提供的资源信息推送装置的一实施例的结构示意图。如 图3所示,该资源信息推送装置200可包括信息获取模块201、优先级确定模块202、目标确定模块203和推送模块204。FIG. 3 is a schematic structural diagram of an embodiment of the resource information pushing device provided by this application. As shown in FIG. 3, the resource information pushing device 200 may include an information acquiring module 201, a priority determining module 202, a target determining module 203, and a pushing module 204.
信息获取模块201,用于获取用户的资源消耗信息和资源提供方信息。The information acquisition module 201 is used to acquire user resource consumption information and resource provider information.
在一些示例中,资源消耗信息包括以下的一项或多项:用户信息、用户消耗资源时间偏好信息、用户消耗资源的资源提供方偏好信息、用户消耗资源类型偏好信息、用户消耗资源的资源提供方类型偏好信息、用户消耗资源的资源提供方位置信息、用户消耗资源的资源提供方关联性信息。In some examples, the resource consumption information includes one or more of the following: user information, user resource consumption time preference information, user resource consumption resource provider preference information, user resource consumption type preference information, user resource consumption resource provision Party type preference information, location information of the resource provider where the user consumes resources, and information about the relevance of the resource provider where the user consumes resources.
在一些示例中,资源提供方信息包括以下的一项或多项:资源提供方名称信息,资源提供方位置信息、资源提供方所属区域位置信息、资源提供方类型信息、资源提供方评价信息、资源提供方热度信息、资源提供方的资源提供关联性信息。In some examples, the resource provider information includes one or more of the following: resource provider name information, resource provider location information, resource provider location information, resource provider type information, resource provider evaluation information, Information on the popularity of the resource provider and the relevance information provided by the resource provider’s resources.
优先级确定模块202,用于基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。The priority determination module 202 is configured to determine the priority of the user based on the resource consumption information and the resource provider information, and determine the priority of the resource information associated with the user.
在一些示例中,资源信息包括资源提供方、资源类别、资源提供方案中的一项或多项。In some examples, the resource information includes one or more of resource providers, resource types, and resource provision schemes.
在一些示例中,资源信息的优先级包括资源信息包括资源提供方的优先级、资源类别的优先级、资源提供方案的优先级中的一项或多项。In some examples, the priority of the resource information includes one or more of the resource information including the priority of the resource provider, the priority of the resource category, and the priority of the resource providing solution.
目标确定模块203,用于将优先级高于第一预设标准优先级的用户作为目标用户,将与用户关联的优先级高于第二预设标准优先级的资源信息作为目标资源信息。The target determination module 203 is configured to use a user with a priority higher than a first preset standard priority as a target user, and use resource information associated with the user with a priority higher than a second preset standard priority as the target resource information.
推送模块204,用于向目标用户推送与目标用户关联的目标资源信息。The pushing module 204 is used to push target resource information associated with the target user to the target user.
在一些示例中,推送模块204可具体用于:利用实时数据流向目标用户推送与目标用户关联的目标资源信息。In some examples, the pushing module 204 may be specifically configured to push the target resource information associated with the target user to the target user using real-time data streams.
在本申请实施例中,基于资源消耗信息和资源提供方信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。向优先级高于第一预设标准优先级的用户推送与该用户关联的优先级高于第二预设标准优先级的资源信息。资源提供方信息表征资源提供方的相关信息,可弥补资源 消耗信息所缺失的部分。结合资源消耗信息和资源提供方信息,利用优先级,实现了资源信息向用户的精确推送,提高了向用户推送资源信息的精确度。In the embodiment of the present application, based on the resource consumption information and the resource provider information, the priority of the user is determined, and the priority of the resource information associated with the user is determined. Push resource information associated with the user whose priority is higher than the second preset standard priority to a user whose priority is higher than the first preset standard priority. The resource provider information represents the relevant information of the resource provider and can make up for the missing part of the resource consumption information. Combining resource consumption information and resource provider information, and using priority, the accurate push of resource information to users is realized, and the accuracy of pushing resource information to users is improved.
