Information recommendation method, system and recommendation server
Through the collaborative work of recommendation servers and trading servers, using user and product characteristics, combining history and simulated transaction information, the target investment evaluation value of investment strategies is calculated, and the personalization and efficiency of investment strategy recommendations in the financial field is solved, achieving efficient Investment strategy recommendation.
Patent Information
- Application Number
- CN202210524498.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-13
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2042-05-13
AI Technical Summary
In the financial field, it is difficult for the existing technology to effectively apply big data processing technology to recommend investment strategies, resulting in large labor costs and the inability to provide personalized recommendation services.
Through the coordinated work of the recommendation server and the trading server, the target user identification and configuration information sent by the client can obtain user characteristics and product characteristics, calculate the similarity, combine historical trading scenario information and simulated trading information, and calculate the target investment evaluation of the investment strategy. Value, recommend personalized investment strategies.
It has achieved personalized investment strategy recommendations for target users, reduced the amount of subsequent data processing, improved the efficiency of information recommendation, and reduced labor costs.
Smart Images

Figure CN114943582B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of big data processing, and in particular to an information recommendation method, system and recommendation server. Background Art
[0002] At present, big data processing technology has been applied to various fields. An important application of big data processing technology in various fields is to generate recommendation information in the corresponding field based on a large amount of historical data and output it to users. For example, the video system can generate recommended videos based on historical data and output them to users; for example, in the field of e-commerce, it is also possible to generate product recommendation information based on historical data and output it to users, etc.
[0003] However, big data processing technology has not been widely used in the financial field. For example, when recommending investment strategies, professionals still manually conduct quantitative analysis of historical data to obtain target investment strategies and recommend them to users. Obviously, this method requires a large labor cost and cannot provide personalized recommendation services for each user. Summary of the invention
[0004] The purpose of the embodiments of the present invention is to provide an information recommendation method, system and recommendation server to provide personalized recommendation services for each user and improve the efficiency of information recommendation. The specific technical solution is as follows:
[0005] In a first aspect, in order to achieve the above-mentioned object, an embodiment of the present invention provides an information recommendation method, the method is applied to a recommendation server in a recommendation system, the recommendation system also includes a client; the recommendation server is communicatively connected with a transaction server, the method includes:
[0006] Receive an information recommendation request sent by the client; wherein the information recommendation request includes a target user identifier and configuration information of the information to be recommended; in the case where the information to be recommended is determined to be an investment strategy based on the configuration information, obtain multiple investment strategies to be recommended based on the configuration information; send a corresponding user feature acquisition request to the transaction server based on the target user identifier; so that the transaction server obtains the user features of the target user based on the historical investment data of the target user, and returns the request to the recommendation server; for each investment strategy to be recommended, send a corresponding product feature acquisition request to the transaction server; so that the transaction server obtains the product features of the investment product corresponding to the investment strategy to be recommended, and returns the request to the recommendation server; calculate the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, respectively, as the similarity between the investment strategy to be recommended and the target user; obtain the investment strategy to be recommended whose similarity with the target user meets the preset selection condition, and for candidate investment strategies; for each candidate investment strategy, sending a corresponding investment evaluation value acquisition request to the transaction server; so that the transaction server obtains historical transaction scenario information within a first historical time period in response to the investment evaluation value acquisition request, and calculates the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtains second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and based on the first transaction information and the second transaction information, calculates the target investment evaluation value of the candidate investment strategy and returns it to the recommendation server; from each candidate investment strategy, determines the target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions; sends the target investment strategy as recommendation information to the client, so that the client displays the recommendation information.
[0007] Optionally, the configuration information includes: configuration parameter values of various strategy configuration items used to determine the investment strategy; the configuration parameter values are input by the target user through a configuration page; and a plurality of strategy configuration items are displayed on the configuration page;
[0008] In the case where it is determined that the information to be recommended is an investment strategy based on the configuration information, a plurality of investment strategies to be recommended are obtained based on the configuration information, including: generating a plurality of investment strategies to be recommended based on the configuration parameter values of each strategy configuration item input by the target user; wherein each investment strategy to be recommended includes screening conditions and transaction methods.
[0009] Optionally, the configuration information further includes: a plurality of target evaluation indicators determined by the target user through the configuration page; the investment evaluation value acquisition request includes the plurality of target evaluation indicators, so that the transaction server determines the target investment evaluation value of the candidate investment strategy based on the target evaluation indicators, the first transaction information and the second transaction information;
[0010] or,
[0011] The investment evaluation value acquisition request includes the target user identifier, so that the transaction server determines an evaluation indicator that matches the user characteristics of the target user from the preset evaluation indicators as the target evaluation indicator; based on the target evaluation indicator, the first transaction information and the second transaction information, the target investment evaluation value of the candidate investment strategy is determined.
[0012] Optionally, the configuration information includes: a strategy identifier of an investment strategy;
[0013] In the case where it is determined based on the configuration information that the information to be recommended is an investment strategy, a plurality of investment strategies to be recommended are obtained based on the configuration information, including: determining the investment strategy to which the strategy identifier belongs from each investment strategy recorded locally, and obtaining a plurality of investment strategies to be recommended; wherein each investment strategy to be recommended includes screening conditions and transaction methods.
