WO2020244468A1 - 金融产品推荐方法、装置、电子设备及计算机存储介质 - Google Patents

金融产品推荐方法、装置、电子设备及计算机存储介质 Download PDF

Info

Publication number
WO2020244468A1
WO2020244468A1 PCT/CN2020/093503 CN2020093503W WO2020244468A1 WO 2020244468 A1 WO2020244468 A1 WO 2020244468A1 CN 2020093503 W CN2020093503 W CN 2020093503W WO 2020244468 A1 WO2020244468 A1 WO 2020244468A1
Authority
WO
WIPO (PCT)
Prior art keywords
product
financial product
recommendation
financial
user
Prior art date
Application number
PCT/CN2020/093503
Other languages
English (en)
French (fr)
Inventor
杨凡
黄斐
Original Assignee
腾讯科技（深圳）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 腾讯科技（深圳）有限公司 filed Critical 腾讯科技（深圳）有限公司
Priority to JP2021541595A priority Critical patent/JP7430191B2/ja
Publication of WO2020244468A1 publication Critical patent/WO2020244468A1/zh
Priority to US17/337,284 priority patent/US20210287295A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/06—Asset management; Financial planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation

Definitions

  • This application relates to the field of computer technology, and in particular to a method, device, device, and computer-readable storage medium for recommending financial products.
  • Online financial management means that users can independently choose their own financial management methods based on their own economic conditions through the Internet financial management platform. As long as there is a network around them, users can search for financial projects they are interested in online anytime, anywhere, and enjoy it. A brand new financial management model for customers.
  • wealth management platforms usually have multiple financial products on the wealth management platform, providing users with multiple purchase options. These financial products may come from the same financial institution or from different financial institutions. In order to avoid the potential financial risks caused by the excessively high amount of a single product, it is necessary to limit the purchase limit of a single financial product. In this way, different users need to be allocated to the financial product, that is, the financial product needs to be diverted to make different financial products. Products correspond to different user groups.
  • the embodiments of the application provide a financial product recommendation method, device, equipment, and computer-readable storage medium, which can determine the user recommendation ratio according to the set parameters of each financial product, so as to improve the accuracy and stability of the financial product distribution, thereby ensuring Improve the security of user data.
  • the embodiment of the present application provides a method for recommending financial products, and the method includes:
  • the recommended financial product is determined according to the user recommendation ratio of each financial product to respond to the recommended financial product request.
  • the embodiment of the application provides a method for recommending financial products.
  • the method is executed by a server.
  • the server includes one or more processors and memories, and one or more programs, wherein the one or more The program is stored in the memory, the program may include one or more units each corresponding to a set of instructions, and the one or more processors are configured to execute the instructions; the method includes:
  • the recommended financial product is determined according to the user recommendation ratio of each financial product to respond to the recommended financial product request.
  • An embodiment of the application provides a financial product recommendation device, and the device includes:
  • the feature construction unit is configured to construct M-type product recommendation features of each financial product according to the historical data of the set parameters of each financial product in the N financial products, and N and M are both positive integers;
  • the feature integration unit is configured to respectively recommend features for each category of the product recommendation features of the M categories, and obtain comprehensive product recommendation features corresponding to the category;
  • the recommendation ratio determining unit is configured to determine the user recommendation ratio of each financial product according to the degree of deviation of various product recommendation features of each financial product with respect to the comprehensive product recommendation features corresponding to the category, and the user recommendation ratio is The proportion of recommended users corresponding to each financial product to all users;
  • the product recommendation unit is configured to determine the recommended financial product according to the user recommendation ratio of each financial product to respond to the recommended financial product request.
  • An embodiment of the present application provides an electronic device, including a memory and a processor
  • the memory is used to store a computer program
  • the processor is configured to implement the method described in the foregoing aspect when executing the program.
  • An embodiment of the present application provides a computer-readable storage medium that stores processor-executable instructions, and the processor implements the method described in the foregoing aspect when the processor executes the executable instructions.
  • the wealth management platform constructs product recommendation features based on the historical data of the set parameters of each financial product, and obtains the comprehensive product recommendation features of all financial products, and then according to the product recommendation features of each financial product relative to the corresponding category
  • the deviation degree of the product recommendation feature is integrated to determine the user recommendation ratio of the financial product, and finally the financial product is recommended to the user based on the user recommendation ratio of each financial product. Therefore, according to the parameters of each financial product, the proportion of user recommendations related to the parameters of the financial product can be determined, which improves the accuracy of the diversion of financial products, and the diversion of financial products will not be affected by non-self parameters, thereby improving financial management
  • the stability of the platform improves the security of user data.
  • FIGS. 1A-1B are schematic diagrams of application scenarios provided by embodiments of this application.
  • FIG. 2 is a schematic diagram of a display page of a wealth management platform provided by an embodiment of the application
  • FIG. 3 is a schematic flowchart of a method for recommending financial products according to an embodiment of the application
  • FIG. 4 is a schematic flowchart of a process of determining a user recommendation ratio provided by an embodiment of the application
  • FIG. 5 is a schematic flowchart of a process of determining a user recommendation ratio provided by an embodiment of the application
  • FIG. 6 is a schematic flowchart of a process of determining a user recommendation ratio provided by an embodiment of the application
  • FIG. 7 is a schematic flowchart of a process of determining a user recommendation ratio provided by an embodiment of the application.
  • FIG. 8 is a schematic structural diagram of a financial product recommendation device provided by an embodiment of the application.
  • FIG. 9 is a schematic structural diagram of an electronic device provided by an embodiment of the application.
  • Financial products refer to various carriers of the financial process, including currency, gold, foreign exchange, securities, etc. These financial products are the objects of purchase and sale in the financial market, and the prices of financial products are formed by the supply and demand parties through the principle of market competition. Such as interest rate or yield, the transaction is finally completed to achieve the purpose of financing.
  • financial products generally refer to products that can be traded through Internet finance.
  • Internet finance refers to a new type of finance that traditional financial institutions and Internet companies use Internet technology and information communication technology to achieve financial communication, payment, investment, and information intermediary services.
  • Business model, Internet finance is a new model and new business created to meet new demands on the level of network technology such as security and mobile. Among them, the circulation of Internet financial products is generally based on electronic money.
  • Wealth management platform or financial product platform, generally a platform provided by Internet companies for users to purchase financial products, such as trading platforms of various banks or trading platforms provided by other financial institutions.
  • traffic refers to users in the wealth management platform.
  • the wealth management platform usually requires Assigning different users to multiple wealth management products requires traffic distribution.
  • different user groups U 1 , U 2 and U 3 need to be assigned to different The financial products A, B and C of the corresponding, the users in the user group U 1 see the financial product A, the users in the user group U 2 see the financial product B, and the user group U 3 What users see is financial product C.
  • Object of embodiments of the present application is how to determine users U 1, U 2 and the ratio of the number of users U 3 in total user, user group U 1, U 2 and U 3 may be completely different from the user, there may be some The intersection.
  • Blockchain An encrypted, chain-type transaction storage structure formed by blocks.
  • Blockchain Network A series of nodes that incorporate new blocks into the blockchain through consensus.
  • wealth management platforms usually use the method of dividing the flow evenly among multiple financial products to distribute the flow. That is, the users corresponding to each financial product account for the same proportion of the total users. Then when recommending financial products to users, Recommendations will be made according to this set ratio.
  • financial products are distinguished by their advantages and disadvantages relative to users. This way, the flow is evenly divided among multiple financial products. The way of products will prevent better financial products from being seen by more users, which is obviously not good for the overall user experience. Therefore, how to allocate traffic more effectively and make the financial products recommended to users more accurate is a technical problem that needs to be solved urgently.
  • the embodiments of the present application provide a method for allocating the flow of financial products.
  • the wealth management platform constructs product recommendation features based on the historical data of the set parameters of each financial product, thereby obtaining comprehensive product recommendation features of all financial products , And then determine the user recommendation ratio of the financial product based on the deviation degree of the product recommendation feature of each financial product relative to the comprehensive product recommendation feature of the corresponding category, and finally recommend the financial product to the user based on the user recommendation ratio of each financial product.
  • the set parameters are the parameters of each financial product, so it can reflect the characteristics of the financial products to a certain extent, so that the product recommendation features constructed based on the set parameters and the comprehensive product recommendation features of all products deviate, and the user recommendation ratio is determined It is directly related to the parameters of financial products, so that the user recommendation ratio of each financial product is determined by the characteristics of each product itself. For example, the corresponding user recommendation ratio can be determined based on the quality of the product, then it can be better Financial products allocate a higher user recommendation ratio, so that better financial products can be seen by more users, thereby improving the accuracy of recommending financial products and thus improving the overall user experience. And according to the parameters of each financial product, the user recommendation ratio related to the financial product's own parameters can be determined, and the diversion of financial products will not be affected by non-self parameters, thereby improving the stability of the wealth management platform, thereby improving user data Security.
  • FIG. 1A is a schematic diagram of a scenario to which the embodiment of the invention can be applied.
  • the scenario may include a server 101, multiple terminals 102 (exemplarily showing terminals 102-1 to 102-L), and multiple financial Institution 103 (illustrating financial institutions 103-1 to financial institutions 103-P by way of example), where L and P are both positive integers, and the values of L and P represent the total number of users and financial institutions, respectively.
  • the embodiments of this application do not No restrictions.
  • the financial institution 103 may represent the equipment of each financial institution, each financial institution may provide one or more financial products, and the financial institution 103 may calculate and store the income data of each financial product.
  • the financial institution 103 may include one or more processors 1031, a memory 1032, and an I/O interface 1033 that interacts with a server.
  • the processor 1031 may calculate the income data of each financial product and store it in the memory 1032.
  • the income data of each financial product can be sent to the server 101 through the I/O interface 1033 that interacts with the server.
  • the server 101 may be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.
  • a cloud server packetaged with a financial product recommendation program
  • the user calls the financial product recommendation service in the cloud service through the terminal, so that the server deployed in the cloud calls the income data of the financial institution 103, and the server calls the packaged financial product
  • the recommended program determines the proportion of each financial product's traffic, and determines the financial product provided for the user according to the proportion, and pushes the financial product to the terminal to display the recommended financial product on the terminal’s display interface product.
  • the server 101 may include one or more processors 1011, a memory 1012, an I/O interface 1013 for interacting with a terminal, an I/O interface 1014 for interacting with a financial institution, and the like.
  • the server 101 can also be configured with a database 1015, which can be used to store user information, historical operation information and other user-related information of each user, and can also store information about financial products provided by financial institutions, such as income data, financial Institutional related information, etc.
  • the memory 1012 of the server 101 can store program instructions of the financial product flow distribution method provided by the embodiment of the present application. These program instructions can be used to realize the flow distribution of the financial product provided by the embodiment of the present application when executed by the processor 1011.
  • the steps of the method are to determine the flow allocated to the financial products provided by each financial institution based on the income data of each financial product. For example, the proportion of the flow allocated to each financial product can be finally determined, then when a new user joins In the case of a financial platform, the financial products that need to be presented to new users can be determined based on the determined ratio, so as to control the flow ratio of each financial product to maintain the above determined ratio.
  • the terminal 102 can be a terminal device such as a mobile phone, a personal computer (PC), or a tablet computer.
  • the terminal 102 can present a display page of a wealth management platform.
  • the terminal 102 can install an application (APP) provided by the wealth management platform to access
  • APP application
  • the display page of the wealth management platform can be opened in the APP provided by the wealth management platform; or the display page of the wealth management platform can be presented through the browser on the terminal 102; or, the display page of the wealth management platform can also be opened in other applications.
  • the wealth management platform can exist in the APP in the form of a light application or the wealth management platform can be provided to users as a function of the APP, such as the applet, official account or plug-in in WeChat.
  • the terminal 102 may include one or more processors 1021, a memory 1022, an I/O interface 1023 that interacts with the server 101, a display panel 1024, and the like.
  • the memory 1022 of the terminal 102 may store program instructions for implementing the functions of the wealth management platform. These program instructions can be used to implement the functions of the wealth management platform when executed by the processor 1021 and display the corresponding display page of the wealth management platform on the display panel 1024.
  • the server 101 when a new user registers for the account of the wealth management platform and enters the page of the wealth management platform, the server 101 will determine the financial product provided for the new user based on the predetermined traffic distribution of each financial product, and The financial product is pushed to the user, so that the user can view the financial product through the display interface of the wealth management platform.
  • Figure 2 it is a schematic diagram of a display page of the wealth management platform. On the display page of the wealth management platform, you can view the name 201 of the financial product assigned to the user, as shown in Figure 2 Product A" and the income data 202 of the financial product can also be displayed. The user can choose whether to subscribe for the financial product according to his own situation.
  • the account balance displayed when entering the display page of the wealth management platform for the first time is zero, and when the user subscribes for financial products, it will be shown on the right in Figure 2.
  • the account balance shown in the figure is not zero, and as time grows, the income gradually increases, and the account balance and cumulative income amount will also increase.
  • the network 104 may be a wired network or a wireless network.
  • the wireless network may be a mobile cellular network, or may be a WIreless-Fidelity (WIFI) network, of course, it may also be other possible networks. This application is implemented The example does not limit this.
  • FIG. 1B is a schematic diagram of a scenario to which the embodiment of the invention can be applied.
  • FIG. 1B shows that the network 104 in FIG. 1A is a blockchain network (the consensus node 1041-1 to the consensus node 1041- 5).
  • the type of blockchain network is flexible and diverse, for example, it can be any of a public chain, a private chain, or a consortium chain.
  • the electronic equipment of any business entity can access the blockchain network without authorization as a consensus of the blockchain Nodes, for example, the server 101 is mapped to the consensus node 1041 ⁇ 1 in the blockchain network, the financial institution 103-1 is mapped to the consensus node 1041 ⁇ 2 in the blockchain network, and the terminal 102-1 is mapped to the blockchain network
  • the electronic equipment under the jurisdiction of the business entity can access the blockchain network after being authorized, such as the server 101, the terminal 102 and the financial institution 103.
  • the terminal 102 when a user browses wealth management information on the terminal 102 (including the wealth management client), the terminal 102 initiates a request for recommending financial products to the server 101 (financial management platform), and the server 101 is mapped to the consensus nodes 1041-1 in the blockchain network. Among them, the terminal 102 generates a transaction corresponding to the update operation according to the financial product request.
  • the smart contract that needs to be called to implement the update operation and the parameters passed to the smart contract are specified in the transaction.
  • the transaction also carries the digital certificate of the terminal 102 and the signed number. Sign (for example, use the private key in the digital certificate of the terminal 102 to encrypt the summary of the transaction), and broadcast the transaction to the consensus node in the blockchain network.
  • the server 101 when the server 101 (consensus nodes 1041 ⁇ 1) receives the transaction, it verifies the digital certificate and digital signature carried in the transaction. After the verification is successful, it confirms whether the terminal 102 has the transaction authority according to the identity carried in the transaction. Any one of verification and authorization verification will result in transaction failure. After the verification is successful, the node’s own digital signature is signed (for example, the private key of the consensus node 1041 ⁇ 1 encrypts the transaction summary), and the smart contract integrated with financial product recommendations is called to obtain the setting parameters of each financial product.