在一些示例中,优先级确定模块202可具体用于:基于资源消耗信息和资源提供方信息,设置用户的活跃度信息和资源消耗潜力信息;根据用户的活跃度信息和资源消耗潜力信息,确定用户的优先级,以及确定与用户关联的资源信息的优先级。In some examples, the priority determination module 202 may be specifically configured to: set user activity information and resource consumption potential information based on resource consumption information and resource provider information; determine according to user activity information and resource consumption potential information The priority of the user, and the priority of determining the resource information associated with the user.
图4为本申请提供的资源信息推送装置的另一实施例的结构示意图。图4与图3的不同之处在于,图4所示的资源信息推送装置200还可包括采集模块205和更新模块206。FIG. 4 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application. The difference between FIG. 4 and FIG. 3 is that the resource information pushing device 200 shown in FIG. 4 may further include a collection module 205 and an update module 206.
上述实施例中的优先级确定模块202可具体用于:对于每个用户,将用户的资源消耗信息和资源提供方信息作为每个目标决策模型的输入,得到每个目标决策模型的输出结果;计算每个目标决策模型的输出结果与目标决策模型的权重的乘积之和;根据乘积之和确定用户的优先级和/或与用户关联的资源信息的优先级。The priority determining module 202 in the foregoing embodiment can be specifically used to: for each user, take the user's resource consumption information and resource provider information as the input of each target decision model, and obtain the output result of each target decision model; Calculate the sum of the product of the output result of each target decision model and the weight of the target decision model; determine the priority of the user and/or the priority of the resource information associated with the user according to the sum of the products.
其中,输出结果表征在目标决策模型对应的目标要求下,用户的优先级和/或与用户关联的资源信息的优先级Among them, the output result represents the priority of the user and/or the priority of the resource information associated with the user under the target requirements corresponding to the target decision model
采集模块205用于采集目标用户接收到目标资源信息后反馈的资源消耗结果信息。The collection module 205 is configured to collect resource consumption result information fed back by the target user after receiving the target resource information.
更新模块206用于利用资源消耗结果信息更新目标决策模型。The update module 206 is used to update the target decision model using the resource consumption result information.
图5为本申请提供的资源信息推送装置的又一实施例的结构示意图。图5与图3的不同之处在于,图5所示的资源信息推送装置200还可包括可信度计算模块207。FIG. 5 is a schematic structural diagram of another embodiment of the resource information pushing device provided by this application. The difference between FIG. 5 and FIG. 3 is that the resource information pushing device 200 shown in FIG. 5 may further include a credibility calculation module 207.
可信度计算模块207可用于利用可信度计算模型,计算资源消耗信息的可信度和资源提供方信息的可信度。The credibility calculation module 207 can be used to use the credibility calculation model to calculate the credibility of resource consumption information and the credibility of resource provider information.
上述实施例中的优先级确定模块202可具体用于:基于资源消耗信息和资源提供方信息,根据资源消耗信息的可信度和资源提供方信息的可信度,确定用户的优先级,以及确定与用户关联的资源信息的优先级。The priority determining module 202 in the foregoing embodiment may be specifically used to determine the priority of the user based on the resource consumption information and the resource provider information, according to the credibility of the resource consumption information and the credibility of the resource provider information, and Determine the priority of the resource information associated with the user.
图6为本申请提供的资源信息推送装置的再一实施例的结构示意图。 图6与图3的不同之处在于,图6所示的资源信息推送装置200还可包括聚类模块208。FIG. 6 is a schematic structural diagram of still another embodiment of the resource information pushing device provided by this application. The difference between FIG. 6 and FIG. 3 is that the resource information pushing device 200 shown in FIG. 6 may further include a clustering module 208.
聚类模块208可用于:对用户进行聚类,得到至少两个簇;对于每个簇,选取一个用户的资源消耗信息和资源提供方信息作为簇的簇中心;计算簇中除簇中心对应的用户外的其他用户的资源消耗信息和资源提供方信息与簇中心的距离;根据距离,更新簇以及簇的簇中心,直至更新的簇中心与更新前的簇中心相同。The clustering module 208 can be used to: cluster users to obtain at least two clusters; for each cluster, select the resource consumption information and resource provider information of a user as the cluster center; The resource consumption information of users other than the user and the distance between the resource provider information and the cluster center; according to the distance, the cluster and the cluster center of the cluster are updated until the updated cluster center is the same as the cluster center before the update.