[0014] Optionally, respectively calculating the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user as the similarity between the investment strategy to be recommended and the target user includes: for each investment strategy to be recommended, performing mapping processing on the product features corresponding to the investment strategy to be recommended to obtain a first feature vector corresponding to the investment strategy to be recommended; performing mapping processing on the user features of the target user to obtain a second feature vector corresponding to the target user; calculating the similarity between the first feature vector and the second feature vector to obtain the similarity between the product features corresponding to the investment strategy to be recommended and the user features of the target user as the similarity between the investment strategy to be recommended and the target user;
[0015] The step of obtaining the investment strategy to be recommended, whose similarity with the target user satisfies the preset selection condition, as the candidate investment strategy includes:
[0016] Determining a preset number of investment strategies to be recommended from among the investment strategies to be recommended as candidate investment strategies in descending order of similarity with the target user;
[0017] or,
[0018] From among the investment strategies to be recommended, an investment strategy to be recommended whose similarity with the target user is greater than a preset threshold is determined as a candidate investment strategy.
[0019] In a second aspect, in order to achieve the above-mentioned purpose, an embodiment of the present invention provides an information recommendation method, the method is applied to a transaction server, the transaction server is communicatively connected with a recommendation server in a recommendation system; the recommendation system also includes a client; the method includes:
[0020] Upon receiving a user feature acquisition request sent by the recommendation server based on the target user identifier included in the information recommendation request, the user features of the target user are obtained based on the historical investment data of the target user and returned to the recommendation server; wherein the information recommendation request is sent by the client to the recommendation server, and the information recommendation request includes the target user identifier and the configuration information of the information to be recommended; for each investment strategy to be recommended, upon receiving a product feature acquisition request sent by the recommendation server, the product features of the investment product corresponding to the investment strategy to be recommended are obtained and returned to the recommendation server; so that the recommendation server calculates the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, respectively, as the similarity between the investment strategy to be recommended and the target user; obtains the investment strategy to be recommended whose similarity with the target user meets the preset selection condition as a candidate investment strategy; wherein the investment strategy to be recommended is: the recommendation server determines that the information to be recommended is an investment strategy based on the configuration information In the case of omission, it is obtained based on the configuration information; for each candidate investment strategy, when receiving the investment evaluation value acquisition request sent by the recommendation server, for the investment evaluation value acquisition request, obtain the historical transaction scenario information within the first historical time period, and calculate the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein, the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtain the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy; wherein, the second historical time period is a time period for simulated transactions based on the candidate investment strategy; based on the first transaction information and the second transaction information, calculate the target investment evaluation value of the candidate investment strategy, and return it to the recommendation server; so that the recommendation server determines the target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions from each candidate investment strategy; send the target investment strategy as recommendation information to the client, so that the client displays the recommendation information.
[0021] Optionally, the obtaining of historical transaction scenario information within the first historical time period, and calculating the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information, includes: obtaining product information of each investment product in the historical transaction scenario within the first historical time period as the first product information; determining, from among the investment products in the historical transaction scenario within the first historical time period, each investment product that matches the screening condition in the candidate investment strategy as the first investment product; and calculating, based on the transaction method in the candidate investment strategy and the first product information of the first investment product, transaction information that meets the target evaluation indicator when investing in accordance with the candidate investment strategy within the first historical time period as the first transaction information;
[0022] The obtaining of second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy includes: obtaining product information of each investment product in a historical transaction scenario within the second historical time period as second product information; determining, from among the investment products in the historical transaction scenarios within the second historical time period, each investment product that matches a screening condition in the candidate investment strategy as a second investment product; and performing simulated transactions within the second historical time period based on a transaction method in the candidate investment strategy and the second product information of the second investment product to obtain transaction information that meets the target evaluation index as second transaction information.
[0023] Optionally, the investment evaluation value acquisition request carries multiple target evaluation indicators;
[0024] The method of calculating the target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information includes: for each target evaluation indicator, based on the first transaction information that meets the target evaluation indicator, calculating the indicator value of the target evaluation indicator as the first indicator value; calculating the weighted sum of the first indicator values of each target evaluation indicator as the first investment evaluation value; based on the second transaction information that meets the target evaluation indicator, calculating the indicator value of the target evaluation indicator as the second indicator value; calculating the weighted sum of the second indicator values of each target evaluation indicator as the second investment evaluation value; and calculating the weighted sum of the first investment evaluation value and the second investment evaluation value to obtain the target investment evaluation value of the candidate investment strategy.