  • each financial product determines the user recommendation ratio of each financial product, determine the financial product recommended by the user according to the user recommendation ratio of each financial product, and according to the user recommendation of each financial product Proportion and generate a transaction for the financial product recommended by the user, and broadcast the transaction to the consensus node in the blockchain network.
  • a financial institution 103-1 (consensus nodes 1041-2) receives a transaction, it verifies the digital certificate and digital signature carried by the transaction. After the verification is successful, the signing node's own digital signature (for example, the consensus node 101-2) The private key encrypts the transaction summary), broadcasts the signed transaction to the consensus node in the blockchain network, and continues the consensus.
  • the server 101 (financial management platform) receives the transaction again, it continues to verify the digital certificate and digital signature carried by the transaction. After the verification is successful, it signs the node’s own digital signature and returns it to the client. When the client receives the transaction, The digital certificate and digital signature carried by the transaction are verified. When the verification is passed and the number of successful transactions by consensus exceeds the consensus threshold, the reliability of the transaction result can be confirmed, that is, the user recommendation ratio of each financial product and the The safety of recommended financial products.
  • the user recommendation ratio of each financial product and the financial product recommended for users can be determined through the blockchain network, which can ensure the fairness and transparency of calculations .
  • users can make safe shopping according to the recommended financial products.
  • the method provided in the embodiment of the present application is not limited to be used in the application scenarios shown in FIGS. 1A-1B, and may also be used in other possible application scenarios, which is not limited by the embodiment of the present application.
  • the functions that can be implemented by each device in the application scenarios shown in FIGS. 1A-1B will be described together in the subsequent method embodiments, and will not be repeated here.
  • FIG. 3 is a schematic flowchart of a method for recommending financial products according to an embodiment of this application.
  • the method can be executed by an electronic device, for example, by the server in FIG. 1A.
  • Step 301 Receive the recommended financial product request from the client, and construct the M-type product recommendation feature of each financial product according to the historical data of the setting parameters of each of the N financial products.
  • the client when a user browses financial related information on the client, the client automatically generates a recommended financial product request and sends a recommended financial product request to the server. After the server receives the recommended financial product request, it will follow the settings of each financial product
  • the historical data of the parameters is used to construct the M-type product recommendation features of each financial product, so that subsequent processing can be performed according to the product recommendation features.
  • multiple financial products may exist in the wealth management platform, and the N financial products may be all financial products in the wealth management products, or may be part of all financial products.
  • a wealth management platform includes 5 financial products, then N financial products can refer to these 5 financial products; or, when one of the 5 financial products, financial product A, uses a fixed user recommendation ratio, for example, its user recommendation ratio is 1/5, then N financial products can refer to the remaining 4 financial products except financial product A, and the total amount of user recommendation ratios that can be allocated for these 4 financial products is 4/5.
  • the setting parameter can be any possible parameter of a financial product, such as a financial product that focuses on returns, and the setting parameter can be a return rate, for example, a financial product that focuses on risk, and the setting parameter can be a risk rate, etc.
  • the setting parameter can be a return rate, for example, a financial product that focuses on risk
  • the setting parameter can be a risk rate, etc.
  • users subscribe for financial products they generally care about the return rate of financial products. Therefore, the following will take setting parameters as an example to introduce the financial product recommendation method of the embodiment of the present application.
  • the rate of return can be 10,000 shares, 7-day annualized, 30-day annualized, or annualized rate of return, while for non-monetary fund financial products, the rate of return can be the most recent Indicators such as January earnings or recent March earnings.
  • the rate of return of each financial product can be calculated by the wealth management platform based on the income data of each financial product; or, since each financial institution will calculate the rate of return and other indicators for its own financial products, in order to prevent the financial platform from There is a deviation between the calculation method and the calculation of the financial institution at the time, making the rate of return different from the rate of return calculated by the financial institution.
  • the wealth management platform can also obtain data such as the rate of return directly from the financial institution, so as to avoid the large number of users and the amount of financial products. Large and inappropriate shunt calculations lead to huge consumption of computing resources and are prone to errors. It also saves a certain amount of calculation for the financial management platform and reduces the computing pressure on the server.
  • the server of the wealth management platform can periodically obtain the rate of return data from the financial institution. For example, if the rate of return is updated once a day, the server can obtain the rate of return from the financial institution regularly every day Or, if the data on the rate of return is updated once a month, the server can obtain the data on the rate of return from financial institutions regularly every month.
  • the server may also adopt a pre-arranged agreement with the financial institution to obtain data on the rate of return from the electronic equipment of the financial institution.
  • the electronic device of the financial institution After the electronic device calculates the return rate data, the electronic device of the financial institution provides the return rate data to the server.
  • the server obtains the data of the rate of return, the data of the rate of return can be stored in a unified manner, for example, stored in a database. When the data of the rate of return needs to be used, it can be directly read from the database.
  • the server can construct the M-type product recommendation feature of each financial product according to the historical data of the setting parameters of each financial product in the N financial products.
  • N and M are both positive integers.
  • the recommended features of M category products include at least one of the following features:
  • the recommended features of the M-category product may be any one of the recommended features for the above-mentioned products, or may be a combination of the recommended features of multiple types of products.
  • the process of constructing product recommendation features is independent.
  • the first set time period is the statistical time period T 1 of the set parameter, and the length of T 1 can be set according to the situation. For example, it may be the last month, or the last two months, etc.
  • the embodiment of the present application does not limit this.
  • construct the product recommendation feature of the financial product which can be based on the set parameters of the financial product in each sub-time period within the first set time period.
  • the data value of and the weight value corresponding to each sub-period to obtain the average value of the set parameter of each financial product in the first set period of time.
  • the average value of the set parameter in the first set time period can be obtained in the above-mentioned manner.
  • the weight value corresponding to the sub-period can be used to distinguish long-term or short-term data. For example, if you pay more attention to long-term data, you can set the weight value of the sub-period farther from the current time to be higher. On the contrary Yes, if you pay more attention to short-term data, you can set the weight value of the sub-period closer to the current time higher.
  • the second set time period is the statistical time period T 2 of the set parameter
  • the length of T 2 may be equal to T 1 is the same but may be different from T 1 . Since the volatility in a short period of time may not be very large, the length of T 2 can be set to a longer period of time, for example, it can be set to the most recent month, the most recent six months, or the most recent year.
  • the volatility of the financial product's setting parameter in each sub-time period can be obtained according to the data value of the financial product's setting parameter in each sub-time period.
  • the volatility of each financial product represents the degree of change in the rate of return of the financial product. Volatility can be obtained through the following process:
  • the degree of deviation of the change rate corresponding to each sub-time period of the financial product from the average change rate in the second set time period.
  • the average rate of change is the mean value of the rate of change in the second set time period, and the degree of deviation can be represented by variance or standard deviation.
  • the volatility of the set parameters of the financial product in each sub-period is obtained.
  • degree of deviation represented by the variance the volatility can be expressed as the square root of the variance of the ratio of T 2.
  • the volatility of each sub-time period and the corresponding weight value of each sub-time period can be obtained according to the set parameters of the financial product
  • the average value of the volatility of the set parameter of each financial product in the second set time period can be obtained for each financial product.
  • the combined feature may be a combination of volatility and set parameters.
  • the volatility of the rate of return can be lower when the rate of return continues to be low, but the financial product with the lower rate of return will obviously not be the better financial product.
  • the yield also needs to be considered, that is, the combination characteristics can be constructed based on the volatility and the yield.
  • the value of the combination feature can be positively correlated with the rate of return, and negatively correlated with the rate of return, which means that the higher the rate of return and the smaller the volatility, the financial product is the better product.
  • the average value of the combined feature of each financial product in the second set time period can be obtained.
  • a certain weight value can also be assigned to each sub-time period.
  • the method of assigning the weight value please refer to the description of the calculation of the average value of the set parameter in the first set time period.
  • Step 302 For each category of product recommendation features in the M category product recommendation features, obtain comprehensive product recommendation features corresponding to the category.
  • the product recommendation feature is used to characterize the feature of one financial product among the N financial products
  • the comprehensive product recommendation feature is used to characterize the overall feature of the N financial product.
  • the comprehensive product recommendation feature can be represented by the mean and variance of the product recommendation feature. Then, after obtaining the product recommendation features of each financial product through the process of step 301, the comprehensive product recommendation features of N financial products can be obtained by calculating the mean and variance of the product recommendation features of each financial product.
  • Step 303 Determine the user recommendation ratio of each financial product according to the deviation degree of the various product recommendation features of each financial product relative to the comprehensive product recommendation features corresponding to the category.
  • the user recommendation ratio is the ratio of recommended users corresponding to each financial product to all users.
  • the user recommendation ratio of each financial product can be determined according to the degree of deviation of the product recommendation feature from the determined comprehensive product recommendation feature.
  • the deviation degree can refer to the absolute deviation degree, that is, the difference between the product recommendation characteristics of a financial product and the average value of the product recommendation characteristics of N financial products; or the deviation degree can also refer to the relative deviation degree, that is, the absolute deviation degree.
  • the ratio of the value of to the variance can refer to the absolute deviation degree, that is, the difference between the product recommendation characteristics of a financial product and the average value of the product recommendation characteristics of N financial products; or the deviation degree can also refer to the relative deviation degree, that is, the absolute deviation degree.
  • the ratio of the value of to the variance is the ratio of the value of to the variance.
  • the user recommendation ratio of each financial product can be obtained based on the deviation degree, wherein the user recommendation ratio of each financial product is positively correlated with the deviation degree.
  • the user recommendation sub-proportions corresponding to the various product recommendation features can be obtained according to the deviation degree corresponding to the various product recommendation features of each financial product , And then calculate the final user recommendation ratio based on the user recommendation weights of various product recommendation features.
  • the process of respectively obtaining the user recommendation sub-proportions corresponding to various product recommendation features is the same as the calculation process described above when the M category product recommendation features include only one of the above product recommendation features. Therefore, you can refer to the above description. This will not be repeated here.
  • the sum of the user recommendation weights of various product recommendation features is 100%, so the user recommendation weights of various product recommendation features can be obtained through an optimal solution process.
  • fixed user recommendation weights may also be set for various product recommendation features, which is not limited in the embodiments of the present application.
  • f i is the user recommendation ratio of the i-th financial product
  • ⁇ j is the user recommendation weight of the j-th product recommendation feature
  • the above calculation formula can be used as the objective function, and the sum of the user recommendation weights of various product recommendation features is 100% as the constraint condition to calculate the optimal user recommendation weight.
  • the constraint conditions for example, the user recommendation ratio of all financial products is a fixed value.
  • steps 301 to 303 in the embodiment of the present application may be repeated multiple times, for example, may be repeated periodically, or may be repeated at After the change value of the setting parameter is greater than or equal to a certain threshold, the user recommendation ratio is determined again.
  • the rate of return is generally updated periodically, such as once a day or once a month. Therefore, the corresponding determination of the user recommendation ratio can be done once a day, or It is done once a month.
  • Step 304 Recommend financial products for users according to the user recommendation ratio of each financial product, in response to the request for recommending financial products.
  • the user after the user recommendation ratio of each financial product, the user can recommend financial products based on the user recommendation ratio of each financial product.
  • the recommendation is based on the determined user recommendation ratio of each financial product, so that the number of recommended users corresponding to each financial product accounts for the ratio of all users to the user recommendation ratio of each financial product Close to or the same.
  • the user recommendation ratio used is generally the user recommendation ratio obtained last time.
  • the server can send the status data of the financial product recommended by the user to the user.
  • the financial product recommended by the user can be displayed on the display page.
  • the status data of the product such as the display interface shown in Figure 2.
  • the status data may include data such as the name of the financial product, the rate of return, the user's subscription status, and the user's income status.
  • the following will show several examples of obtaining the user recommendation ratio, in which the setting parameter is based on the profit rate.
  • the process of determining the user recommendation ratio is introduced by taking the product recommendation feature as the average value of the return rate in the first set time period as an example.
  • Step 401 Obtain product recommendation features of a single financial product.
  • the product recommendation feature is the average value of the rate of return in the first set time period, where the first set time period is the statistical time period T 1 of the set parameters, and the length of T 1 can be determined according to the situation
  • the setting for example, may be the last month, or the last two months, etc.
  • the embodiment of the present application does not limit this.
  • a sub-time period can be set to one day, then the average value of the set parameter in the first set time period can be calculated as shown in formula (2) :
  • Step 402 Obtain comprehensive product recommendation features of N financial products.
  • the calculation method of the comprehensive product recommendation feature can be as shown in formulas (3) and (4):
  • the mean and standard deviation can also be used as the comprehensive product recommendation features.
  • other possible adoption numbers can also be used as the comprehensive product recommendation features. No restrictions.
  • Step 403 Obtain the relative deviation degree between the product recommendation feature of each financial product and the comprehensive product recommendation feature.
  • the deviation degree here takes the relative deviation degree as an example.
  • the calculation method of the relative deviation between the product recommendation features of each financial product and the comprehensive product recommendation features can be as shown in formula (5):
  • k ai is the relative deviation between the i-th feature Recommendations integrated financial product and product feature recommendation, where the subscript a represents a characteristic corresponding to the recommended product yields mean T 1 as in.
  • Step 404 Determine the user recommendation ratio based on the relative deviation corresponding to each financial product.
  • ⁇ is the distribution coefficient
  • ⁇ is used to represent the proportion of the total flow that can be allocated, and ⁇ can be set as a fixed value or a variable value.
  • the return rate of each financial product is high or low, so it is possible that the relative deviation of the financial product is negative. Therefore, in order to ensure the minimum relative deviation, that is, the financial product with the furthest negative deviation can be allocated
  • the value of ⁇ can be set to a value that satisfies the following conditions, as shown in formula (7):
  • the process of calculating the user recommendation ratio is similar to the above process, that is, the product recommendation feature is replaced with the set parameter in the second set time period
  • the average value of the volatility within the range is sufficient. Therefore, when the product recommendation feature is the average value of the volatility of the set parameter in the second set time period, the process of calculating the user recommendation ratio can be referred to the above description. This will not be repeated here.
  • the process of determining the user recommendation ratio is introduced by taking the product recommendation feature as an example of the mean value of the combined feature in the second set time period.
  • Step 501 Obtain the volatility of the yield of a single financial product.
  • the product recommendation feature is the average value of the combined feature in the second set time period, where the second set time period is the statistical time period T 2 of the set parameters, and the length of T 2 can be based on the situation
  • the setting may be, for example, the last month, the last six months, or the last year, etc.
  • the embodiment of the present application does not limit this.
  • the combination feature can be a combination feature composed of the return rate and the volatility of the return rate. Therefore, before obtaining the mean value of the combination feature, the volatility of the return rate of each financial product needs to be obtained first.