在一些示例中,资源信息包括资源提供方、资源类别、资源提供方案中的一项或多项,向同一簇中用户推送的资源信息中至少一项相同。In some examples, the resource information includes one or more of a resource provider, a resource category, and a resource providing solution, and at least one of the resource information pushed to users in the same cluster is the same.
图7为本申请提供的服务器的一实施例的结构示意图。如图7所示,服务器300包括存储器301、处理器302及存储在存储器301上并可在处理器302上运行的计算机程序。FIG. 7 is a schematic structural diagram of an embodiment of a server provided by this application. As shown in FIG. 7, the server 300 includes a memory 301, a processor 302, and a computer program stored on the memory 301 and running on the processor 302.
在一个示例中,上述处理器302可以包括中央处理器(CPU),或者特定集成电路(ASIC),或者可以被配置成实施本申请实施例的一个或多个集成电路。In an example, the aforementioned processor 302 may include a central processing unit (CPU), or a specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
存储器301可以包括用于数据或指令的大容量存储器。举例来说而非限制,存储器301可包括HDD、软盘驱动器、闪存、光盘、磁光盘、磁带或通用串行总线(USB)驱动器或者两个或更多个以上这些的组合。在合适的情况下,存储器301可包括可移除或不可移除(或固定)的介质。在合适的情况下,存储器301可在终端热点开启服务器300的内部或外部。在特定实施例中,存储器301是非易失性固态存储器。在特定实施例中,存储器301包括只读存储器(ROM)。在合适的情况下,该ROM可以是掩模编程的ROM、可编程ROM(PROM)、可擦除PROM(EPROM)、电可擦除PROM(EEPROM)、电可改写ROM(EAROM)或闪存或者两个或更多个以上这些的组合。The memory 301 may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory 301 may include an HDD, a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the storage 301 may include removable or non-removable (or fixed) media. Under appropriate circumstances, the storage 301 can be opened inside or outside the server 300 in a terminal hotspot. In a particular embodiment, the memory 301 is a non-volatile solid state memory. In a particular embodiment, the memory 301 includes read-only memory (ROM). Where appropriate, the ROM can be mask-programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically rewritable ROM (EAROM) or flash memory or A combination of two or more of these.
处理器302通过读取存储器301中存储的可执行程序代码来运行与可执行程序代码对应的计算机程序,以用于实现上述实施例中的资源信息推送方法。The processor 302 runs the computer program corresponding to the executable program code by reading the executable program code stored in the memory 301, so as to implement the resource information pushing method in the foregoing embodiment.
在一个示例中,服务器300还可包括通信接口303和总线304。其中,如图7所示,存储器301、处理器302、通信接口303通过总线304连接并完成相互间的通信。In an example, the server 300 may further include a communication interface 303 and a bus 304. Among them, as shown in FIG. 7, the memory 301, the processor 302, and the communication interface 303 are connected through the bus 304 and complete the communication with each other.
通信接口303,主要用于实现本申请实施例中各模块、装置、单元和/或设备之间的通信。也可通过通信接口303接入输入设备和/或输出设备。The communication interface 303 is mainly used to implement communication between various modules, devices, units and/or devices in the embodiments of the present application. The input device and/or output device can also be accessed through the communication interface 303.
总线304包括硬件、软件或两者,将服务器300的部件彼此耦接在一起。举例来说而非限制,总线304可包括加速图形端口(AGP)或其他图形总线、增强工业标准架构(EISA)总线、前端总线(FSB)、超传输(HT)互连、工业标准架构(ISA)总线、无限带宽互连、低引脚数(LPC)总线、存储器总线、微信道架构(MCA)总线、外围组件互连(PCI)总线、PCI-Express(PCI-X)总线、串行高级技术附件(SATA)总线、视频电子标准协会局部(VLB)总线或其他合适的总线或者两个或更多个以上这些的组合。在合适的情况下,总线304可包括一个或多个总线。尽管本申请实施例描述和示出了特定的总线,但本申请考虑任何合适的总线或互连。The bus 304 includes hardware, software, or both, and couples the components of the server 300 to each other. For example and not limitation, the bus 304 may include an accelerated graphics port (AGP) or other graphics bus, an enhanced industry standard architecture (EISA) bus, a front side bus (FSB), a hypertransport (HT) interconnect, an industry standard architecture (ISA) ) Bus, unlimited bandwidth interconnect, low pin count (LPC) bus, memory bus, microchannel architecture (MCA) bus, peripheral component interconnect (PCI) bus, PCI-Express (PCI-X) bus, serial advanced Technical Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of these. Where appropriate, the bus 304 may include one or more buses. Although the embodiments of this application describe and show a specific bus, this application considers any suitable bus or interconnection.