[0025] In a third aspect, in order to achieve the above-mentioned purpose, an embodiment of the present invention provides an information recommendation system, the recommendation system comprising: a client and a recommendation server; the recommendation server is communicatively connected with a transaction server;
[0026] The client is used to send an information recommendation request including a target user identifier and configuration information of the information to be recommended to the recommendation server; the recommendation server is used to receive the information recommendation request; in the case where the information to be recommended is determined to be an investment strategy based on the configuration information, a plurality of investment strategies to be recommended are obtained based on the configuration information; a corresponding user feature acquisition request is sent to the transaction server based on the target user identifier; the transaction server is used to obtain the user features of the target user based on the historical investment data of the target user, and return the same to the recommendation server; the recommendation server is also used to send a corresponding product feature acquisition request to the transaction server for each investment strategy to be recommended; the transaction server is also used to obtain the product features of the investment product corresponding to the investment strategy to be recommended, and return the same to the recommendation server; the recommendation server is also used to respectively calculate the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user; the similarity obtained with the target user satisfies a preset The investment strategy to be recommended according to the selection conditions is selected as a candidate investment strategy; for each candidate investment strategy, a corresponding investment evaluation value acquisition request is sent to the transaction server; the transaction server is also used to obtain historical transaction scenario information within a first historical time period in response to the investment evaluation value acquisition request, and calculate the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtain the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and based on the first transaction information and the second transaction information, a target investment evaluation value of the candidate investment strategy is calculated and returned to the recommendation server; the recommendation server is also used to determine, from each candidate investment strategy, a target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions; and send the target investment strategy as recommendation information to the client; the client is also used to display the recommendation information.
[0027] In a fourth aspect, in order to achieve the above-mentioned object, an embodiment of the present invention provides a recommendation server, wherein the recommendation server includes:
[0028] an information recommendation request receiving module, used for receiving an information recommendation request sent by the client; wherein the information recommendation request includes a target user identifier and configuration information of the information to be recommended; a module for obtaining investment strategies to be recommended, used for obtaining multiple investment strategies to be recommended based on the configuration information when the information to be recommended is determined to be an investment strategy based on the configuration information; a user feature obtaining module, used for sending a corresponding user feature obtaining request to the transaction server based on the target user identifier; so that the transaction server obtains the user features of the target user based on the transaction data of the target user's historical investments, and returns the same to the recommendation server; a product feature obtaining module, used for sending a corresponding product feature obtaining request to the transaction server for each investment strategy to be recommended; so that the transaction server obtains the product features of the investment product corresponding to the investment strategy to be recommended, and returns the same to the recommendation server; a similarity calculation module, used for respectively calculating the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user; a candidate investment strategy obtaining module, used for obtaining the product features of the investment product corresponding to the target user The investment strategies to be recommended whose similarity meets the preset selection conditions are used as candidate investment strategies; an investment evaluation value acquisition module is used to send a corresponding investment evaluation value acquisition request to the transaction server for each candidate investment strategy; so that the transaction server obtains historical transaction scenario information within a first historical time period in response to the investment evaluation value acquisition request, and calculates the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtains the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and based on the first transaction information and the second transaction information, calculates the target investment evaluation value of the candidate investment strategy and returns it to the recommendation server; a target investment strategy acquisition module is used to determine, from each candidate investment strategy, a target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions; a recommendation information output module is used to send the target investment strategy as recommendation information to the client, so that the client displays the recommendation information.
[0029] Optionally, the configuration information includes: configuration parameter values of various strategy configuration items used to determine the investment strategy; the configuration parameter values are input by the target user through a configuration page; and a plurality of strategy configuration items are displayed on the configuration page;
[0030] The investment strategy acquisition module to be recommended is specifically used to generate multiple investment strategies to be recommended based on the configuration parameter values of each strategy configuration item input by the target user; wherein each investment strategy to be recommended includes a screening condition and a transaction method.
[0031] Optionally, the configuration information further includes: a plurality of target evaluation indicators determined by the target user through the configuration page; the investment evaluation value acquisition request includes the plurality of target evaluation indicators, so that the transaction server determines the target investment evaluation value of the candidate investment strategy based on the target evaluation indicators, the first transaction information and the second transaction information;
[0032] or,
[0033] The investment evaluation value acquisition request includes the target user identifier, so that the transaction server determines an evaluation indicator that matches the user characteristics of the target user from the preset evaluation indicators as the target evaluation indicator; based on the target evaluation indicator, the first transaction information and the second transaction information, the target investment evaluation value of the candidate investment strategy is determined.
[0034] Optionally, the configuration information includes: a strategy identifier of an investment strategy;
[0035] The investment strategy acquisition module to be recommended is specifically used to determine the investment strategy to which the strategy identifier belongs from the various investment strategies recorded locally, and obtain multiple investment strategies to be recommended; wherein each investment strategy to be recommended includes a screening condition and a transaction method.
[0036] Optionally, the similarity calculation module is specifically used to, for each investment strategy to be recommended, perform mapping processing on the product features corresponding to the investment strategy to be recommended, to obtain a first feature vector corresponding to the investment strategy to be recommended; perform mapping processing on the user features of the target user, to obtain a second feature vector corresponding to the target user; calculate the similarity between the first feature vector and the second feature vector, to obtain the similarity between the product features corresponding to the investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user;
[0037] The candidate investment strategy acquisition module is specifically used to determine a preset number of investment strategies to be recommended from the investment strategies to be recommended as candidate investment strategies in descending order of similarity with the target user;
[0038] or,
[0039] From among the investment strategies to be recommended, an investment strategy to be recommended whose similarity with the target user is greater than a preset threshold is determined as a candidate investment strategy.