  • the relative change characteristics of the financial can be constructed based on the rate of return of the financial product in the second set time period.
  • the calculation method of the relative change characteristics can be as follows Formula (8) shows:
  • Is the data value of the i-th financial product in the t-th sub-period, compared to the rate of change of the data value in the t-1-th sub-period, t 1, 2 , 3...T 2 .
  • the volatility of the rate of return in the second set time period can be understood as the degree of dispersion of the rate of change in the second set time period. Therefore, it can be calculated as follows: The mean and variance of are shown in formulas (9) and (10):
  • ⁇ c is The variance in the second set time period.
  • the volatility of the return rate of the t-th sub-period is calculated based on the data from the t-th sub-period to T 2 sub-periods before the t-th sub-period. For example, if the statistical time period is half a year, then the volatility rate of the day is calculated based on the data of the current day and half a year before the current day, and the volatility rate of yesterday is based on the data of yesterday and the half year before yesterday. Basic calculations.
  • Step 502 Construct a combination feature of each financial product based on the volatility of the yield.
  • the volatility of the rate of return may also be lower, but the financial product with a lower rate of return will obviously not be a better financial product. Therefore, the user recommendation ratio of the financial product is determined.
  • the rate of return needs to be considered at the same time, that is, the combination characteristics can be constructed based on the volatility and the rate of return. Among them, the value of the combination feature can be positively correlated with the rate of return, and negatively correlated with the rate of return, which means that the higher the rate of return and the smaller the volatility, the financial product is the better product. Therefore, the combination feature can be passed as Formula (12) expresses:
  • Step 503 Obtain product recommendation features of a single financial product.
  • the product recommendation feature is the average value of the combined feature within the second set time period, where the average value of the combined feature within the second set time period can be calculated as shown in formula (13):
  • Is the mean value of the combination characteristics of the i-th financial product in the second set time period, i 1, 2, 3...N.
  • a sub-time period can be set to one day.
  • Step 504 Obtain comprehensive product recommendation features of N financial products.
  • the calculation method of the comprehensive product recommendation feature can be as shown in formulas (14) and (15):
  • the mean and standard deviation can also be used as the comprehensive product recommendation features.
  • other possible adoption numbers can also be used as the comprehensive product recommendation features. No restrictions.
  • Step 505 Obtain the relative deviation degree between the product recommendation feature of each financial product and the comprehensive product recommendation feature.
  • the deviation degree here takes the relative deviation degree as an example.
  • the calculation method of the relative deviation between the product recommendation features of each financial product and the comprehensive product recommendation features can be as shown in formula (16):
  • k bi is the relative deviation between the product recommendation feature of the i-th financial product and the comprehensive product recommendation feature, where the subscript b indicates that the corresponding product recommendation feature is the mean value of the combined features in T 2 .
  • Step 506 Determine the user recommendation ratio based on the relative deviation corresponding to each financial product.
  • the user recommendation ratio should be higher. In this way, the number of users who can see the financial product is greater, and the overall user experience can be improved, and the user's stickiness to the financial platform can be improved. Therefore, the calculation method of the user recommendation ratio can be as shown in formula (17):
  • ⁇ is the distribution coefficient
  • ⁇ is used to represent the proportion of the total flow that can be allocated, and ⁇ can be set as a fixed value or a variable value.
  • the value of ⁇ can be set to a value that satisfies the following conditions, as shown in formula (18):
  • the product recommendation feature includes the average value of the return rate in the first set time period and the average value of the combined feature in the second set time period to introduce the process of determining the user recommendation ratio.
  • the average value of the return rate in the first time period is the recommended feature of the first product
  • the average of the combined features in the second set time period is the recommended feature of the second product.
  • Step 601 Determine the user recommendation sub-proportion corresponding to the first product recommendation feature according to the first product recommendation feature.
  • Step 602 Determine the user recommendation sub-proportion corresponding to the second product recommendation feature according to the second product recommendation feature.
  • step 601 and step 602 can be performed at the same time or sequentially, for example, step 601 is performed first, and then performed Step 602, Figure 6 takes this as an example, or, step 602 is performed first, and then step 601 is performed.
  • Step 603 Obtain the user recommendation ratio of financial products based on the user recommendation sub-proportions corresponding to various product recommendation features and the user recommendation weights corresponding to various product recommendation features.
  • the user recommendation weights corresponding to various product recommendation features may be fixed weights, or may be calculated through an optimal solution method.
  • the calculation method of the user recommendation ratio can be as shown in formulas (19) and (20):
  • f i is the user recommendation ratio of the i-th financial product
  • ⁇ a is the user recommendation weight corresponding to the first product recommendation feature
  • ⁇ b is the user recommendation weight corresponding to the second product recommendation feature
  • the attractiveness of various financial products to users may be different, and these attractiveness are not only brought about by the return rate or the stability of the return rate, but may also be related to Other factors related to financial products, such as the brand awareness of financial products and the popularity of product managers, will affect whether users subscribe for financial products.
  • the attractiveness of financial products can be measured by the user conversion rate of the financial product.
  • the conversion rate of financial products to users can also be taken into consideration, that is, the user conversion rate can be combined with any of the above-mentioned M-type product recommendation features to construct a new combined product recommendation feature. As shown in FIG. 7, the following takes the combination of the average value of the user conversion rate and the profit rate in the first set time period as an example to introduce the process of determining the user recommendation ratio.
  • Step 701 Obtain the user conversion rate of each financial product.
  • the user conversion rate refers to the proportion of the number of users who actually use the financial product among the recommended users corresponding to the financial product. Then the user conversion rate can be calculated as shown in formula (21):
  • ⁇ i is the user conversion rate of the i-th financial product
  • u i is the ratio of the number of users who actually use the i-th financial product to all users.
  • the number of users who actually use the i-th financial product can also be directly compared with the number of users who use the i-th financial product.
  • the ratio of the number of recommended users corresponding to financial products is used as the user conversion rate.
  • Step 702 Constructing each product recommendation feature based on the user conversion rate.
  • the combination feature can be expressed by the following formula (22):
  • Step 703 Obtain comprehensive product recommendation features of N financial products.
  • Step 704 Obtain the relative deviation degree between the product recommendation feature of each financial product and the comprehensive product recommendation feature.
  • Step 705 Determine the user recommendation ratio based on the relative deviation corresponding to each financial product.
  • Steps 703 to 705 are similar to steps 402 to 404, or steps 504 to 507. Therefore, for the steps 703 to 705, please refer to the descriptions of steps 402 to 404 or steps 504 to 507, which will not be repeated here.
  • the product recommendation feature is constructed based on the historical data of the setting parameters of each financial product, thereby obtaining the comprehensive product recommendation feature of all financial products, and then according to the product recommendation feature of each financial product relative to According to the deviation degree of the comprehensive product recommendation characteristics of the corresponding category, the user recommendation ratio of the financial product is determined, and finally the financial product is recommended to the user based on the user recommendation ratio of each financial product.
  • the setting parameters are the parameters of each financial product itself.
  • the determined user recommendation ratio is directly related to the parameters of the financial product, and each The user recommendation ratio of financial products is determined by the characteristics of each product. For example, the corresponding user recommendation ratio can be determined based on the quality of the product, and then a higher user recommendation ratio can be assigned to better financial products, so that more Excellent financial products can be seen by more users, thereby improving the overall user experience.
  • the financial product recommendation method of the embodiments of this application not only can the limit on the same product be met to ensure potential financial risks, but also higher quality financial products can be allocated to more traffic as much as possible, and the accuracy of recommendation can be improved.
  • the user experience can also prevent financial product providers from penetrating the flow distribution strategy through high and short-term revenue penetration, improving the stability of the platform and guiding financial asset companies to provide users with better assets.
  • it can also improve the efficiency of platform traffic usage.
  • an embodiment of the present application further provides a financial product recommending device 80.
  • the device may be, for example, the server shown in Fig. 1A, and the device includes:
  • the feature construction unit 801 is configured to receive a request for recommended financial products from the client; respectively, according to the historical data of the set parameters of each financial product in the N financial products, construct the recommended features of the M-type products of each financial product, and both N and M Is a positive integer;
  • the feature synthesis unit 802 is configured to obtain a comprehensive product recommendation feature corresponding to the category for each category of product recommendation features in the M category product recommendation features respectively;
  • the recommendation ratio determining unit 803 is configured to determine the user recommendation ratio of each financial product according to the degree of deviation of the various product recommendation features of each financial product with respect to the comprehensive product recommendation features corresponding to the category, and the user recommendation ratio The ratio of recommended users to all users corresponding to each financial product;
  • the product recommendation unit 804 is configured to determine a recommended financial product according to the user recommendation ratio of each financial product to respond to the recommended financial product request.
  • the recommended features of M category products include at least one of the following features:
  • the feature construction unit 801 is configured to: respectively according to the data value of each sub-time period within the first set time period according to the setting parameter of each financial product, and the corresponding data value of each sub-time period The weight value obtains the average value of the setting parameter of each financial product in the first setting time period.
  • the feature construction unit 801 is configured to obtain the setting of each financial product according to the data value of each sub-time period within the second set time period of the setting parameter of each financial product.
  • the volatility of the parameter in each sub-time period respectively according to the volatility of the set parameters of each financial product in each sub-time period, and the weight value corresponding to each sub-time period, Obtain the average value of the volatility of the setting parameter of each financial product in the second setting time period.
  • the feature construction unit 801 is configured to obtain the setting of each financial product according to the data value of each sub-time period within the second set time period of the setting parameter of each financial product.
  • the volatility of the parameter in each sub-time period ; construct the volatility of each financial product according to the setting parameter of each financial product and the volatility of the setting parameter of each financial product in each sub-time period. And obtaining the combined feature of each financial product in the second set time period.
  • the feature constructing unit 801 is configured to: obtain the data value of the setting parameter of each financial product in each of the sub-periods, compared with the data value of the previous sub-period of the sub-period Obtain the degree of deviation of the change rate corresponding to each sub-time period of each financial product from the average change rate in the second set time period; according to each financial product The degree of deviation corresponding to a sub-time period is obtained, and the volatility of the set parameter of each financial product in each sub-time period is obtained.
  • the recommendation ratio determining unit 803 is configured to: obtain the deviation degree of the various product recommendation features of each financial product relative to the comprehensive product recommendation features corresponding to the category; The deviation degree corresponding to the recommended feature determines the user recommendation ratio of each financial product; wherein the user recommendation ratio of each financial product is positively correlated with the deviation degree.
  • the recommendation ratio determining unit 803 is configured to: obtain the user recommendation sub-proportions corresponding to various product recommendation features according to the deviation values corresponding to the various product recommendation features of each financial product; User recommendation weights corresponding to various product recommendation features; among them, the sum of user recommendation weights corresponding to various product recommendation features is 100%; according to the user recommendation sub-proportions corresponding to the various product recommendation features and the various product recommendation features According to the user recommendation weight, the user recommendation ratio of each financial product is obtained.
  • the device further includes a conversion rate obtaining unit 805 configured to obtain a user conversion rate of each financial product, where the user conversion rate is determined by the number of users who actually use the financial product among the recommended users corresponding to the financial product. As a percentage of the total number of recommended users;
  • the feature construction unit 801 is further configured to obtain the average value of the setting parameter of each financial product in the first set time period according to the historical data of the setting parameter of each financial product, and according to each The average value of the set parameter of a financial product in the first set time period and the user conversion rate construct the product recommendation feature of each financial product.
  • the device further includes a data sending unit 806 configured to send the status data of the financial product recommended for the user to the user, so that after logging in to the user's corresponding account through the user equipment, the user equipment is displayed on the display page
  • a data sending unit 806 configured to send the status data of the financial product recommended for the user to the user, so that after logging in to the user's corresponding account through the user equipment, the user equipment is displayed on the display page
  • the above display is the status data of the financial product recommended by the user, and the status data includes the name and rate of return of the financial product.
  • the device can be used to execute the methods shown in the embodiments shown in FIGS. 3-7. Therefore, for the functions that can be realized by the functional modules of the device, please refer to the examples of the embodiments shown in FIGS. 3-7 Description, not much to repeat. Among them, the conversion rate acquiring unit 805 and the data sending unit 806 are not mandatory functional units, so they are shown in dotted lines in FIG. 8.
  • an embodiment of the present application further provides an electronic device 90 that may include a memory 901 and a processor 902.
  • the memory 901 is configured to store a computer program executed by the processor 902.
  • the memory 901 may mainly include a program storage area and a data storage area.
  • the program storage area may store an operating system, an application program required by at least one function, etc.; the data storage area may store data created according to the use of an electronic device, etc.
  • the processor 902 may be a central processing unit (Central Processing Unit, CPU), or a digital processing unit or the like.
  • the embodiment of the present application does not limit the connection medium between the foregoing memory 901 and the processor 902.
  • the memory 901 and the processor 902 are connected through a bus 903, and the bus 903 is represented by a thick line in FIG. 9.
  • the connection mode between other components is only for schematic illustration, and is not Limited.
  • the bus 903 can be divided into an address bus, a data bus, a control bus, and so on. For ease of representation, only one thick line is used in FIG. 9, but it does not mean that there is only one bus or one type of bus.
  • the memory 901 may be a volatile memory (Volatile Memory), such as a random access memory (Random Access memory, RAM); the memory 901 may also be a non-volatile memory (Non-volatile Memory), such as a read memory, flash memory (flash memory), Hard Disk Drive (HDD) or Solid State Drive (SSD), or memory 901 can be used to carry or store desired program codes in the form of instructions or data structures and can be stored by a computer Any other media taken, but not limited to this.
  • the memory 901 may be a combination of the above-mentioned memories.
  • the processor 902 is configured to execute the method executed by the device in the embodiments shown in FIGS. 3 to 7 when calling the computer program stored in the memory 901.
  • various aspects of the method provided in this application can also be implemented in the form of a program product, which includes program code, and when the program product runs on an electronic device, the program code is used for
  • the electronic device is allowed to execute the steps in the method according to various exemplary embodiments of the application described above in this specification.
  • the electronic device can execute the steps performed by the device in the embodiments shown in FIGS. 3 to 7 Methods.
  • the program product may adopt any combination of one or more computer-readable storage media.
  • the computer-readable storage medium may be a readable signal medium or a readable storage medium.
  • the readable storage medium may be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or a combination of any of the above, for example.
  • examples of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), personal read memory (ROM), erasable Programmable readable memory (EPROM or flash memory), optical fiber, portable compact disk read memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • the server constructs product recommendation features based on the historical data of the setting parameters of each financial product, and obtains the comprehensive product recommendation features of all financial products.
  • the product recommendation features of each financial product are compared to the comprehensive product of the corresponding category.
  • the degree of deviation of the recommended features determines the user recommendation ratio of financial products, and recommends financial products to users according to the user recommendation ratio of each financial product.
  • the user recommendation ratio related to the parameters of the financial product can be determined, which improves the accuracy of financial product diversion and the security of user data.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • Operations Research (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