本申请一实施例还提供一种计算机可读存储介质,该计算机可读存储介质上存储有计算机程序,该计算机程序被处理器执行时可实现上述实施例中的资源信息推送方法,且能达到相同的技术效果,为避免重复,这里不再赘述。其中,所述的计算机可读存储介质可包括非暂态计算机可读存储介质,如只读存储器(Read-Only Memory,简称ROM)、随机存取存储器(Random Access Memory,简称RAM)、磁碟或者光盘等。An embodiment of the present application also provides a computer-readable storage medium having a computer program stored on the computer-readable storage medium. When the computer program is executed by a processor, the method for pushing resource information in the above-mentioned embodiment can be implemented, and can achieve The same technical effect, in order to avoid repetition, will not be repeated here. Wherein, the computer-readable storage medium may include non-transitory computer-readable storage media, such as read-only memory (Read-Only Memory, ROM for short), Random Access Memory (RAM for short), and magnetic disks. Or CD etc.
需要明确的是,本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同或相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。对于装置实施例、服务器实施例和计算机可读存储介质实施例而言,相关之处可以参见方法实施例的说明部分。本申请并不局限于上文所描述并在图中示出的特定步骤和结构。本领域的技术人员可以在领会本申请的精神之后,作出各种改变、修改和添加,或者改变步骤之间的顺序。并且,为了简明起见,这里省略对已知方法技术的详细描述。It should be clear that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. Place. For the device embodiment, the server embodiment, and the computer-readable storage medium embodiment, see the description part of the method embodiment for relevant details. The application is not limited to the specific steps and structures described above and shown in the figures. Those skilled in the art can make various changes, modifications and additions, or change the order between steps after grasping the spirit of the application. And, for the sake of brevity, a detailed description of the known method and technology is omitted here.
上面参考根据本申请的实施例的方法、装置(系统)和机器程序产品 的流程图和/或框图描述了本申请的各方面。应当理解,流程图和/或框图中的每个方框以及流程图和/或框图中各方框的组合可以由程序或指令实现。这些程序或指令可被提供给通用计算机、专用计算机、或其它可编程数据处理装置的处理器,以产生一种机器,使得经由计算机或其它可编程数据处理装置的处理器执行的这些程序或指令使能对流程图和/或框图的一个或多个方框中指定的功能/动作的实现。这种处理器可以是但不限于是通用处理器、专用处理器、特殊应用处理器或者现场可编程逻辑电路。还可理解,框图和/或流程图中的每个方框以及框图和/或流程图中的方框的组合,也可以由执行指定的功能或动作的专用硬件来实现,或可由专用硬件和计算机指令的组合来实现。The above describes various aspects of the present application with reference to the flowcharts and/or block diagrams of the methods, devices (systems) and machine program products according to the embodiments of the present application. It should be understood that each block in the flowchart and/or block diagram and the combination of each block in the flowchart and/or block diagram can be implemented by a program or instruction. These programs or instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to generate a machine so that these programs or instructions are executed by the processor of the computer or other programmable data processing device Enables the realization of functions/actions specified in one or more blocks of the flowchart and/or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a dedicated processor, a special application processor, or a field programmable logic circuit. It can also be understood that each block in the block diagram and/or flowchart and the combination of the blocks in the block diagram and/or flowchart can also be implemented by dedicated hardware that performs specified functions or actions, or can be implemented by dedicated hardware and A combination of computer instructions.
本领域技术人员应能理解,上述实施例均是示例性而非限制性的。在不同实施例中出现的不同技术特征可以进行组合,以取得有益效果。本领域技术人员在研究附图、说明书及权利要求书的基础上,应能理解并实现所揭示的实施例的其他变化的实施例。在权利要求书中,术语“包括”并不排除其他装置或步骤;数量词“一个”不排除多个;术语“第一”、“第二”用于标示名称而非用于表示任何特定的顺序。权利要求中的任何附图标记均不应被理解为对保护范围的限制。权利要求中出现的多个部分的功能可以由一个单独的硬件或软件模块来实现。某些技术特征出现在不同的从属权利要求中并不意味着不能将这些技术特征进行组合以取得有益效果。Those skilled in the art should understand that the above-mentioned embodiments are all exemplary rather than restrictive. Different technical features appearing in different embodiments can be combined to achieve beneficial effects. Those skilled in the art should be able to understand and implement other modified embodiments of the disclosed embodiments on the basis of studying the drawings, the description, and the claims. In the claims, the term "comprising" does not exclude other means or steps; the quantifier "a" does not exclude multiple; the terms "first" and "second" are used to denote names rather than to indicate any specific order . Any reference signs in the claims should not be construed as limiting the scope of protection. The functions of multiple parts appearing in the claims can be implemented by a single hardware or software module. The appearance of certain technical features in different dependent claims does not mean that these technical features cannot be combined to achieve beneficial effects.