[0040] In a fifth aspect, in order to achieve the above-mentioned object, an embodiment of the present invention provides a transaction server, the transaction server comprising:
[0041] a user feature acquisition module, for, upon receiving a user feature acquisition request sent by the recommendation server based on the target user identifier included in the information recommendation request, obtaining the user features of the target user based on the historical investment data of the target user, and returning the same to the recommendation server; wherein the information recommendation request is sent by the client to the recommendation server, and the information recommendation request includes the target user identifier and the configuration information of the information to be recommended; a product feature acquisition module, for, for each investment strategy to be recommended, upon receiving a product feature acquisition request sent by the recommendation server, obtaining the product features of the investment product corresponding to the investment strategy to be recommended, and returning the same to the recommendation server; so that the recommendation server calculates the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, respectively, as the similarity between the investment strategy to be recommended and the target user; obtaining the investment strategy to be recommended whose similarity with the target user meets the preset selection condition, as a candidate investment strategy; wherein the investment strategy to be recommended is: the recommendation server determines that the information to be recommended is an investment strategy based on the configuration information In the case of omission, the information is obtained based on the configuration information; a transaction information acquisition module is used for, for each candidate investment strategy, when receiving the investment evaluation value acquisition request sent by the recommendation server, obtaining the historical transaction scenario information within the first historical time period for the investment evaluation value acquisition request, and calculating the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtaining the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; an investment evaluation value acquisition module is used to calculate the target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information, and return it to the recommendation server; so that the recommendation server determines the target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions from each candidate investment strategy; and sends the target investment strategy as recommendation information to the client, so that the client displays the recommendation information.
[0042] Optionally, the transaction information acquisition module is specifically used to acquire product information of each investment product in the historical transaction scenario within the first historical time period as the first product information; determine each investment product that matches the screening condition in the candidate investment strategy from the investment products in the historical transaction scenario within the first historical time period as the first investment product; based on the transaction method in the candidate investment strategy and the first product information of the first investment product, calculate the transaction information that meets the target evaluation indicator when investing according to the candidate investment strategy within the first historical time period as the first transaction information;
[0043] The transaction information acquisition module is specifically used to obtain product information of each investment product in the historical transaction scenarios within the second historical time period as the second product information; from each investment product in the historical transaction scenarios within the second historical time period, determine each investment product that matches the screening conditions in the candidate investment strategy as the second investment product; within the second historical time period, simulate transactions based on the transaction method in the candidate investment strategy and the second product information of the second investment product to obtain transaction information that meets the target evaluation index as the second transaction information.
[0044] Optionally, the investment evaluation value acquisition request carries multiple target evaluation indicators;
[0045] The investment evaluation value acquisition module is specifically used to calculate, for each target evaluation indicator, the indicator value of the target evaluation indicator based on the first transaction information that meets the target evaluation indicator, as the first indicator value; calculate the weighted sum of the first indicator values of each target evaluation indicator as the first investment evaluation value; calculate the indicator value of the target evaluation indicator based on the second transaction information that meets the target evaluation indicator as the second indicator value; calculate the weighted sum of the second indicator values of each target evaluation indicator as the second investment evaluation value; calculate the weighted sum of the first investment evaluation value and the second investment evaluation value to obtain the target investment evaluation value of the candidate investment strategy.
[0046] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the information recommendation method steps described in any one of the first aspect or the second aspect are implemented.
[0047] An embodiment of the present invention further provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the information recommendation method described in any one of the first aspect or the second aspect.
[0048] According to the technical solution provided by the embodiment of the present invention, the client sends an information recommendation request including a target user identifier and configuration information of the information to be recommended to the recommendation server. When the recommendation server determines that the information to be recommended is an investment strategy based on the configuration information in the received information recommendation request, it obtains multiple investment strategies to be recommended based on the configuration information; and sends a corresponding user feature acquisition request to the transaction server based on the target user identifier. The transaction server obtains the user features of the target user based on the historical investment data of the target user and returns them to the recommendation server. For each investment strategy to be recommended, the recommendation server sends a corresponding product feature acquisition request to the transaction server. The transaction server obtains the product features of the investment product corresponding to the investment strategy to be recommended and returns them to the recommendation server. The recommendation server calculates the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user respectively, as the similarity between the investment strategy to be recommended and the target user; obtains the investment strategy to be recommended whose similarity with the target user meets the preset selection condition as a candidate investment strategy; and sends a corresponding investment evaluation value acquisition request to the transaction server for each candidate investment strategy. In response to the investment evaluation value acquisition request, the transaction server obtains the historical transaction scenario information in the first historical time period, and calculates the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information. The first historical time period is a time period earlier than the moment when the information recommendation request is received; and the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy is obtained; the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and based on the first transaction information and the second transaction information, the target investment evaluation value of the candidate investment strategy is calculated and returned to the recommendation server. The recommendation server determines the target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions from each candidate investment strategy; and sends the target investment strategy as recommendation information to the client. The client displays the recommendation information.