一种金融产品推荐方法、装置、设备及计算机可读存储介质，该方法包括：接收客户端的推荐金融产品请求，分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数（301）；分别针对M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征（302）；根据每一个金融产品的各类产品推荐特征相对于与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例（302），用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；根据各金融产品的用户推荐比例确定推荐金融产品，以响应推荐金融产品请求（304）。

Description

金融产品推荐方法、装置、电子设备及计算机存储介质
相关申请的交叉引用
本申请基于申请号为201910490545.8、申请日为2019年06月06日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及计算机技术领域，特别涉及一种金融产品推荐方法、装置、设备及计算机可读存储介质。
背景技术
在线理财是指用户根据自身经济情况，通过网络的理财平台自主选择适合自己的理财方式进行理财，只要身边有网络，用户就可以随时随地的在网上寻找他们感兴趣的理财项目，享受足不出户的全新理财模式。
例如，理财平台作为金融产品的重要获取途径，通常理财平台上都拥有着多个金融产品，为用户提供多种购买选择，这些金融产品可能来自相同的金融机构，也可能来自不同的金融机构。为了避免单个产品的金额过高而带来的潜在的金融风险，需要对单个金融产品有申购上限限制，这样就需要给金融产品分配不同的用户，即需要对金融产品进行分流，使得不同的金融产品对应着不同的用户群。
在相关技术中，不恰当地分流会影响理财平台的稳定性和用户数据的安全性。
发明内容
本申请实施例提供一种金融产品推荐方法、装置、设备及计算机可读 存储介质，能够依据各金融产品的设定参数确定用户推荐比例，以提高金融产品分流的精准度、稳定性，从而保障了用户数据的安全性。
本申请实施例提供一种金融产品推荐方法，所述方法包括：
接收客户端的推荐金融产品请求；
分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数；
分别针对所述M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征；
根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，所述用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；
根据各金融产品的用户推荐比例确定推荐金融产品，以响应所述推荐金融产品请求。
本申请实施例提供一种金融产品推荐方法，所述方法由服务器执行，所述服务器包括有一个或多个处理器以及存储器，以及一个或一个以上的程序，其中，所述一个或一个以上的程序存储于存储器中，所述程序可以包括一个或一个以上的每一个对应于一组指令的单元，所述一个或多个处理器被配置为执行指令；所述方法包括：
接收客户端的推荐金融产品请求；
分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数；
分别针对所述M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征；
根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，所述用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；
根据各金融产品的用户推荐比例确定推荐金融产品，以响应所述推荐金融产品请求。
本申请实施例提供一种金融产品推荐装置，所述装置包括：
特征构建单元，配置为分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数；
特征综合单元，配置为分别针对所述M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征；
推荐比例确定单元，配置为根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，所述用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；
产品推荐单元，配置为根据各金融产品的用户推荐比例确定推荐金融产品，以响应所述推荐金融产品请求。
本申请实施例提供一种电子设备，包括存储器和处理器；
其中，所述存储器用于存储计算机程序；
所述处理器用于执行所述程序时实现上述方面所述的方法。
本申请实施例提供一种计算机可读存储介质，存储有处理器可执行指令，处理器执行所述可执行指令时实现上述方面所述的方法。
本申请实施例中，理财平台基于各金融产品的设定参数的历史数据，构建产品推荐特征，并获取所有金融产品的综合产品推荐特征，进而根据各金融产品的产品推荐特征相对于对应类别的综合产品推荐特征的偏离度，来确定该金融产品的用户推荐比例，最终基于各金融产品的用户推荐比例为用户推荐金融产品。从而根据各金融产品自身的参数，即可确定出与金融产品的自身参数相关的用户推荐比例，提高了金融产品分流的精准度，且金融产品分流不会受到非自身参数的影响，从而提高理财平台的稳定性， 进而提高用户数据的安全性。
附图说明
图1A-1B为本申请实施例提供的应用场景的示意图；
图2为本申请实施例提供的理财平台的一种显示页面的示意图；
图3为本申请实施例提供的金融产品推荐方法的流程示意图；
图4为本申请实施例提供的用户推荐比例的确定过程的流程示意图；
图5为本申请实施例提供的用户推荐比例的确定过程的流程示意图；
图6为本申请实施例提供的用户推荐比例的确定过程的流程示意图；
图7为本申请实施例提供的用户推荐比例的确定过程的流程示意图；
图8为本申请实施例提供的金融产品推荐装置的一种结构示意图；
图9为本申请实施例提供的电子设备的一种结构示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚明白，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
为便于理解本申请实施例提供的技术方案，这里先对本申请实施例使用的一些关键名词进行解释：
1)金融产品：金融产品指资金融通过程的各种载体，它包括货币、黄金、外汇、有价证券等，这些金融产品就是金融市场的买卖对象，供求双 方通过市场竞争原则形成金融产品价格，如利率或收益率，最终完成交易，达到融通资金的目的。本申请实施例中，金融产品一般是指能够通过互联网金融交易的产品，互联网金融是指传统金融机构与互联网企业利用互联网技术和信息通信技术实现资金融通、支付、投资和信息中介服务的新型金融业务模式，互联网金融是在实现安全以及移动等网络技术水平上，为适应新的需求而产生的新模式及新业务。其中，互联网金融产品的流通一般是以电子货币为基础的。
2)理财平台：或称金融产品平台，一般是互联网企业提供的用于用户购买金融产品的平台，例如各个银行的交易平台或者其他理财机构提供的交易平台。
3)流量分配：本申请实施例中，流量即是指理财平台中的用户。在同一理财平台中，一般会有众多的金融产品，为了避免单个金融产品的金额过高带来的潜在的金融风险，会对单个理财产品(金融产品)有申购上限限制，因此理财平台通常需要给多个理财产品分配不同的用户，即需要进行流量分配，例如有3个金融产品，即A、B和C时，则需要将不同的用户群U 1、U 2和U 3对应分配给不同的金融产品A、B和C，相对应的，用户群U 1中的用户所看到的就是金融产品A，用户群U 2中的用户所看到的就是金融产品B，用户群U 3中的用户所看到的就是金融产品C。本申请实施例的目的主要在于如何确定用户群U 1、U 2和U 3中用户的数量占总用户的比例，用户群U 1、U 2和U 3中用户可以完全不同，也可以有一定的交集。
4)区块链(Blockchain)：由区块(Block)形成的加密的、链式的交易的存储结构。
5)区块链网络(Blockchain Network)：通过共识的方式将新区块纳入区块链的一系列的节点的集合。
另外，本文中术语“和/或”，仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存 在A和B，单独存在B这三种情况。另外，本文中字符“/”，在不做特别说明的情况下，一般表示前后关联对象是一种“或”的关系。
相关技术中，理财平台通常采用将流量均分给多个金融产品的方式来进行流量分配，即各金融产品所对应的用户所占总用户的比例相同，那么在为用户推荐金融产品时，则会按照这种设定好的比例进行推荐，但是，由于不同产品的属性值有一定的差异，因而使得金融产品相对于用户而言存在优劣之分，这种将流量均分给多个金融产品的方式，会使得较优的金融产品不能被更多的用户所见，对于整体用户体验而言显然是不佳的。因此，如何更为有效的进行流量分配，使得推荐给用户的金融产品的精准度较高是目前亟待解决的技术问题。
鉴于上述的问题，在本申请实施例中发现正是因为目前的流量分配方式是直接均分的，所有的金融产品的用户推荐比例均相同，而并未考虑到各金融产品本身的特性，从而才使得一些较优的金融产品并不能分配得到较多的流量，因此，为了解决上述问题，则需要在确定各金融产品的用户推荐比例时，将各金融产品的自身特性作为考虑因素。
因此，本申请实施例提供一种金融产品的流量分配方法，在该方法中，理财平台基于各金融产品的设定参数的历史数据，构建产品推荐特征，从而获取所有金融产品的综合产品推荐特征，进而根据各金融产品的产品推荐特征相对于对应类别的综合产品推荐特征的偏离度，来确定该金融产品的用户推荐比例，最终基于各金融产品的用户推荐比例为用户推荐金融产品，这样，设定参数是各金融产品自身的参数，因而一定程度上能够反映出金融产品的特性，从而基于设定参数构造的产品推荐特征与所有产品的综合产品推荐特征的偏离度，确定的用户推荐比例是直接与金融产品的参数相关的，从而各金融产品的用户推荐比例是由各产品自身的特性决定的，例如可以基于产品的优劣确定相对应的用户推荐比例，那么则可以为较优的金融产品分配更高的用户推荐比例，使得较优的金融产品能够被更多的 用户所见，从而提高推荐金融产品的精准度，进而提高整体用户使用体验。并根据各金融产品自身的参数，即可确定出与金融产品的自身参数相关的用户推荐比例，则金融产品分流不会受到非自身参数的影响，从而提高理财平台的稳定性，进而提高用户数据的安全性。
在介绍完本申请实施例的设计思想之后，下面对本申请实施例的技术方案能够适用的应用场景做一些介绍，需要说明的是，以下介绍的应用场景仅用于说明本申请实施例而非限定。可以根据需要灵活地应用本申请实施例提供的技术方案。
请参见图1A所示，为发明实施例能够适用的一种场景示意图，该场景中可以包括服务器101、多个终端102(示例性示出终端102～1至终端102～L)以及多个金融机构103(示例性示出金融机构103～1至金融机构103～P)，其中，L、P均为正整数，L、P的值分别代表用户和金融机构的总数量，本申请实施例并不进行限制。
金融机构103可以表示各金融机构的设备，各金融机构可以提供一个或者多个金融产品，金融机构103可以计算得到各金融产品的收益数据，并进行存储。其中，金融机构103可以包括一个或多个处理器1031、存储器1032以及与服务器交互的I/O接口1033等，处理器1031可以计算得到各金融产品的收益数据，并存储于存储器1032中，还可通过与服务器交互的I/O接口1033将各金融产品的收益数据发送给服务器101。
服务器101(理财平台的后台服务器)可以是独立的物理服务器，也可以是多个物理服务器构成的服务器集群或者分布式系统，还可以是提供云计算服务的云服务器。以云端的服务器(封装有金融产品推荐的程序)为例，用户通过终端调用云服务中的金融产品推荐服务，以使部署在云端的服务器调用金融机构103的收益数据，服务器调用封装的金融产品推荐的程序，确定出各金融产品所分得的流量所占的比例，并根据比例确定为用户提供的金融产品，并将该金融产品推送给终端，以在终端的显示界面上 显示推荐的金融产品。
服务器101可以包括一个或多个处理器1011、存储器1012、与终端交互的I/O接口1013以及与金融机构交互的I/O接口1014等。此外，服务器101还可以配置数据库1015，数据库1015可以用于存储各用户的用户信息、历史操作信息等与用户相关的信息，以及还可以存储金融机构提供的金融产品的信息，例如收益数据、金融机构相关信息等。其中，服务器101的存储器1012中可以存储本申请实施例提供的金融产品的流量分配方法的程序指令，这些程序指令被处理器1011执行时能够用以实现本申请实施例提供的金融产品的流量分配方法的步骤，即根据各金融产品的收益数据，确定为各金融机构提供的金融产品分配的流量，例如最终可以确定出各金融产品所分得的流量所占的比例，那么当有新用户加入金融平台时，则可以基于确定的比例确定需要为新用户呈现的金融产品，从而控制各金融产品的流量比例维持在上述确定出的比例。
终端102可以为手机、个人电脑(personal computer，PC)或者平板电脑等终端设备，终端102可以呈现理财平台的显示页面，例如终端102可以安装理财平台提供的应用程序(application，APP)，从而在理财平台提供的APP中打开理财平台的显示页面；或者，通过终端102上的浏览器呈现理财平台的显示页面；或者，还可以在其他应用中打开理财平台的显示页面，其他应用是指非理财平台自身提供的APP，例如理财平台可以在APP中以轻应用的形式存在或者理财平台可以作为APP的一种功能提供给用户，例如微信中的小程序、公众号或者插件等形式。
终端102可以包括一个或多个处理器1021、存储器1022、与服务器101交互的I/O接口1023、显示面板1024等。其中，终端102的存储器1022中可以存储实现理财平台功能的程序指令，这些程序指令被处理器1021执行时能够用以实现理财平台的功能，以及在显示面板1024显示理财平台的相应显示页面。