Claims (26)

  1. 一种资源信息推送方法,包括:A method for pushing resource information, including:
    获取用户的资源消耗信息和资源提供方信息;Obtain user resource consumption information and resource provider information;
    基于所述资源消耗信息和所述资源提供方信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级;Determining the priority of the user based on the resource consumption information and the resource provider information, and determining the priority of resource information associated with the user;
    将优先级高于第一预设标准优先级的用户作为目标用户,将与所述用户关联的优先级高于第二预设标准优先级的资源信息作为目标资源信息;Taking a user whose priority is higher than the first preset standard priority as the target user, and taking the resource information associated with the user whose priority is higher than the second preset standard priority as the target resource information;
    向所述目标用户推送与所述目标用户关联的所述目标资源信息。Push the target resource information associated with the target user to the target user.
  2. 根据权利要求1所述的方法,其中,所述基于所述资源消耗信息和所述资源提供方信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级,包括:The method according to claim 1, wherein the determining the priority of the user based on the resource consumption information and the resource provider information, and determining the priority of the resource information associated with the user comprises :
    基于所述资源消耗信息和所述资源提供方信息,设置所述用户的活跃度信息和资源消耗潜力信息;Based on the resource consumption information and the resource provider information, setting the user's activity information and resource consumption potential information;
    根据所述用户的所述活跃度信息和所述资源消耗潜力信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级。According to the activity information of the user and the resource consumption potential information, the priority of the user is determined, and the priority of the resource information associated with the user is determined.
  3. 根据权利要求1所述的方法,其中,所述基于所述资源消耗信息和所述资源提供方信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级,包括:The method according to claim 1, wherein the determining the priority of the user based on the resource consumption information and the resource provider information, and determining the priority of the resource information associated with the user comprises :
    对于每个所述用户,将所述用户的所述资源消耗信息和所述资源提供方信息作为每个目标决策模型的输入,得到每个目标决策模型的输出结果,所述输出结果表征在所述目标决策模型对应的目标要求下,所述用户的优先级和/或与所述用户关联的资源信息的优先级;For each user, use the user's resource consumption information and the resource provider information as the input of each target decision model, and obtain the output result of each target decision model. Under the target requirements corresponding to the target decision model, the priority of the user and/or the priority of the resource information associated with the user;
    计算每个所述目标决策模型的输出结果与所述目标决策模型的权重的乘积之和;Calculating the sum of the product of the output result of each target decision model and the weight of the target decision model;
    根据所述乘积之和确定所述用户的优先级和/或与所述用户关联的资源信息的优先级。The priority of the user and/or the priority of the resource information associated with the user is determined according to the sum of the products.
  4. 根据权利要求3所述的方法,其中,还包括:The method according to claim 3, further comprising:
    采集所述目标用户接收到所述目标资源信息后反馈的资源消耗结果信 息;Collecting resource consumption result information fed back by the target user after receiving the target resource information;
    利用所述资源消耗结果信息更新所述目标决策模型。Update the target decision model using the resource consumption result information.
  5. 根据权利要求1所述的方法,其中,在所述基于所述资源消耗信息和所述资源提供方信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级之前,还包括:The method according to claim 1, wherein before the determining the priority of the user based on the resource consumption information and the resource provider information, and determining the priority of the resource information associated with the user ,Also includes:
    利用可信度计算模型,计算所述资源消耗信息的可信度和所述资源提供方信息的可信度。The credibility calculation model is used to calculate the credibility of the resource consumption information and the credibility of the resource provider information.
  6. 根据权利要求5所述的方法,其中,所述基于所述资源消耗信息和所述资源提供方信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级,包括:The method according to claim 5, wherein the determining the priority of the user based on the resource consumption information and the resource provider information, and determining the priority of the resource information associated with the user comprises :
    基于所述资源消耗信息和所述资源提供方信息,根据所述资源消耗信息的可信度和所述资源提供方信息的可信度,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级。Based on the resource consumption information and the resource provider information, according to the credibility of the resource consumption information and the credibility of the resource provider information, determine the priority of the user, and determine the relationship with the user The priority of the associated resource information.
  7. 根据权利要求1所述的方法,其中,还包括:The method according to claim 1, further comprising:
    对所述用户进行聚类,得到至少两个簇;Clustering the users to obtain at least two clusters;
    对于每个簇,选取一个所述用户的所述资源消耗信息和所述资源提供方信息作为所述簇的簇中心;For each cluster, select one of the user's resource consumption information and the resource provider information as the cluster center of the cluster;
    计算所述簇中除所述簇中心对应的所述用户外的其他所述用户的所述资源消耗信息和所述资源提供方信息与所述簇中心的距离;Calculating the resource consumption information and the distance between the resource provider information and the cluster center of the users other than the user corresponding to the cluster center in the cluster;