[0049] Based on the above processing, based on the user characteristics of the target user, the candidate investment strategies whose similarity with the user characteristics meets the preset selection conditions are determined, and the target investment strategies recommended to the user are determined from the candidate investment strategies. The target investment strategies are investment strategies that meet the real needs of the target users. It is possible to provide personalized recommendation services to the target users, and the candidate investment strategies are determined from the investment strategies to be recommended, which can reduce the amount of data for subsequent processing and improve the efficiency of information recommendation.
[0050] Of course, it is not necessary to achieve all of the advantages described above at the same time to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.
[0052] Figure 1 A structural diagram of an information recommendation system provided by an embodiment of the present invention;
[0053] Figure 2 A flowchart of an information recommendation method provided by an embodiment of the present invention;
[0054] Figure 3 A flowchart of another information recommendation method provided by an embodiment of the present invention;
[0055] Figure 4 A specific flow chart of obtaining transaction information in an embodiment of the present invention;
[0056] Figure 5 A specific flow chart for determining an investment evaluation value in an embodiment of the present invention;
[0057] Figure 6 A schematic diagram of a configuration page displayed by a client in an embodiment of the present invention;
[0058] Figure 7 A schematic diagram of the principle of an information recommendation method provided by an embodiment of the present invention;
[0059] Figure 8 A structural diagram of a recommendation server provided in an embodiment of the present invention;
[0060] Fig. 9 A structural diagram of a transaction server provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field based on this application belong to the scope of protection of the present invention.
[0062] In the related technologies, big data processing technology has not been widely used in the financial field. For example, when recommending investment strategies, professionals still manually conduct quantitative analysis on historical data to obtain target investment strategies and recommend them to users. This method requires high labor costs and cannot provide personalized recommendation services for each user.
[0063] To solve the above problems, see Figure 1 , Figure 1 The structure diagram of an information recommendation system provided by an embodiment of the present invention includes a client 101 and a recommendation server 102 , and the recommendation server 102 is connected to a transaction server 103 in communication. Figure 1 The connection relationship between the client 101, the recommendation server 102 and the transaction server 103 in the figure represents the logical relationship when implementing the information recommendation method provided in the embodiment of the present invention. In actual implementation, the recommendation server 102 and the transaction server 103 can be different servers in the server cluster. In this case, the recommendation server 102 and the transaction server 103 are connected to each other through an external network. Alternatively, the recommendation server 102 and the transaction server 103 can also be deployed on the same physical device. In this case, the recommendation server 102 and the transaction server 103 can be connected to each other through an internal communication protocol.
[0064] The client can be a terminal such as a mobile phone or a computer. A program for obtaining investment strategies is running in the client. For example, when the client is a mobile phone, the program for obtaining investment strategies is installed in the mobile phone in the form of an APP (Application). When the client is a computer, the program for obtaining investment strategies runs in the form of a web page in the computer's web browser.
[0065] When the target user needs to obtain an investment strategy, the target user can input a strategy acquisition instruction to the client. For example, an "obtain investment strategy" operation button may be provided in the display interface of the client, and the user may click the "obtain investment strategy" operation button to input a strategy acquisition instruction to the client. Upon receiving the strategy acquisition instruction, the client may send a corresponding information recommendation request to the recommendation server. Accordingly, the recommendation server may interact with the transaction server based on the method provided in the embodiment of the present invention, determine the target investment strategy recommended to the target user, and send recommendation information containing the target investment strategy to the client. Furthermore, the client may display the recommended information to the user in the display interface for the user to browse.
[0066] Based on the information recommendation system provided by the embodiment of the present invention, a program for obtaining investment strategies is formed in the form of code, and the program for obtaining investment strategies is provided to users. The user only needs to enter a strategy acquisition instruction in the client running the program for obtaining investment strategies, and the client can display the target investment strategy to the user, and the target investment strategy is an investment strategy that meets the real needs of the target user, which can provide personalized recommendation services for the target user and improve the efficiency of information recommendation.
[0067] See also Figure 2 , Figure 2A flowchart of an information recommendation method provided by an embodiment of the present invention is applied to an information recommendation system, the recommendation system comprising: a client and a recommendation server; the recommendation server is connected in communication with a transaction server, and the method may include the following steps:
[0068] S201: The client sends an information recommendation request including a target user identifier and configuration information of the information to be recommended to a recommendation server.
[0069] S202: The recommendation server receives an information recommendation request; when it is determined based on the configuration information that the information to be recommended is an investment strategy, a plurality of investment strategies to be recommended are obtained based on the configuration information.
[0070] S203: The recommendation server sends a corresponding user feature acquisition request to the transaction server based on the target user identifier.
[0071] S204: The transaction server obtains user characteristics of the target user based on the historical investment data of the target user.
[0072] S205: The transaction server returns the user characteristics of the target user to the recommendation server.
[0073] S206: The recommendation server sends a corresponding product feature acquisition request to the transaction server for each investment strategy to be recommended.
[0074] S207: The transaction server obtains product characteristics of the investment product corresponding to the investment strategy to be recommended.
[0075] S208: The transaction server returns the product features corresponding to the investment strategy to be recommended to the recommendation server.
[0076] S209: The recommendation server calculates the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user respectively as the similarity between the investment strategy to be recommended and the target user; obtains the investment strategy to be recommended whose similarity with the target user meets the preset selection condition as the candidate investment strategy.