示例性的，当有新用户注册理财平台的账号，并进入理财平台的页面时，服务器101都会基于预先确定的各金融产品的流量分配情况，来确定为该新用户提供的金融产品，并将该金融产品推送给用户，这样，用户通过理财平台的显示界面即可查看到该金融产品。如图2所示，为理财平台的一种显示页面的示意图，其中，在理财平台的显示页面上，则可以查看到为该用户分配的金融产品的名称201，如图2所示的“理财产品A”，以及还可以显示该金融产品的收益数据202，用户可以根据自身情况选择是否申购该金融产品，若是选择申购，则可通过操作按钮203中的“转入”按钮进行金额的转入，以申购金融产品；若是选择不申购，则可通过页面跳转按钮204退出该显示界面。在用户新加入到理财平台时，由于该用户未曾申购任何金融产品，因此初次进入理财平台的显示页面时，所显示账户余额为零，而当用户申购了金融产品之后，则会图2中右图所示账户余额不为零，且随着时间的增长，收益逐渐增加，账户余额以及累计收益金额也就会随之增加。在申购金融产品之后，若是用户需要兑现可流通货币，那么可通过操作按钮中的“转出”按钮进行金额的转出，以实现金融产品到货币的转换。
服务器101与终端102之间，以及服务器101与金融机构103之间可以通过一个或者多个网络104进行通信连接。该网络104可以是有线网络，也可以是无线网络，例如无线网络可以是移动蜂窝网络，或者可以是无线保真(WIreless-Fidelity，WIFI)网络，当然还可以是其他可能的网络，本申请实施例对此不做限制。
请参见图1B所示，为发明实施例能够适用的一种场景示意图，图1B示出图1A中的网络104为区块链网络(示例性示出了共识节点1041-1至共识节点1041-5)。区块链网络的类型是灵活多样的，例如可以为公有链、私有链或联盟链中的任意一种。以公有链为例，任何业务主体的电子设备例如服务器101(理财平台)、终端102和金融机构103，都可以在不需要授权的情况下接入区块链网络，以作为区块链的共识节点，例如服务器10 1映射为区块链网络中的共识节点1041～1，金融机构103～1映射为区块链网络中的共识节点1041～2，终端102～1映射为区块链网络中的共识节点1041～3；以联盟链为例，业务主体在获得授权后其下辖的电子设备可以接入区块链网络，例如服务器101、终端102和金融机构103。
例如，用户在终端102(包含理财客户端)上浏览理财信息时，终端102向服务器101(理财平台)发起推荐金融产品请求，服务器101映射为区块链网络中的共识节点1041～1。其中，终端102根据金融产品请求生成对应更新操作的交易，在交易中指定了实现更新操作需要调用的智能合约、以及向智能合约传递的参数，交易还携带了终端102的数字证书、签署的数字签名(例如，使用终端102的数字证书中的私钥，对交易的摘要进行加密得到)，并将交易广播到区块链网络中的共识节点。
例如，服务器101(共识节点1041～1)接收到交易时，对交易携带的数字证书和数字签名进行验证，验证成功后，根据交易中携带的身份，确认终端102是否是具有交易权限，数字签名和权限验证中的任何一个验证判断都将导致交易失败。验证成功后签署节点自己的数字签名(例如，共识节点1041～1的私钥对交易的摘要进行加密得到)，并调用集成有金融产品推荐的智能合约，获取每一个金融产品的设定参数，并根据每一个金融产品的设定参数的历史数据，确定出每一个金融产品的用户推荐比例，根据各金融产品的用户推荐比例确定为用户推荐的金融产品，并根据每一个金融产品的用户推荐比例以及为用户推荐的金融产品生成交易，将该交易广播到区块链网络中的共识节点中。
例如，金融机构103～1(共识节点1041～2)接收到交易时，对交易携带的数字证书和数字签名进行验证，验证成功后，签署节点自己的数字签名(例如，共识节点1041～2的私钥对交易的摘要进行加密得到)，将签署标签后的交易广播到区块链网络中的共识节点中，继续进行共识。
当服务器101(理财平台)再次接收到交易时，继续对交易携带的数字 证书和数字签名进行验证，验证成功后，签署节点自己的数字签名，并返回客户端，客户端接收到交易时，对交易携带的数字证书和数字签名进行验证，当验证通过且确定交易被共识成功的次数超过共识阈值时，则可确认交易结果的可靠性，即保证了每一个金融产品的用户推荐比例以及为用户推荐的金融产品的安全性。
因此，基于区块链网络去中心化、分布式存储和不可篡改的特性，通过区块链网络确定各金融产品的用户推荐比例以及为用户推荐的金融产品，可以保证计算的公平性和透明性，保障了理财平台的安全性，用户可以根据推荐的金融产品进行安全购物。
当然，本申请实施例提供的方法并不限用于图1A-1B所示的应用场景中，还可以用于其他可能的应用场景，本申请实施例并不进行限制。对于图1A-1B所示的应用场景的各个设备所能实现的功能将在后续的方法实施例中一并进行描述，在此先不过多赘述。
请参见图3，为本申请实施例提供的金融产品推荐方法的流程示意图，该方法可以通过电子设备来执行，例如可以由图1A中的服务器来执行。
步骤301：接收客户端的推荐金融产品请求，分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征。
例如，用户在客户端上浏览理财的相关信息时，客户端自动生成推荐金融产品请求，并向服务器发送推荐金融产品请求，服务器接收到该推荐金融产品请求后，根据每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，以便根据产品推荐特征进行后续处理。
本申请实施例中，理财平台中可以存在多个金融产品，N个金融产品可以是理财产品中的全部金融产品，或者可以是全部金融产品中的部分。例如理财平台中包括5个金融产品，那么N个金融产品可以就是指这5个金融产品；或者，当5个金融产品中的其中一个金融产品A采用固定用户 推荐比例，例如其用户推荐比例为1/5，那么N个金融产品可以就是指除金融产品A之外的其余4个金融产品，且这4个金融产品还可分配的用户推荐比例总和为4/5。
例如，因为通常用户是可以同时申购不同类型的金融产品的，即不同类型的金融产品一般不存在用户之间的竞争问题，因此流量分配一般是针对同类型的金融产品而言的。
本申请实施例中，设定参数可以为金融产品的任意可能的参数，例如关注收益的金融产品，设定参数可以为收益率，例如关注风险的金融产品，设定参数可以为风险率等，其中，由于用户在申购金融产品时，一般较为关心金融产品的收益率，因此后续将以设定参数为收益率为例，对本申请实施例的金融产品推荐方法进行介绍。
例如，对于货币基金类的金融产品，收益率可以为万份收益、7日年化、30日年化或者年化收益率等指标，而对于非货币基金类的金融产品，收益率可以为最近1月收益或者最近3月收益等指标。
例如，各金融产品的收益率可以是理财平台根据各金融产品的收益数据计算得到的；或者，由于各金融机构对于自家的金融产品都会进行收益率等指标的统计，因此，为了防止理财平台的计算方式和金融机构的计算当时存在偏差，使得收益率与金融机构计算的收益率不同，理财平台还可以直接从金融机构获取收益率等数据，这样，可以避免由于用户人数多且理财产品的金额大，不恰当的分流计算导致计算资源的巨大消耗，且容易出错，还为理财平台节省了一定的计算量，减轻服务器的计算压力。由于收益率一般是周期性更新的，那么理财平台的服务器可以周期性的从金融机构获取收益率的数据，例如，若是收益率的数据每天更新一次，那么服务器可以每天定时从金融机构获取收益率的数据；或者，若是收益率的数据每月更新一次，那么服务器可以每月定时从金融机构获取收益率的数据。
例如，服务器除了可以采用主动向金融机构的电子设备申请获取收益 率的数据，以接收金融机构的电子设备返回的收益率的数据的方式之外，还可以采用与金融机构预先约定在金融机构的电子设备计算完成收益率的数据之后，金融机构的电子设备将收益率的数据提供给服务器的方式。当服务器获取收益率的数据之后，可以将收益率的数据统一进行存储，例如存储至数据库中，在需要使用收益率的数据时，直接从数据库读取即可。
本申请实施例中，服务器能够分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征。其中，N、M均为正整数。
例如，M类产品推荐特征包括如下特征的至少之一：
设定参数在第一设定时间段内的均值，即平均收益率；
设定参数在第二设定时间段内的波动率的均值，即平均收益波动率；
在第二设定时间段内的组合特征的均值，其中，组合特征与设定参数呈正相关，且与设定参数的波动率呈负相关。
在一些实施例中，M类产品推荐特征可以是为上述产品推荐特征中的任意一种，也可以是多类产品推荐特征的组合。但是，不管是多少类产品推荐特征，构建产品推荐特征的过程均是独立的。
例如，当产品推荐特征为设定参数在第一设定时间段内的均值时，其中，第一设定时间段为设定参数的统计时间周期T 1，T 1的长度可以根据情况进行设定，例如可以为最近一个月，或者最近两个月等，本申请实施例对此不做限制。针对每一个金融产品而言，基于其设定参数的历史数据，构建该金融产品的产品推荐特征，可以是根据该金融产品的设定参数在第一设定时间段内的每一个子时间段的数据值，以及每一个子时间段对应的权重值，获取每一个金融产品的设定参数在第一设定时间段内的均值。其中，针对每一个金融产品，都可以通过上述方式获取设定参数在第一设定时间段内的均值。
子时间段对应的权重值可以用于区分更关注长期或者短期的数据，例 如，若是更为关注长期的数据，则可以将距离当前时间更远的子时间段的权重值设置的更高，相反的，若是更为关注短期的数据，则可以将距离当前时间更近的子时间段的权重值设置的更高。
例如，当产品推荐特征为设定参数在第二设定时间段内的波动率的均值时，其中，第二设定时间段为设定参数的统计时间周期T 2，T 2的长度可以与T 1相同，也可以与T 1不相同。由于短时间内的波动率可能并不会很大，因此T 2的长度可以设置为较长的时间段，例如可以设置为最近一月、最近半年或者最近一年等。
获取设定参数在第二设定时间段内的波动率的均值，那么必然需要获取各金融产品的波动率。针对每一个金融产品，可以根据该金融产品的设定参数在每一个子时间段的数据值，获取该金融产品的设定参数在每一个子时间段的波动率。
例如，每个金融产品的波动率表征了该个金融产品的收益率的变化程度。波动率可以通过如下过程获取：
首先，获取该金融产品的设定参数在每一个子时间段的数据值，相较于该子时间段的上一个子时间段的数据值的变化率。例如，设定参数在子时间段t 1的数据值为A，设定参数在子时间段t 1的上一个子时间段t 2的数据值为B，那么变化率则可以为ln(A/B)。
其次，获取该金融产品的每一个子时间段对应的变化率相较于第二设定时间段内的平均变化率的偏离程度。其中，平均变化率为第二设定时间段内的变化率的均值，偏离程度可以通过方差或者标准差进行表示。
最后，根据该金融产品的每一个子时间段对应的偏离程度，获取该金融产品的设定参数在每一个子时间段的波动率。例如，偏离程度通过方差表示，那么波动率则可以表示为方差与T 2的平方根的比值。
本申请实施例中，在获取各金融产品的波动率之后，则可以根据该金融产品的设定参数在每一个子时间段的波动率，以及每一个子时间段对应 的权重值，则可以获取每一个金融产品的设定参数在第二设定时间段内的波动率的均值。其中，针对每一个金融产品，都可以通过上述方式获取设定参数在第二设定时间段内的波动率的均值。
例如，当产品推荐特征为在第二设定时间段内的组合特征的均值时，组合特征可以是波动率和设定参数的组合。举例来讲，当设定参数为收益率时，在收益率持续较低时收益率的波动率也可以较低，但是收益率较低的金融产品显然不会是较优的金融产品，因而在确定金融产品的用户推荐比例时，除考虑收益率的波动率时，同时还需要考虑收益率，即可以基于波动率与收益率构建组合特征。其中，该组合特征的值可以与收益率呈正相关，且与波动率呈负相关，即表示收益率越高，且波动率越小的金融产品为更优的产品。
例如，根据各子时间段的设定参数和波动率获取各子时间段的组合特征值之后，则可以获取各金融产品的组合特征在第二设定时间段内的均值。当然，在计算均值，也可以为每一个子时间段赋予一定的权重值，对于赋予权重值的方式可以参见上述计算设定参数在第一设定时间段内的均值部分的描述。
步骤302：分别针对M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征。
本申请实施例中，产品推荐特征用于表征N个金融产品中的一个金融产品的特征，而综合产品推荐特征用于表征N个金融产品的整体特征。
例如，综合产品推荐特征可以通过产品推荐特征的均值和方差进行表示。那么在通过步骤301的过程获取各金融产品的产品推荐特征之后，则可以通过计算各金融产品的产品推荐特征的均值和方差的方式获取得到N个金融产品的综合产品推荐特征。
步骤303：根据每一个金融产品的各类产品推荐特征相对于与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例。
本申请实施例中，用户推荐比例为将每一个金融产品所对应的被推荐用户占所有用户的比例。
例如，当M类产品推荐特征仅包括上述产品推荐特征中的其中一个时，则可以根据该产品推荐特征相对于确定的综合产品推荐特征的偏离度，确定各个金融产品的用户推荐比例。
其中，偏离度可以是指绝对偏离度，即一个金融产品的产品推荐特征与N个金融产品的产品推荐特征的均值的差值；或者，偏离度也可以是指相对偏离度，即绝对偏离度的值与方差的比值。
在获取各金融产品对应的偏离度之后，则可以基于偏离度获取各金融产品的用户推荐比例，其中，各金融产品的用户推荐比例与偏离度成正相关。
例如，当M类产品推荐特征仅包括上述产品推荐特征中的多个时，则可以根据各金融产品的各类产品推荐特征对应的偏离度，分别获取各类产品推荐特征对应的用户推荐子比例，再根据各类产品推荐特征的用户推荐权重，计算得到最终的用户推荐比例。其中，分别获取各类产品推荐特征对应的用户推荐子比例的过程，与上述当M类产品推荐特征仅包括上述产品推荐特征中的其中一个时的计算过程相同，因此可以参见上述的描述，在此不再进行赘述。
例如，各类产品推荐特征的用户推荐权重的总和为100％，因此可以通过最优求解过程获取得到各类产品推荐特征的用户推荐权重。当然，在一些实施例中，也可以为各类产品推荐特征设置固定的用户推荐权重，本申请实施例对此不做限制。
示例性的，若最终的用户推荐比例计算公式如公式(1)所示：
Figure PCTCN2020093503-appb-000001
其中，f i为第i个金融产品的用户推荐比例，ω j为第j类产品推荐特征 的用户推荐权重，
Figure PCTCN2020093503-appb-000002
为第i个金融产品的第j类产品推荐特征对应的用户推荐子比例。