    根据所述距离,更新所述簇以及所述簇的所述簇中心,直至更新的所述簇中心与更新前的所述簇中心相同。According to the distance, the cluster and the cluster center of the cluster are updated until the updated cluster center is the same as the cluster center before the update.
  8. 根据权利要求7所述的方法,其中,所述资源信息包括资源提供方、资源类别、资源提供方案中的一项或多项,The method according to claim 7, wherein the resource information includes one or more of resource providers, resource types, and resource provision solutions,
    向同一所述簇中所述用户推送的资源信息中至少一项相同。At least one item of resource information pushed to the users in the same cluster is the same.
  9. 根据权利要求1所述的方法,其中,所述资源信息的优先级包括资源信息包括资源提供方的优先级、资源类别的优先级、资源提供方案的优先级中的一项或多项。The method according to claim 1, wherein the priority of the resource information includes resource information including one or more of the priority of the resource provider, the priority of the resource category, and the priority of the resource provision scheme.
  10. 根据权利要求1所述的方法,其中,所述向所述目标用户推送与所述目标用户关联的所述目标资源信息,包括:The method according to claim 1, wherein said pushing said target resource information associated with said target user to said target user comprises:
    利用实时数据流向所述目标用户推送与所述目标用户关联的所述目标资源信息。Push the target resource information associated with the target user to the target user using real-time data streams.
  11. 根据权利要求1至10中任一项所述的方法,其中,所述资源消耗信息包括以下的一项或多项:The method according to any one of claims 1 to 10, wherein the resource consumption information includes one or more of the following:
    用户信息、用户消耗资源时间偏好信息、用户消耗资源的资源提供方偏好信息、用户消耗资源类型偏好信息、用户消耗资源的资源提供方类型偏好信息、用户消耗资源的资源提供方位置信息、用户消耗资源的资源提供方关联性信息。User information, user resource consumption time preference information, user resource consumption resource provider preference information, user resource consumption type preference information, user resource consumption resource type preference information, user resource consumption resource provider location information, user consumption Relevance information of the resource provider of the resource.
  12. 根据权利要求1至10中任意一项所述的方法,其中,所述资源提供方信息包括以下的一项或多项:The method according to any one of claims 1 to 10, wherein the resource provider information includes one or more of the following:
    资源提供方名称信息,资源提供方位置信息、资源提供方所属区域位置信息、资源提供方类型信息、资源提供方评价信息、资源提供方热度信息、资源提供方的资源提供关联性信息。Resource provider name information, resource provider location information, resource provider location information, resource provider type information, resource provider evaluation information, resource provider popularity information, and resource provider’s resource providing relevance information.
  13. 一种资源信息推送装置,包括:A resource information pushing device, including:
    信息获取模块,用于获取用户的资源消耗信息和资源提供方信息;The information acquisition module is used to acquire the user's resource consumption information and resource provider information;
    优先级确定模块,用于基于所述资源消耗信息和所述资源提供方信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级;A priority determining module, configured to determine the priority of the user based on the resource consumption information and the resource provider information, and determine the priority of resource information associated with the user;
    目标确定模块,用于将优先级高于第一预设标准优先级的用户作为目标用户,将与所述用户关联的优先级高于第二预设标准优先级的资源信息作为目标资源信息;The target determination module is configured to use a user with a priority higher than a first preset standard priority as a target user, and use resource information associated with the user with a priority higher than a second preset standard priority as the target resource information;
    推送模块,用于向所述目标用户推送与所述目标用户关联的所述目标资源信息。The pushing module is configured to push the target resource information associated with the target user to the target user.
  14. 根据权利要求13所述的装置,其中,所述优先级确定模块具体用于:The device according to claim 13, wherein the priority determining module is specifically configured to:
    基于所述资源消耗信息和所述资源提供方信息,设置所述用户的活跃度信息和资源消耗潜力信息;Based on the resource consumption information and the resource provider information, setting the user's activity information and resource consumption potential information;
    根据所述用户的所述活跃度信息和所述资源消耗潜力信息,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级。According to the activity information of the user and the resource consumption potential information, the priority of the user is determined, and the priority of the resource information associated with the user is determined.
  15. 根据权利要求13所述的装置,其中,所述优先级确定模块具体用于:The device according to claim 13, wherein the priority determining module is specifically configured to:
    对于每个所述用户,将所述用户的所述资源消耗信息和所述资源提供方信息作为每个目标决策模型的输入,得到每个目标决策模型的输出结果,所述输出结果表征在所述目标决策模型对应的目标要求下,所述用户的优先级和/或与所述用户关联的资源信息的优先级;For each user, use the user's resource consumption information and the resource provider information as the input of each target decision model, and obtain the output result of each target decision model. Under the target requirements corresponding to the target decision model, the priority of the user and/or the priority of the resource information associated with the user;
    计算每个所述目标决策模型的输出结果与所述目标决策模型的权重的乘积之和;Calculating the sum of the product of the output result of each target decision model and the weight of the target decision model;
    根据所述乘积之和确定所述用户的优先级和/或与所述用户关联的资源信息的优先级。The priority of the user and/or the priority of the resource information associated with the user is determined according to the sum of the products.
  16. 根据权利要求15所述的装置,其中,还包括:The device according to claim 15, further comprising:
    采集模块,用于采集所述目标用户接收到所述目标资源信息后反馈的资源消耗结果信息;A collection module, configured to collect resource consumption result information fed back after the target user receives the target resource information;
    更新模块,用于利用所述资源消耗结果信息更新所述目标决策模型。The update module is used to update the target decision model by using the resource consumption result information.
  17. 根据权利要求13所述的装置,其中,还包括:The device according to claim 13, further comprising:
    可信度计算模块,用于利用可信度计算模型,计算所述资源消耗信息的可信度和所述资源提供方信息的可信度。The credibility calculation module is used to use the credibility calculation model to calculate the credibility of the resource consumption information and the credibility of the resource provider information.
  18. 根据权利要求17所述的装置,其中,所述优先级确定模块具体用于:The device according to claim 17, wherein the priority determining module is specifically configured to:
    基于所述资源消耗信息和所述资源提供方信息,根据所述资源消耗信息的可信度和所述资源提供方信息的可信度,确定所述用户的优先级,以及确定与所述用户关联的资源信息的优先级。Based on the resource consumption information and the resource provider information, according to the credibility of the resource consumption information and the credibility of the resource provider information, determine the priority of the user, and determine the relationship with the user The priority of the associated resource information.
  19. 根据权利要求13所述的装置,其中,还包括聚类模块,所述聚类模块用于:The device according to claim 13, further comprising a clustering module, the clustering module being configured to:
    对所述用户进行聚类,得到至少两个簇;Clustering the users to obtain at least two clusters;
    对于每个簇,选取一个所述用户的所述资源消耗信息和所述资源提供方信息作为所述簇的簇中心;For each cluster, select one of the user's resource consumption information and the resource provider information as the cluster center of the cluster;
    计算所述簇中除所述簇中心对应的所述用户外的其他所述用户的所述资源消耗信息和所述资源提供方信息与所述簇中心的距离;Calculating the resource consumption information and the distance between the resource provider information and the cluster center of the users other than the user corresponding to the cluster center in the cluster;
    根据所述距离,更新所述簇以及所述簇的所述簇中心,直至更新的所述簇中心与更新前的所述簇中心相同。According to the distance, the cluster and the cluster center of the cluster are updated until the updated cluster center is the same as the cluster center before the update.
  20. 根据权利要求19所述的装置,其中,所述资源信息包括资源提供方、资源类别、资源提供方案中的一项或多项,The device according to claim 19, wherein the resource information includes one or more of resource providers, resource types, and resource provision solutions,
    向同一所述簇中所述用户推送的资源信息中至少一项相同。At least one item of resource information pushed to the users in the same cluster is the same.
  21. 根据权利要求13所述的装置,其中,所述资源信息的优先级包括资源信息包括资源提供方的优先级、资源类别的优先级、资源提供方案的优先级中的一项或多项。The apparatus according to claim 13, wherein the priority of the resource information includes resource information including one or more of the priority of the resource provider, the priority of the resource category, and the priority of the resource provision scheme.
  22. 根据权利要求13所述的装置,其中,所述推送模块具体用于:The device according to claim 13, wherein the pushing module is specifically configured to:
    利用实时数据流向所述目标用户推送与所述目标用户关联的所述目标资源信息。Push the target resource information associated with the target user to the target user using real-time data streams.
  23. 根据权利要求13至22中任一项所述的装置,其中,所述资源消耗信息包括以下的一项或多项:The device according to any one of claims 13 to 22, wherein the resource consumption information includes one or more of the following:
    用户信息、用户消耗资源时间偏好信息、用户消耗资源的资源提供方偏好信息、用户消耗资源类型偏好信息、用户消耗资源的资源提供方类型偏好信息、用户消耗资源的资源提供方位置信息、用户消耗资源的资源提供方关联性信息。User information, user resource consumption time preference information, user resource consumption resource provider preference information, user resource consumption type preference information, user resource consumption resource type preference information, user resource consumption resource provider location information, user consumption Relevance information of the resource provider of the resource.
  24. 根据权利要求13至22中任意一项所述的装置,其中,所述资源提供方信息包括以下的一项或多项:The apparatus according to any one of claims 13 to 22, wherein the resource provider information includes one or more of the following:
    资源提供方名称信息,资源提供方位置信息、资源提供方所属区域位置信息、资源提供方类型信息、资源提供方评价信息、资源提供方热度信息、资源提供方的资源提供关联性信息。Resource provider name information, resource provider location information, resource provider location information, resource provider type information, resource provider evaluation information, resource provider popularity information, and resource provider’s resource providing relevance information.
  25. 一种服务器,包括处理器、存储器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现如权利要求1至12中任意一项所述的资源信息推送方法。A server comprising a processor, a memory, and a computer program stored on the memory and capable of running on the processor. The computer program is executed by the processor to implement any one of claims 1 to 12 The resource information push method described in item.
  26. 一种计算机可读存储介质,所述计算机可读存储介质上存储计算机程序,所述计算机程序被处理器执行时实现如权利要求1至12中任意一项所述的资源信息推送方法。A computer-readable storage medium storing a computer program on the computer-readable storage medium, and when the computer program is executed by a processor, the method for pushing resource information according to any one of claims 1 to 12 is realized.
PCT/CN2021/072131 2020-02-24 2021-01-15 Resource information pushing method and device, server, and storage medium WO2021169659A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010113562.2 2020-02-24
CN202010113562.2A CN111371853B (en) 2020-02-24 2020-02-24 Resource information pushing method, device, server and storage medium