[0077] S210: The recommendation server sends a corresponding investment evaluation value acquisition request to the transaction server for each candidate investment strategy.
[0078] S211: The transaction server obtains historical transaction scenario information within a first historical time period in response to a request for obtaining an investment evaluation value, and calculates first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; and obtains second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy; and calculates a target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information.
[0079] The first historical time period is a time period earlier than the moment when the information recommendation request is received; and the second historical time period is a time period for simulated transactions based on the candidate investment strategy.
[0080] S212: The transaction server returns the target investment evaluation value of the candidate investment strategy to the recommendation server.
[0081] S213: The recommendation server determines, from among the candidate investment strategies, a target investment strategy whose corresponding target investment evaluation value satisfies a preset recommendation condition.
[0082] S214: The recommendation server sends the target investment strategy as recommendation information to the client.
[0083] S215: The client displays the recommendation information.
[0084] Based on the information recommendation method provided in the embodiment of the present invention, based on the user characteristics of the target user, the candidate investment strategies whose similarity with the user characteristics of the user meets the preset selection conditions are determined, and the target investment strategy recommended to the user is determined from the candidate investment strategies. The target investment strategy is an investment strategy that meets the real needs of the target user. It can provide personalized recommendation services for the target user, and the candidate investment strategy is determined from the investment strategies to be recommended, which can reduce the amount of data to be processed subsequently and improve the efficiency of information recommendation.
[0085] For step S201 and step S202, the information to be recommended may be: videos, music, commodities, and investment strategies, etc. The target user may be any user who currently needs information recommendation, and the target user identifier may be the name, mobile phone number, and account number of the target user for logging into the client, etc. An investment strategy includes: screening conditions for selecting investment products and a trading method for investment. Investment products may include: stocks, funds, treasury bonds, futures, options, convertible bonds, etc. The trading method may include: the number, time, and amount of funds for buying investment products that meet the screening, and the number, time, and amount of funds for selling the purchased investment products, etc.
[0086] When the target user needs to obtain recommended information, the target user can input an information obtaining instruction to the client. Upon receiving the information obtaining instruction, the client can send an information recommendation request including the target user identifier and configuration information of the information to be recommended to the recommendation server.
[0087] In one implementation, the configuration information includes: a policy identifier of the investment policy. Accordingly, step S202 may include the following steps: the recommendation server determines the investment policy to which the policy identifier belongs from each locally recorded investment policy, and obtains a plurality of investment policies to be recommended; wherein each investment policy to be recommended includes a screening condition and a transaction method.
[0088] When the client receives the information acquisition instruction, the strategy identifiers of the preset investment strategies can be displayed. The target user can select the investment strategy that he wants to know from the preset investment strategies, and input a selection instruction carrying the strategy identifier of the selected investment strategy to the client. Accordingly, when the client receives the selection instruction for selecting an investment strategy, it sends an information recommendation request carrying the strategy identifier of the investment strategy selected by the user to the recommendation server. Accordingly, the recommendation server can determine the investment strategy to which the strategy identifier carried in the information recommendation request belongs from the multiple investment strategies recorded locally as the investment strategy to be recommended. The investment strategies recorded in the recommendation server can be set by professionals who understand financial investment knowledge (for example, investment analysis specialists, investment managers, and staff of securities companies).
[0089] In another implementation, the configuration information of the information to be recommended includes: configuration parameter values of various strategy configuration items used to determine the investment strategy; the configuration parameter values are input by the target user through a configuration page; and a plurality of strategy configuration items are displayed in the configuration page.
[0090] Accordingly, step S202 may include the following steps: the recommendation server generates a plurality of investment strategies to be recommended based on the configuration parameter values of each strategy configuration item input by the target user; wherein each investment strategy to be recommended includes a product screening condition and a transaction method.
[0091] The strategy configuration items are used to generate investment strategies, and the strategy configuration items may include: strategy configuration items for determining the screening conditions of investment products (which may be referred to as first strategy configuration items), and strategy configuration items for determining the transaction method for investment (which may be referred to as second strategy configuration items). For example, the first strategy configuration item may include: the opening price, price-to-book ratio, operating profit, K-line chart type, etc. corresponding to the investment product, and the second strategy configuration item may include: the number, time, and amount of funds purchased, the number, time, and amount of funds sold, etc. of the investment product.
[0092] When the client receives the strategy acquisition instruction, it can display a configuration page containing multiple strategy configuration items. The target user can then enter the configuration parameter values of each strategy configuration item in the configuration page. For example, the user can enter the corresponding opening price of the investment product greater than 1.01 yuan, the price-to-book ratio greater than 2 times, the operating profit greater than 10 million, the K-line chart type is three crows, and the number of investment products purchased is 3, the time is each trading day, the amount of funds is 500,000, the number of sales is 1, the time is every 3 trading days, the amount of funds is 100,000, etc.
[0093] Furthermore, the client can send configuration information including configuration parameter values of each strategy configuration item to the recommendation server. Accordingly, the recommendation server can determine the screening conditions and transaction methods including the configuration parameter values of each strategy configuration item based on the configuration parameter values of each strategy configuration item input by the user, and thus obtain the corresponding investment strategy as the investment strategy to be recommended.