例如，在计算用户推荐权重时，则可以以上述计算公式作为目标函数，以及各类产品推荐特征的用户推荐权重的总和为100％作为约束条件来计算最优的用户推荐权重。当然，约束条件还可以增加其他条件，例如所有金融产品的用户推荐比例为一固定值等。
例如，由于设定参数可能会随着时间的推移而会发生一定的变化，因此本申请实施例的步骤301～303可以是多次重复进行的，例如可以是周期性重复进行的，或者可以在设定参数的变化值大于或者等于一定阈值之后，再次进行用户推荐比例的确定。例如设定参数为金融产品的收益率时，收益率一般是周期性更新的，例如每天更新一次，或者每月更新一次，因此对应的，用户推荐比例的确定则可以是每天进行一次，或者可以是每月进行一次。
步骤304：根据各金融产品的用户推荐比例为用户推荐金融产品，以响应推荐金融产品请求。
本申请实施例中，在各金融产品的用户推荐比例之后，则可以基于各金融产品的用户推荐比例为用户推荐金融产品。
由于新用户加入理财平台的时间并不是固定的，且在新用户加入理财平台之后，一般就需要将为其推荐的金融产品显示在理财平台的页面上，因此理财平台无法在已有用户的基础上统一进行金融产品的流量分配，而是在新用户已进入理财平台时，就需要为其推荐金融产品。
例如，在为用户推荐金融产品时，是基于确定的各金融产品的用户推荐比例进行推荐，以使得各金融产品所对应的被推荐用户的数量占所有用户的比例与各金融产品的用户推荐比例接近或者相同。其中，所利用的用户推荐比例一般是最近一次获取到的用户推荐比例，
在确定为用户推荐的金融产品之后，服务器则可以将为用户推荐的金 融产品的状态数据发送给用户，这样，通过用户设备登录用户对应的账号后，能够在显示页面上显示为用户推荐的金融产品的状态数据，例如如图2所示的显示界面。其中，状态数据可以包括金融产品的名称、收益率、用户申购情况以及用户收益情况等数据。
下面将示出几个获取用户推荐比例的例子，其中，设定参数以收益率为例。
如图4所示，为以产品推荐特征为收益率在第一设定时间段内的均值为例对用户推荐比例的确定过程进行介绍。
步骤401：获取单个金融产品的产品推荐特征。
本申请实施例中，产品推荐特征为收益率在第一设定时间段内的均值，其中，第一设定时间段为设定参数的统计时间周期T 1，T 1的长度可以根据情况进行设定，例如可以为最近一个月，或者最近两个月等，本申请实施例对此不做限制。
例如，由于金融产品的收益率的更新周期通常为一天，因此一个子时间段可以设置为一天，那么设定参数在第一设定时间段内的均值的计算方式可以如公式(2)所示：
Figure PCTCN2020093503-appb-000003
其中，
Figure PCTCN2020093503-appb-000004
为第i个金融产品的收益率在第一设定时间段内的均值，i＝1,2,3…N；
Figure PCTCN2020093503-appb-000005
为第i个金融产品在第t个子时间段的收益率，t＝1,2,3…T 1；
Figure PCTCN2020093503-appb-000006
为第t个子时间段对应的权重值，用于区分更关注长期或者短期的数据，例如，若是更为关注长期的数据，则可以将距离当前时间更远的子时间段的权重值设置的更高，相反的，若是更为关注长期的数据，则可以将距离当前时间更近的子时间段的权重值设置的更高。
例如，当
Figure PCTCN2020093503-appb-000007
时，
Figure PCTCN2020093503-appb-000008
为几何均值，即每个子时间段的权重均相等；当
Figure PCTCN2020093503-appb-000009
t＝1,2,3…T 1时，
Figure PCTCN2020093503-appb-000010
为线性权重，则表示距离当前时间越近， 权重值越大。当然，
Figure PCTCN2020093503-appb-000011
还可以是其它可能的权重函数，例如指数函数或者对数函数等，本申请实施例对此不做限制。
通过上述的产品推荐特征的获取过程，可以获取到所有金融产品的产品推荐特征。
步骤402：获取N个金融产品的综合产品推荐特征。
本申请实施例中，这里以综合产品推荐特征为产品推荐特征的均值和方差为例，那么综合产品推荐特征的计算方式可以如公式(3)、(4)所示：
Figure PCTCN2020093503-appb-000012
Figure PCTCN2020093503-appb-000013
其中，
Figure PCTCN2020093503-appb-000014
为N个金融产品的产品推荐特征的均值，δ a为N个金融产品的产品推荐特征的方差。
当然，除了将均值和方差作为综合产品推荐特征之外，还可以将均值和标准差作为综合产品推荐特征，当然，还可以将其他可能的采纳数作为综合产品推荐特征，本申请实施例对此不做限制。
步骤403：获取各金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度。
本申请实施例中，这里的偏离度以相对偏离度为例。各金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度的计算方式可以如公式(5)所示：
Figure PCTCN2020093503-appb-000015
其中，k ai为第i个金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度，这里的下标a表示对应的产品推荐特征为收益率在T 1的均值。
步骤404：基于各金融产品对应的相对偏离度确定用户推荐比例。
本申请实施例中，很容易理解到，金融产品的收益率越高时，该金融 产品对应的相对偏离度的值应是越大的，且金融产品的收益率越高时，应该为该金融产品分配更多的流量，即用户推荐比例应更高，因此，金融产品对应的相对偏离度的值越大，该金融产品的用户推荐比例应更高。这样，能够见到该金融产品的用户的数量才更多，才能整体上提高用户的使用体验，提高用户对于金融平台的黏性。因此，用户推荐比例的计算方式可以如公式(6)所示：
Figure PCTCN2020093503-appb-000016
其中，
Figure PCTCN2020093503-appb-000017
为第i个金融产品的用户推荐比例；α是分配系数，α用于表征可分配的总流量比例，α可以设定为固定值，亦可设定为变化的值。
例如，各金融产品的收益率有高有低，因此有可能出现有金融产品的相对偏离度为负值的情况，因此，为了保证相对偏离度最小，即负向偏离最远的金融产品能够分配到流量，且为了避免流量过于集中，可将α的值设置为满足以下条件的值，如公式(7)所示：
Figure PCTCN2020093503-appb-000018
在基于上述计算过程获取各金融产品的用户推荐比例，则可以基于各金融产品的用户推荐比例为用户推荐金融产品了。
对于产品推荐特征为设定参数在第二设定时间段内的波动率的均值时，计算用户推荐比例的过程与上述过程类似，即将产品推荐特征更换为设定参数在第二设定时间段内的波动率的均值即可，因此对于产品推荐特征为设定参数在第二设定时间段内的波动率的均值时，计算用户推荐比例的过程可以参见上述的描述，本申请实施例对此不再进行赘述。
如图5所示，为以产品推荐特征为在第二设定时间段内的组合特征的均值为例对用户推荐比例的确定过程进行介绍。
步骤501：获取单个金融产品的收益率的波动率。
本申请实施例中，产品推荐特征为在第二设定时间段内的组合特征的 均值，其中，第二设定时间段为设定参数的统计时间周期T 2，T 2的长度可以根据情况进行设定，例如可以为最近一个月、最近半年或者最近一年等，本申请实施例对此不做限制。
例如，组合特征可以为收益率与收益率的波动率构成的组合特征，因此在获取组合特征的均值之前，需要首先获取各金融产品的收益率的波动率。
例如，针对一个金融产品而言，在计算收益率的波动率时，可以基于该金融产品在第二设定时间段内的收益率构造该金融的相对变化特征，相对变化特征的计算方式可以如公式(8)所示：
Figure PCTCN2020093503-appb-000019
其中，
Figure PCTCN2020093503-appb-000020
为第i个金融产品在第t个子时间段的数据值，相较于第t-1个子时间段的数据值的变化率，t＝1,2,3…T 2。
在第二设定时间段内的收益率的波动率可以理解为在第二设定时间段内的变化率的离散程度，因此，可以通过计算如下方式计算
Figure PCTCN2020093503-appb-000021
的均值和方差，如公式(9)、(10)所示：
Figure PCTCN2020093503-appb-000022
Figure PCTCN2020093503-appb-000023
其中，
Figure PCTCN2020093503-appb-000024
为
Figure PCTCN2020093503-appb-000025
在第二设定时间段内的均值，δ c为
Figure PCTCN2020093503-appb-000026
在第二设定时间段内的方差。
那么，一个金融产品的收益率的波动率则可以通过公式(11)进行计算：
Figure PCTCN2020093503-appb-000027
其中，
Figure PCTCN2020093503-appb-000028
为第i个金融产品在第t个子时间段的收益率的波动率。对于第t个子时间段而言，第t个子时间段的收益率的波动率是基于第t个子时 间段至第t个子时间段之前的T 2个子时间段内的数据为基础计算得到的。举例来讲，若统计时间周期为半年，那么当日的波动率是基于当日以及当日之前的半年内的数据为基础计算得到的，而昨日的波动率是基于昨日以及昨日之前的半年内的数据为基础计算得到的。
步骤502：基于收益率的波动率构建各金融产品的组合特征。
本申请实施例中，当收益率持续较低时收益率的波动率也可以较低，但是收益率较低的金融产品显然不会是较优的金融产品，因而在确定金融产品的用户推荐比例时，除考虑收益率的波动率时，同时还需要考虑收益率，即可以基于波动率与收益率构建组合特征。其中，该组合特征的值可以与收益率呈正相关，且与波动率呈负相关，即表示收益率越高，且波动率越小的金融产品为更优的产品，因此，组合特征可以通过如公式(12)进行表示：
Figure PCTCN2020093503-appb-000029
其中，
Figure PCTCN2020093503-appb-000030
为第i个金融产品的在第t个子时间段的组合特征。当然，上述方式只是组合特征的一种表达方式，还可以采用其他可能的且满足上述组合特征的规律的其他方式，本申请实施例对此不做限制。
步骤503：获取单个金融产品的产品推荐特征。
本申请实施例中，产品推荐特征为第二设定时间段内的组合特征的均值，其中，第二设定时间段内的组合特征的均值的计算方式可以如公式(13)所示：
Figure PCTCN2020093503-appb-000031
其中，
Figure PCTCN2020093503-appb-000032
为第i个金融产品在第二设定时间段内的组合特征的均值，i＝1,2,3…N。
例如，由于金融产品的收益率的更新周期通常为一天，因此一个子时间段既可以设置为一天。
Figure PCTCN2020093503-appb-000033
为第t个子时间段对应的权重值，用于区分更关注长期或者短期的数据，例如，若是更为关注长期的数据，则可以将距离当前时间更远的子时间段的权重值设置的更高，相反的，若是更为关注长期的数据，则可以将距离当前时间更近的子时间段的权重值设置的更高。
例如，当
Figure PCTCN2020093503-appb-000034
时，
Figure PCTCN2020093503-appb-000035
为几何均值，即每个子时间段的权重均相等；当
Figure PCTCN2020093503-appb-000036
t＝1,2,3…T 2时，
Figure PCTCN2020093503-appb-000037
为线性权重，则表示距离当前时间越近，权重值越大。当然，
Figure PCTCN2020093503-appb-000038
还可以是其它可能的权重函数，例如指数函数或者对数函数等，本申请实施例对此不做限制。
通过上述的产品推荐特征的获取过程，可以获取到所有金融产品的产品推荐特征。
步骤504：获取N个金融产品的综合产品推荐特征。
本申请实施例中，这里以综合产品推荐特征为产品推荐特征的均值和方差为例，那么综合产品推荐特征的计算方式可以如公式(14)、(15)所示：
Figure PCTCN2020093503-appb-000039
Figure PCTCN2020093503-appb-000040
其中，
Figure PCTCN2020093503-appb-000041
为N个金融产品的产品推荐特征的均值，δ b为N个金融产品的产品推荐特征的方差。
当然，除了将均值和方差作为综合产品推荐特征之外，还可以将均值和标准差作为综合产品推荐特征，当然，还可以将其他可能的采纳数作为综合产品推荐特征，本申请实施例对此不做限制。
步骤505：获取各金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度。
本申请实施例中，这里的偏离度以相对偏离度为例。各金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度的计算方式可以如公式 (16)所示：
Figure PCTCN2020093503-appb-000042
其中，k bi为第i个金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度，这里的下标b表示对应的产品推荐特征为T 2内的组合特征的均值。
步骤506：基于各金融产品对应的相对偏离度确定用户推荐比例。
本申请实施例中，很容易理解到，金融产品的收益率越高，且波动率越小时，组合特征的值越大，那么金融产品对应的相对偏离度的值应是越大的，且金融产品的收益率越高，且波动率越小时，应该为该金融产品分配更多的流量，即用户推荐比例应更高，因此，金融产品对应的相对偏离度的值越大，该金融产品的用户推荐比例应更高。这样，能够见到该金融产品的用户的数量才更多，才能整体上提高用户的使用体验，提高用户对于金融平台的黏性。因此，用户推荐比例的计算方式可以如公式(17)所示：
Figure PCTCN2020093503-appb-000043
其中，
Figure PCTCN2020093503-appb-000044