Publications (1)

Publication Number Publication Date
WO2021169659A1 true WO2021169659A1 (en) 2021-09-02

Family

ID=71212070

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/072131 WO2021169659A1 (en) 2020-02-24 2021-01-15 Resource information pushing method and device, server, and storage medium

Country Status (3)

Country Link
CN (1) CN111371853B (en)
TW (1) TWI803823B (en)
WO (1) WO2021169659A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113986486A (en) * 2021-10-15 2022-01-28 东华大学 Joint optimization method for data caching and task scheduling in edge environment

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111371853B (en) * 2020-02-24 2022-12-20 中国银联股份有限公司 Resource information pushing method, device, server and storage medium
CN112925982B (en) * 2021-03-12 2023-04-07 上海意略明数字科技股份有限公司 User redirection method and device, storage medium and computer equipment
CN116542443A (en) * 2023-03-29 2023-08-04 浙江中兴慧农信息科技有限公司 Resource allocation method and equipment based on bilateral matching model

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140337424A1 (en) * 2013-05-10 2014-11-13 Apple Inc. Authorizing Push Notifications for Websites
CN105931066A (en) * 2015-09-24 2016-09-07 中国银联股份有限公司 Transaction data processing method and device
CN108446963A (en) * 2018-03-30 2018-08-24 深圳春沐源控股有限公司 Resource information recommends method and resource information commending system
CN109711950A (en) * 2019-01-18 2019-05-03 深圳壹账通智能科技有限公司 Recommended method, device, storage medium and the computer equipment of marketing product
CN111371853A (en) * 2020-02-24 2020-07-03 中国银联股份有限公司 Resource information pushing method and device, server and storage medium

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140123229A1 (en) * 2012-10-26 2014-05-01 Motorola Solutions, Inc. Antenna/actuation key assembly
KR102120911B1 (en) * 2013-09-25 2020-06-09 삼성에스디에스 주식회사 Apparatus and Method for Controlling Demand Based Priority Rank, Apparatus and Method for Determining Priority Rank And Recording Medium Recording the Program Thereof
TW201826187A (en) * 2017-01-11 2018-07-16 富邦人壽保險股份有限公司 Costumer analysis system and method
CN108335138B (en) * 2018-02-01 2020-03-06 百度在线网络技术(北京)有限公司 Information pushing method and device
CN110457570A (en) * 2019-06-20 2019-11-15 拉扎斯网络科技(上海)有限公司 Text method for pushing, device, server and storage medium
CN110717806B (en) * 2019-09-06 2024-04-02 中国平安财产保险股份有限公司 Product information pushing method, device, equipment and storage medium
CN110750658B (en) * 2019-09-18 2022-12-09 平安科技(深圳)有限公司 Recommendation method of media resource, server and computer readable storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140337424A1 (en) * 2013-05-10 2014-11-13 Apple Inc. Authorizing Push Notifications for Websites
CN105931066A (en) * 2015-09-24 2016-09-07 中国银联股份有限公司 Transaction data processing method and device
CN108446963A (en) * 2018-03-30 2018-08-24 深圳春沐源控股有限公司 Resource information recommends method and resource information commending system
CN109711950A (en) * 2019-01-18 2019-05-03 深圳壹账通智能科技有限公司 Recommended method, device, storage medium and the computer equipment of marketing product
CN111371853A (en) * 2020-02-24 2020-07-03 中国银联股份有限公司 Resource information pushing method and device, server and storage medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113986486A (en) * 2021-10-15 2022-01-28 东华大学 Joint optimization method for data caching and task scheduling in edge environment

Also Published As

Publication number Publication date
CN111371853A (en) 2020-07-03
TWI803823B (en) 2023-06-01
TW202133080A (en) 2021-09-01
CN111371853B (en) 2022-12-20

Similar Documents

Publication Publication Date Title
WO2021169659A1 (en) Resource information pushing method and device, server, and storage medium
Mowlaei et al. Aspect-based sentiment analysis using adaptive aspect-based lexicons
Shu et al. Beyond news contents: The role of social context for fake news detection
Tang et al. Exploiting homophily effect for trust prediction
US20180253657A1 (en) Real-time credit risk management system
US7930242B2 (en) Methods and systems for multi-credit reporting agency data modeling
WO2022105525A1 (en) Method and apparatus for predicting user probability, and computer device
Liang et al. Analyzing credit risk among Chinese P2P-lending businesses by integrating text-related soft information
CN111061962A (en) Recommendation method based on user score analysis
CN111783039B (en) Risk determination method, risk determination device, computer system and storage medium
CN112861541A (en) Commodity comment sentiment analysis method based on multi-feature fusion
Ma et al. A credit risk assessment model of borrowers in P2P lending based on BP neural network
CN112232950A (en) Loan risk assessment method and device, equipment and computer-readable storage medium
Zhu et al. Loan default prediction based on convolutional neural network and LightGBM
CN117455529A (en) User electricity utilization characteristic image construction method and system based on big data technology
Duman Social media analytical CRM: a case study in a bank
CN115329207B (en) Intelligent sales information recommendation method and system
US11822609B2 (en) Prediction of future prominence attributes in data set
CN114897607A (en) Data processing method and device for product resources, electronic equipment and storage medium
JP2003323601A (en) Predicting device with reliability scale
CN110852392A (en) User grouping method, device, equipment and medium
US11989243B2 (en) Ranking similar users based on values and personal journeys
CN117743376B (en) Big data mining method, device and storage medium for digital financial service
CN116308615A (en) Product recommendation method and device, electronic equipment and storage medium
Jiang et al. A bank customer credit evaluation based on the decision tree and the simulated annealing algorithm

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21760394

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21760394

Country of ref document: EP

Kind code of ref document: A1