[0094] Based on the above processing, the client can flexibly configure the investment strategy according to the user's instructions and obtain the investment strategy that meets the user's needs, which can meet the user's personalized needs for configuring the investment strategy and improve the user experience.
[0095] With respect to step S203 and step S204, the recommendation server may send a user acquisition request including the target user identifier to the transaction server to obtain the user characteristics of the target user. When the transaction server receives the user characteristics acquisition request, it may obtain the user characteristics of the target user based on the historical investment data of the target user.
[0096] The historical investment data of the target user may include the product characteristics of the investment products that the target user has historically invested in. The user characteristics of the target user include: the risk level, product type, and return situation that the target user prefers when making investments. The transaction server may determine the user characteristics of the target user based on the product characteristics of the investment products that the target user has historically invested in. For example, the investment products that the target user has historically invested in are investment products with a low risk level. The transaction server may determine that the user characteristics of the target user include: the risk type is a stable type, indicating that the target user has a low tolerance for investment risks. Alternatively, the investment products that the target user has historically invested in are technology-related investment products. The transaction server may determine that the user characteristics of the target user include: the product type is technology-related, indicating that the target user prefers technology-related investment products.
[0097] With respect to step S206 and step S207, for each investment strategy to be recommended, the recommendation server may send a product feature acquisition request for the investment strategy to be recommended to the transaction server to obtain the product features corresponding to the investment strategy to be recommended. When the transaction server receives the product feature acquisition request, it may determine the investment product corresponding to the screening condition contained in the investment strategy to be recommended (i.e., the first investment product and the second investment product in the subsequent embodiments), and obtain the feature label of the investment product as the product feature corresponding to the investment strategy to be recommended. The feature label of an investment product may include: the name, code, sector, industry, company, investment risk level, cumulative rate of return, annualized volatility, the amount of funds held by the target user in the investment product, and the holding time.
[0098] In one embodiment, Figure 2 Based on Figure 3, step S209 may include the following steps:
[0099] S2091: The recommendation server performs mapping processing on the product features corresponding to each investment strategy to be recommended, and obtains a first feature vector corresponding to the investment strategy to be recommended.
[0100] S2092: The recommendation server performs mapping processing on the user features of the target user to obtain a second feature vector corresponding to the target user.
[0101] S2093: The recommendation server calculates the similarity between the first feature vector and the second feature vector, and obtains the similarity between the product feature corresponding to the investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user.
[0102] S2094: The recommendation server determines a preset number of investment strategies to be recommended from the investment strategies to be recommended as candidate investment strategies in descending order of similarity with the target user. Alternatively, the recommendation server determines, from the investment strategies to be recommended, investment strategies whose similarity with the target user is greater than a preset threshold as candidate investment strategies.
[0103] For each investment strategy to be recommended, the recommendation server can map the product features corresponding to the investment strategy to be recommended according to a preset encoding method to obtain a first feature vector corresponding to the investment strategy to be recommended. And map the user features of the target user according to a preset encoding method to obtain a second feature vector corresponding to the target user. The preset encoding method can be One-hot encoding, or Embedding encoding, etc.
[0104] The recommendation server may calculate the similarity between the first feature vector and the second feature vector based on a preset similarity algorithm to obtain the similarity between the investment strategy to be recommended and the target user. The preset similarity algorithm may be a cosine similarity algorithm or a Euclidean distance algorithm.
[0105] In one implementation, the recommendation server may determine a preset number of investment strategies to be recommended from among the investment strategies to be recommended as candidate investment strategies in descending order of similarity with the target user.
[0106] In another implementation, the recommendation server determines, from among the investment strategies to be recommended, an investment strategy to be recommended whose similarity with the target user is greater than a preset threshold as a candidate investment strategy.
[0107] With respect to step S210 and step S211, for each candidate investment strategy, the recommendation server may send a corresponding investment evaluation value acquisition request to the transaction server. Accordingly, upon receiving the investment evaluation value acquisition request, the transaction server may acquire the first transaction information and the second transaction information corresponding to the candidate investment strategy to determine the target investment evaluation value of the candidate investment strategy, and return it to the recommendation server.
[0108] The first historical time period is a time period earlier than the moment when the information recommendation request is received; the second historical time period is a time period for simulated trading based on the candidate investment strategy. That is, the first historical time period is earlier than the second historical time period, and the length of the first historical time period and the length of the second historical time period can be set based on demand. For each candidate investment strategy, when the target user configures the candidate investment strategy on the client, the first historical time period can be set as the time period before the moment when the candidate investment strategy is configured, and the second historical time period can be set as the time period after the moment when the candidate investment strategy is configured.
[0109] For example, if the user configures a candidate investment strategy on February 1, 2022, the first historical time period can be set to January 1, 2021 to December 31, 2021, and the second historical time period can be set to February 2, 2022 to April 1, 2022. On February 1, 2022, the trading server can calculate the first transaction information that meets the target evaluation index when investing in the investment market according to the candidate investment strategy during the first historical time period. The trading server can start running the candidate investment strategy in real time from February 2, 2022 (i.e., the start time of the second historical time period), that is, based on the second product information of the second investment product, simulate transactions in real time according to the candidate investment strategy until April 1, 2022 (i.e., the end time of the second historical time period), and obtain the second transaction information that meets the target evaluation index when investing in the investment market according to the candidate investment strategy during the second historical time period.