为第i个金融产品的用户推荐比例；α是分配系数，α用于表征可分配的总流量比例，α可以设定为固定值，亦可设定为变化的值。
例如，各金融产品的收益率有高有低，因此有可能出现有金融产品的相对偏离度为负值的情况，因此，为了保证相对偏离度最小，即负向偏离最远的金融产品能够分配到流量，且为了避免流量过于集中，可将α的值设置为满足以下条件的值，如公式(18)所示：
Figure PCTCN2020093503-appb-000045
在基于上述计算过程获取各金融产品的用户推荐比例，则可以基于各金融产品的用户推荐比例为用户推荐金融产品了。
如图6所示，为以产品推荐特征包括收益率在第一设定时间段内的均 值和在第二设定时间段内的组合特征的均值为例对用户推荐比例的确定过程进行介绍。其中，收益率在第一时间段内的均值为第一产品推荐特征，在第二设定时间段内的组合特征的均值为第二产品推荐特征。
步骤601：根据第一产品推荐特征确定第一产品推荐特征对应的用户推荐子比例。
该步骤的过程可以参见步骤401～404的介绍，在此不再过多赘述。
步骤602：根据第二产品推荐特征确定第二产品推荐特征对应的用户推荐子比例。
该步骤的过程可以参见步骤50～506的介绍，在此不再过多赘述。其中，需要声明的是，步骤601和步骤602并没有实质上的先后顺序关系，在一些实施例中，步骤601和步骤602可以同时执行，也可以先后顺序执行，例如先执行步骤601，再执行步骤602，图6以此为例，或者，先执行步骤602，再执行步骤601。
步骤603：基于各类产品推荐特征对应的用户推荐子比例以及各类产品推荐特征对应的用户推荐权重，获取金融产品的用户推荐比例。
本申请实施例中，各类产品推荐特征对应的用户推荐权重可以为固定的权重，也可以是通过最优解求解方法计算得到的。
例如，用户推荐比例的计算方式可以如公式(19)、(20)所示：
Figure PCTCN2020093503-appb-000046
ω a+ω b＝1     (20)
其中，f i为第i个金融产品的用户推荐比例，ω a为第一产品推荐特征对应的用户推荐权重，
Figure PCTCN2020093503-appb-000047
为第i个金融产品的第一产品推荐特征对应的用户推荐子比例，ω b为第二产品推荐特征对应的用户推荐权重，
Figure PCTCN2020093503-appb-000048
为第i个金融产品的第二产品推荐特征对应的用户推荐子比例。
本申请实施例中，考虑到在为用户推荐金融产品之后，各金融产品对于用户的吸引力可能不同，而这些吸引力不仅仅是收益率或者收益率的稳 定与否带来的，还可能与金融产品的其他因素相关，例如金融产品的品牌知名度、产品管理人知名度等，都会对用户是否申购金融产品产生影响，而金融产品的产品吸引力可以通过该金融产品的用户转化率进行衡量，因此为了综合考虑其他因素，还可以将金融产品对用户的转化率考虑进去，即可以将用户转化率与上述M类产品推荐特征中的任一特征进行组合构建新的组合产品推荐特征。如图7所示，下面以用户转化率与收益率在第一设定时间段内的均值进行组合为例，对用户推荐比例的确定过程进行介绍。
步骤701：获取各金融产品的用户转化率。
其中，用户转化率即金融产品对应的被推荐用户中实际使用该金融产品的用户数量所占的比例，那么用户转化率的计算方式可以如公式(21)所示：
Figure PCTCN2020093503-appb-000049
其中，π i为第i个金融产品的用户转化率，u i为实际使用第i个金融产品的用户数量占所有用户的比例。当然，除了上述使用第i个金融产品的用户数量占所有用户的比例、与用户推荐比例的比值作为用户转化率之外，还可以直接将实际使用第i个金融产品的用户数量与第i个金融产品对应的被推荐用户数量的比值作为用户转化率。
步骤702：基于用户转化率构建各产品推荐特征。
本申请实施例中，当用户转化率越高，以及收益率越高时，则表明该个金融产品为更优的金融产品，因此，组合特征可以通过如下公式(22)进行表示：
Figure PCTCN2020093503-appb-000050
其中，
Figure PCTCN2020093503-appb-000051
为第i个金融产品的基于用户转化率和平均收益率构建的组合产品推荐特征。当然，上述方式只是组合产品推荐特征的一种表达方式，还可以采用其他可能的且满足上述组合特征的规律的其他方式，本申请实 施例对此不做限制。
步骤703：获取N个金融产品的综合产品推荐特征。
步骤704：获取各金融产品的产品推荐特征与综合产品推荐特征之间的相对偏离度。
步骤705：基于各金融产品对应的相对偏离度确定用户推荐比例。
步骤703～705与步骤402～404，或者，步骤504～507的过程类似，因此对于步骤703～705部分可以参见步骤402～404或者步骤504～507部分的描述，在此不再过多赘述。
综上所述，本申请实施例中，基于各金融产品的设定参数的历史数据，构建产品推荐特征，从而获取所有金融产品的综合产品推荐特征，进而根据各金融产品的产品推荐特征相对于对应类别的综合产品推荐特征的偏离度，来确定该金融产品的用户推荐比例，最终基于各金融产品的用户推荐比例为用户推荐金融产品，这样，设定参数是各金融产品自身的参数，因而一定程度上能够反映出金融产品的特性，从而基于设定参数构造的产品推荐特征与所有产品的综合产品推荐特征的偏离度，确定的用户推荐比例是直接与金融产品的参数相关的，从而各金融产品的用户推荐比例是由各产品自身的特性决定的，例如，可以基于产品的优劣确定相对应的用户推荐比例，则可以为较优的金融产品分配更高的用户推荐比例，使得较优的金融产品能够被更多的用户所见，进而提高整体用户使用体验。
通过本申请实施例的金融产品推荐方法，不仅可以满足对同一产品限量以保证潜在的金融风险，还可以实现尽可能让更优质的金融产品分配到更多的流量，提高推荐的精准度，提升用户的体验，同时还可以避免金融产品提供方通过做高短时间的收益洞穿流量分配策略，提升平台的稳定性和引导金融资产公司为用户提供更优质的资产。同时，结合用户转化率，还可以提升平台流量的使用效率。
请参见图8，基于同一发明构思，本申请实施例还提供了一种金融产品 推荐装置80，该装置例如可以为图1A所示的服务器，该装置包括：
特征构建单元801，配置为接收客户端的推荐金融产品请求；分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数；
特征综合单元802，配置为分别针对所述M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征；
推荐比例确定单元803，配置为根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，所述用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；
产品推荐单元804，配置为根据各金融产品的用户推荐比例确定推荐金融产品，以响应所述推荐金融产品请求。
例如，M类产品推荐特征包括如下特征的至少之一：
设定参数在第一设定时间段内的均值；
设定参数在第二设定时间段内的波动率的均值；
在第二设定时间段内的组合特征的均值，其中，组合特征与设定参数呈正相关，且与设定参数的波动率呈负相关。
例如，特征构建单元801，配置为：分别根据每一个金融产品的设定参数在所述第一设定时间段内的每一个子时间段的数据值，以及所述每一个子时间段对应的权重值，获取所述每一个金融产品的设定参数在所述第一设定时间段内的均值。
例如，特征构建单元801，配置为：分别根据每一个金融产品的设定参数在所述第二设定时间段内的每一个子时间段的数据值，获取所述每一个金融产品的设定参数在所述每一个子时间段的波动率；分别根据所述每一个金融产品的设定参数在所述每一个子时间段的波动率，以及所述每一个子时间段对应的权重值，获取所述每一个金融产品的设定参数在所述第二 设定时间段内的波动率的均值。
例如，特征构建单元801，配置为：分别根据每一个金融产品的设定参数在所述第二设定时间段内的每一个子时间段的数据值，获取所述每一个金融产品的设定参数在所述每一个子时间段的波动率；分别根据所述每一个金融产品的设定参数，以及所述每一个金融产品的设定参数在所述每一个子时间段的波动率，构建所述组合特征，并分别获取所述每一个金融产品的组合特征在所述第二设定时间段内的均值。
例如，特征构建单元801，配置为：获取所述每一个金融产品的设定参数在所述每一个子时间段的数据值，相较于所述子时间段的上一个子时间段的数据值的变化率；获取所述每一个金融产品的每一个子时间段对应的变化率相较于所述第二设定时间段内的平均变化率的偏离程度；根据所述每一个金融产品的每一个子时间段对应的偏离程度，获取所述每一个金融产品的设定参数在所述每一个子时间段的波动率。
例如，推荐比例确定单元803，配置为：获取所述每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度；根据所述每一个金融产品各类产品推荐特征对应的偏离度，确定所述每一个金融产品的用户推荐比例；其中，每一个金融产品的用户推荐比例与所述偏离度呈正相关。
例如，推荐比例确定单元803配置为：根据所述每一个金融产品的各类产品推荐特征对应的偏离值，分别获取各类产品推荐特征对应的用户推荐子比例；获取所述每一个金融产品的各类产品推荐特征对应的用户推荐权重；其中，各类产品推荐特征对应的用户推荐权重总和为100％；根据所述各类产品推荐特征对应的用户推荐子比例以及所述各类产品推荐特征对应的用户推荐权重，获取所述每一个金融产品的用户推荐比例。
例如，装置还包括转化率获取单元805，配置为获取所述每一个金融产品的用户转化率，所述用户转化率为金融产品对应的被推荐用户中实际使 用所述金融产品的用户的数量所占总被推荐用户的数量的比例；
特征构建单元801，还配置为根据所述每一个金融产品的设定参数的历史数据，获取所述每一个金融产品的设定参数在所述第一设定时间段内的均值，并根据每一个金融产品的设定参数在所述第一设定时间段内的均值以及所述用户转化率，构建所述每一个金融产品的产品推荐特征。
例如，装置还包括数据发送单元806，配置为将为用户推荐的金融产品的状态数据发送给所述用户，以使得通过用户设备登录所述用户对应的账号后，在所述用户设备的显示页面上显示为所述用户推荐的金融产品的状态数据，所述状态数据包括所述金融产品的名称、收益率。
该装置可以用于执行图3～图7所示的实施例中所示的方法，因此，对于该装置的各功能模块所能够实现的功能等可参考图3～图7所示的实施例的描述，不多赘述。其中，转化率获取单元805和数据发送单元806并不是必选的功能单元，因此在图8中以虚线示出。
请参见图9，基于同一技术构思，本申请实施例还提供了一种电子设备90，可以包括存储器901和处理器902。
所述存储器901，用于存储处理器902执行的计算机程序。存储器901可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序等；存储数据区可存储根据电子设备的使用所创建的数据等。处理器902，可以是一个中央处理单元(Central Processing Unit，CPU)，或者为数字处理单元等等。本申请实施例中不限定上述存储器901和处理器902之间的连接介质。本申请实施例在图9中以存储器901和处理器902之间通过总线903连接，总线903在图9中以粗线表示，其它部件之间的连接方式，仅是进行示意性说明，并不引以为限。所述总线903可以分为地址总线、数据总线、控制总线等。为便于表示，图9中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
存储器901可以是易失性存储器(Volatile Memory)，例如随机存取存 储器(Random Access memory，RAM)；存储器901也可以是非易失性存储器(Non-volatile Memory)，例如个读存储器，快闪存储器(flash memory)，硬盘(Hard Disk Drive，HDD)或固态硬盘(Solid State Drive，SSD)、或者存储器901是能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器901可以是上述存储器的组合。
处理器902，用于调用所述存储器901中存储的计算机程序时执行如图3～图7中所示的实施例中设备所执行的方法。
在一些可能的实施方式中，本申请提供的方法的各个方面还可以实现为一种程序产品的形式，其包括程序代码，当所述程序产品在电子设备上运行时，所述程序代码用于使所述电子设备执行本说明书上述描述的根据本申请各种示例性实施方式的方法中的步骤，例如，所述电子设备可以执行如图3～图7中所示的实施例中设备所执行的方法。
所述程序产品可以采用一个或多个计算机可读存储介质的任意组合。计算机可读存储介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以是、但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。例如，可读存储介质的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、个读存储器(ROM)、可擦式可编程可读存储器(EPROM或闪存)、光纤、便携式紧凑盘个读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
尽管已描述了本申请的优选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例做出另外的变更和修改。所以，所附权利要求意欲解释为包括优选实施例以及落入本申请范围的所有变更和修改。
显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离 本申请的精神和范围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。
工业实用性
本申请实施例中通过服务器根据各金融产品的设定参数的历史数据，构建产品推荐特征，并获取所有金融产品的综合产品推荐特征，根据各金融产品的产品推荐特征相对于对应类别的综合产品推荐特征的偏离度，确定出金融产品的用户推荐比例，并根据各金融产品的用户推荐比例为用户推荐金融产品。如此，根据各金融产品自身的参数，即可确定出与金融产品的自身参数相关的用户推荐比例，提高了金融产品分流的精准度以及用户数据的安全性。

Claims (15)

  1. 一种金融产品推荐方法，所述方法由服务器执行，所述方法包括：
    接收客户端的推荐金融产品请求；
    分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数；
    分别针对所述M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征；
    根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，所述用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；
    根据各金融产品的用户推荐比例确定推荐金融产品，以响应所述推荐金融产品请求。
  2. 如权利要求1所述的方法，其中，所述M类产品推荐特征包括如下特征的至少之一：
    所述设定参数在第一设定时间段内的均值；
    所述设定参数在第二设定时间段内的波动率的均值；
    在第二设定时间段内的组合特征的均值，其中，所述组合特征与所述设定参数呈正相关，且与所述设定参数的波动率呈负相关。
  3. 如权利要求2所述的方法，其中，所述分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，包括：
    分别根据每一个金融产品的设定参数在所述第一设定时间段内的每一个子时间段的数据值，以及所述每一个子时间段对应的权重值，获取所述每一个金融产品的设定参数在所述第一设定时间段内的均值。
  4. 如权利要求2所述的方法，其中，所述分别根据N个金融产品中每 一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，包括：
    分别根据每一个金融产品的设定参数在所述第二设定时间段内的每一个子时间段的数据值，获取所述每一个金融产品的设定参数在所述每一个子时间段的波动率；
    分别根据所述每一个金融产品的设定参数在所述每一个子时间段的波动率，以及所述每一个子时间段对应的权重值，获取所述每一个金融产品的设定参数在所述第二设定时间段内的波动率的均值。
  5. 如权利要求2所述的方法，其中，所述分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，包括：
    分别根据每一个金融产品的设定参数在所述第二设定时间段内的每一个子时间段的数据值，获取所述每一个金融产品的设定参数在所述每一个子时间段的波动率；
    分别根据所述每一个金融产品的设定参数，以及所述每一个金融产品的设定参数在所述每一个子时间段的波动率，构建所述组合特征，并
    分别获取所述每一个金融产品的组合特征在所述第二设定时间段内的均值。
  6. 如权利要求4或5所述的方法，其中，分别根据每一个金融产品的设定参数在所述第二设定时间段内的每一个子时间段的数据值，获取所述每一个金融产品的设定参数在所述每一个子时间段的波动率，包括：
    获取所述每一个金融产品的设定参数在所述每一个子时间段的数据值，相较于所述子时间段的上一个子时间段的数据值的变化率；
    获取所述每一个金融产品的每一个子时间段对应的变化率相较于所述第二设定时间段内的平均变化率的偏离程度；
    根据所述每一个金融产品的每一个子时间段对应的偏离程度，获取所 述每一个金融产品的设定参数在所述每一个子时间段的波动率。
  7. 如权利要求1或2所述的方法，其中，所述根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，包括：
    获取所述每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度；
    根据所述每一个金融产品各类产品推荐特征对应的偏离度，确定所述每一个金融产品的用户推荐比例；其中，每一个金融产品的用户推荐比例与所述偏离度呈正相关。
  8. 如权利要求7所述的方法，其中，根据所述每一个金融产品各类产品推荐特征对应的偏离度，确定所述每一个金融产品的用户推荐比例，包括：
    根据所述每一个金融产品的各类产品推荐特征对应的偏离度，分别获取各类产品推荐特征对应的用户推荐子比例；
    获取所述每一个金融产品的各类产品推荐特征对应的用户推荐权重；其中，各类产品推荐特征对应的用户推荐权重总和为100％；
    根据所述各类产品推荐特征对应的用户推荐子比例以及所述各类产品推荐特征对应的用户推荐权重，获取所述每一个金融产品的用户推荐比例。
  9. 如权利要求1-5任一所述的方法，其中，所述方法还包括：
    获取所述每一个金融产品的用户转化率，所述用户转化率为金融产品对应的被推荐用户中实际使用所述金融产品的用户的数量所占总被推荐用户的数量的比例；
    所述分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，包括：
    根据所述每一个金融产品的设定参数的历史数据，获取所述每一个金融产品的设定参数在所述第一设定时间段内的均值，并根据每一个金融产 品的设定参数在所述第一设定时间段内的均值以及所述用户转化率，构建所述每一个金融产品的产品推荐特征。
  10. 如权利要求1-5任一所述的方法，其中，在所述根据各金融产品的用户推荐比例确定推荐金融产品之后，所述方法还包括：
    将为用户推荐的金融产品的状态数据发送给所述用户，以使得通过用户设备登录所述用户对应的账号后，在所述用户设备的显示页面上显示为所述用户推荐的金融产品的状态数据，所述状态数据包括所述金融产品的名称、收益率。
  11. 一种金融产品推荐装置，所述装置包括：
    特征构建单元，配置为接收客户端的推荐金融产品请求；分别根据N个金融产品中每一个金融产品的设定参数的历史数据，构建每一个金融产品的M类产品推荐特征，N、M均为正整数；
    特征综合单元，配置为分别针对所述M类产品推荐特征中的每一类产品推荐特征，获取与类别对应的综合产品推荐特征；
    推荐比例确定单元，配置为根据每一个金融产品的各类产品推荐特征相对于所述与类别对应的综合产品推荐特征的偏离度，确定每一个金融产品的用户推荐比例，所述用户推荐比例为每一个金融产品所对应的被推荐用户占所有用户的比例；
    产品推荐单元，配置为根据各金融产品的用户推荐比例确定推荐金融产品，以响应所述推荐金融产品请求。
  12. 如权利要求11所述的装置，其中，所述M类产品推荐特征包括如下特征的至少之一：
    所述设定参数在第一设定时间段内的均值；
    所述设定参数在第二设定时间段内的波动率的均值；
    在第二设定时间段内的组合特征的均值，其中，所述组合特征与所述设定参数呈正相关，且与所述设定参数的波动率呈负相关。
  13. 如权利要求11所述的装置，其中，所述推荐比例确定单元配置为：
    根据所述每一个金融产品的各类产品推荐特征对应的偏离度，分别获取各类产品推荐特征对应的用户推荐子比例；
    获取所述每一个金融产品的各类产品推荐特征对应的用户推荐权重；其中，各类产品推荐特征对应的用户推荐权重总和为100％；
    根据所述各类产品推荐特征对应的用户推荐子比例以及所述各类产品推荐特征对应的用户推荐权重，获取所述每一个金融产品的用户推荐比例。
  14. 一种电子设备，包括存储器和处理器；
    其中，所述存储器用于存储计算机程序；
    所述处理器用于执行所述程序时实现如权利要求1至10任一权利要求所述的方法。
  15. 一种计算机可读存储介质，存储有处理器可执行指令，处理器执行所述可执行指令时实现如权利要求1至10任一权利要求所述的方法。
PCT/CN2020/093503 2019-06-06 2020-05-29 金融产品推荐方法、装置、电子设备及计算机存储介质 WO2020244468A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2021541595A JP7430191B2 (ja) 2019-06-06 2020-05-29 金融商品推薦方法、装置、電子機器及びプログラム
US17/337,284 US20210287295A1 (en) 2019-06-06 2021-06-02 Method and apparatus for recommending financial product, electronic device, and computer storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910490545.8 2019-06-06
CN201910490545.8A CN110415123A (zh) 2019-06-06 2019-06-06 金融产品推荐方法、装置和设备及计算机存储介质

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/337,284 Continuation US20210287295A1 (en) 2019-06-06 2021-06-02 Method and apparatus for recommending financial product, electronic device, and computer storage medium

Publications (1)

Publication Number Publication Date
WO2020244468A1 true WO2020244468A1 (zh) 2020-12-10

Family

ID=68358412

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/093503 WO2020244468A1 (zh) 2019-06-06 2020-05-29 金融产品推荐方法、装置、电子设备及计算机存储介质

Country Status (4)

Country Link
US (1) US20210287295A1 (zh)
JP (1) JP7430191B2 (zh)
CN (1) CN110415123A (zh)
WO (1) WO2020244468A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994821A (zh) * 2023-01-09 2023-04-21 中云融拓数据科技发展(深圳)有限公司 基于产业链数字化场景金融模型建立金融风控体系的方法

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110415123A (zh) * 2019-06-06 2019-11-05 财付通支付科技有限公司 金融产品推荐方法、装置和设备及计算机存储介质
CN111681113B (zh) * 2020-05-29 2023-07-18 泰康保险集团股份有限公司 一种基金产品对象配置的系统和服务器
CN114786025B (zh) * 2022-04-01 2024-01-02 北京达佳互联信息技术有限公司 直播数据处理方法、装置、计算机设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106454536A (zh) * 2016-09-19 2017-02-22 广州视源电子科技股份有限公司 信息推荐度的确定方法及装置
US20170337613A1 (en) * 2016-05-23 2017-11-23 Fuji Xerox Co., Ltd. Recording medium, product recommendation system, and product recommendation method
CN108985935A (zh) * 2018-07-06 2018-12-11 兴业证券股份有限公司 金融产品推荐方法及存储介质
CN109461053A (zh) * 2018-10-24 2019-03-12 平安科技（深圳）有限公司 多推荐渠道的动态分流方法、电子装置及存储介质
CN110415123A (zh) * 2019-06-06 2019-11-05 财付通支付科技有限公司 金融产品推荐方法、装置和设备及计算机存储介质

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150363862A1 (en) * 2014-06-13 2015-12-17 Connect Financial LLC Financial product recommendation for a consumer
CN105791157B (zh) * 2016-04-20 2019-10-25 腾讯科技（深圳）有限公司 一种流量的分配方法、分配系统、及服务器
CN105897616B (zh) * 2016-05-17 2020-11-06 腾讯科技（深圳）有限公司 一种资源分配的方法及服务器
CN108156204B (zh) * 2016-12-06 2021-03-12 阿里巴巴集团控股有限公司 一种目标对象推送系统和方法
CN108399565A (zh) * 2017-10-09 2018-08-14 平安科技（深圳）有限公司 金融产品推荐装置、方法及计算机可读存储介质
CN109447728A (zh) * 2018-09-07 2019-03-08 平安科技（深圳）有限公司 金融产品推荐方法、装置、计算机设备及存储介质
CN109300045A (zh) * 2018-10-25 2019-02-01 平安科技（深圳）有限公司 金融产品推荐方法、装置、计算机设备及存储介质
CN109472646A (zh) * 2018-11-16 2019-03-15 广发证券股份有限公司 一种金融产品推荐方法及装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170337613A1 (en) * 2016-05-23 2017-11-23 Fuji Xerox Co., Ltd. Recording medium, product recommendation system, and product recommendation method
CN106454536A (zh) * 2016-09-19 2017-02-22 广州视源电子科技股份有限公司 信息推荐度的确定方法及装置
CN108985935A (zh) * 2018-07-06 2018-12-11 兴业证券股份有限公司 金融产品推荐方法及存储介质
CN109461053A (zh) * 2018-10-24 2019-03-12 平安科技（深圳）有限公司 多推荐渠道的动态分流方法、电子装置及存储介质
CN110415123A (zh) * 2019-06-06 2019-11-05 财付通支付科技有限公司 金融产品推荐方法、装置和设备及计算机存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115994821A (zh) * 2023-01-09 2023-04-21 中云融拓数据科技发展(深圳)有限公司 基于产业链数字化场景金融模型建立金融风控体系的方法

Also Published As

Publication number Publication date
US20210287295A1 (en) 2021-09-16
CN110415123A (zh) 2019-11-05
JP7430191B2 (ja) 2024-02-09
JP2022535636A (ja) 2022-08-10

Similar Documents

Publication Publication Date Title
WO2020244468A1 (zh) 金融产品推荐方法、装置、电子设备及计算机存储介质
US11928652B2 (en) Electronic capital marketplace systems and methods
US9947059B2 (en) Group formation and dynamic pricing for E-commerce in social networks
WO2019040712A1 (en) METHOD AND SYSTEM FOR AUCTION AT DECENTRALIZED MARKET
JP2019535080A (ja) 逆入札型オークションのためのシステムおよび方法
US20220130005A1 (en) Digital asset management systems and methods
US20180315025A1 (en) Technology adapted to configure computer systems to perform management, and enable streamlined user-configuration, of complex autonomous peer-to-peer transaction frameworks
JP2023134689A (ja) セルサイドのマーケットメイキングを促進するプロセス間通信
CN111815372A (zh) 基于区块链的直播处理方法、装置、电子设备及存储介质
US20180357715A1 (en) System and Method For a Virtual Currency Exchange
Huang et al. Virtual standard currency for approximating foreign exchange rates
US8788364B1 (en) System for configuration and implementation of an assignment auction or exchange
US20130173427A1 (en) Apparatus and method of data sharing between online marketplaces
TWI656490B (zh) Method and data collection and distribution system for big data commodity customization and data provider separation, computer readable recording media and computer program products
US20230087580A1 (en) Methods, systems, and computer readable media for providing decentralized finance over blockchains
US20150127517A1 (en) Methods and apparatus for facilitating fairnetting and distribution of currency trades
US20150178840A1 (en) Systems and related techniques for fairnetting and distribution of electronic trades
Malakani et al. Trading 4.0: An online peer-to-peer money lending platform
KR20200048401A (ko) 광고 트래픽 거래 서비스 제공 방법 및 그 장치
KR20190085616A (ko) 암호 화폐 지급 서비스 방법 및 서버
US20230283469A1 (en) Systems and methods for validating transfers between cryptographic addresses
Deck et al. Fixed revenue auctions: Theory and behavior
US20230351463A1 (en) Computer-implemented bidding method, computer equipment and storage medium
WO2015183993A1 (en) Systems and methods for collaborative commerce
CN109299385A (zh) 一种利用支付令牌进行支付方式推荐的方法及其装置

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: 20819263

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021541595

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20819263

Country of ref document: EP

Kind code of ref document: A1