[0110] In one embodiment, see Figure 4 , Figure 4 This is a specific flow chart of obtaining transaction information in an embodiment of the present invention. The method may include the following steps:
[0111] S401: The transaction server obtains product information of each investment product in a historical transaction scenario within a first historical time period as first product information.
[0112] S402: The transaction server determines, from among the investment products in the historical transaction scenarios within the first historical time period, investment products that match the screening conditions in the candidate investment strategy as first investment products.
[0113] S403: The transaction server calculates, based on the transaction method in the candidate investment strategy and the first product information of the first investment product, transaction information that meets the target evaluation indicator when investing according to the candidate investment strategy in the first historical time period as the first transaction information.
[0114] S404: The transaction server obtains product information of each investment product in the historical transaction scenario within the second historical time period as second product information.
[0115] S405: The transaction server determines, from among the investment products in the historical transaction scenarios within the second historical time period, investment products that match the screening conditions in the candidate investment strategy as second investment products.
[0116] S406: The transaction server performs simulated transactions within the second historical time period based on the transaction method in the candidate investment strategy and the second product information of the second investment product, and obtains transaction information that meets the target evaluation index as the second transaction information.
[0117] In one implementation, the transaction server may obtain product information (i.e., first product information) of each investment product in the investment market (i.e., historical transaction scenario) within a first historical time period. The product information of an investment product includes: a market index factor representing the price information of the investment product, such as the opening price, closing price, transaction price, and market net value of the investment product on each trading day within a time period. A financial index factor representing the financial information of the company to which the investment product belongs, such as the operating profit and financial statements of the company to which the investment product belongs on each day within a time period. A morphological index factor representing the type of K-line chart of the investment product, such as two crows, three crows, ascending triangle, descending triangle, head and shoulders top, and head and shoulders bottom. A quantity and price index factor representing the statistical characteristics of the price information of the investment product, such as the daily K-line chart, 5-day K-line chart, minute K-line chart, and net value curve of a time period of the investment product. A public opinion index factor representing the public opinion information of the company to which the investment product belongs, such as the number of news related to the company to which the investment product belongs, the number of views, and the user's evaluation of the company to which the investment product belongs. Characteristic data indicator factors that represent the business information of the company to which the investment product belongs, such as the sales of the company's operating products within a period of time, the number of times the company to which the investment product belongs has been sued, and the number of people who pay social security for the company to which the investment product belongs.
[0118] Then, for each candidate investment strategy, the transaction server may determine the first investment product that matches the screening condition included in the candidate investment strategy from the investment products in the investment market during the first historical time period. For example, the screening condition included in the candidate investment strategy is: the opening price is greater than 1.01 yuan, and the transaction server may determine that the investment product with an opening price greater than 1.01 yuan is the first investment product. The transaction server may calculate the first transaction information that meets the target evaluation index when investing according to the candidate investment strategy during the first historical time period based on the candidate investment strategy and the first product information of the first investment product.
[0119] The transaction server determines the second investment product in a similar manner to the first investment product, and reference may be made to the relevant description of the aforementioned embodiment. Furthermore, the transaction server may simulate transactions according to the transaction method and the second product information of the second investment product in the candidate investment strategy from the start of the second historical time period until the end of the second time period, and obtain the second transaction information that meets the target evaluation index when investing in the investment market according to the candidate investment strategy during the second historical time period.
[0120] The transaction information within a time period includes at least one of the following: the initial amount of funds allowed for investment within the time period, the amount of funds remaining at the end of the last transaction within the time period, the total equity of each trading day within the time period, the daily rate of return, the number of first investment products held, the identification of each first investment product held (for example, code), and the amount of funds held. The total equity of a trading day is: the sum of the amount of funds remaining on the trading day and the total amount of funds held for each first investment product held.
[0121] The transaction information that meets the target evaluation index is the transaction information used to calculate the index value of the target evaluation index. When acquiring the transaction information, the transaction server can determine the transaction information used to calculate the index value of the target evaluation index based on the candidate investment strategy in the first historical time period (i.e., the first transaction information), and determine the transaction information used to calculate the index value of the target evaluation index based on the candidate investment strategy in the second historical time period (i.e., the second transaction information).
[0122] For example, the target evaluation index is: cumulative rate of return, which is determined based on the total equity at the beginning of a period and the total equity at the end of a period. The total equity at the beginning of a period is the initial amount of funds allowed for investment at the beginning of the period, and the total equity at the end of a period is the sum of the remaining funds at the end of the period and the amount of funds held in the investment products. Correspondingly, the first transaction information includes the total equity at the beginning of the period and the total equity at the end of the period in the first historical period, and the transaction server can calculate the total equity at the beginning of the period and the total equity at the end of the period when investing in the investment market according to the candidate investment strategy in the first historical period.
[0123] Based on the above processing, the target evaluation value of the candidate investment strategy is determined by combining the transaction information within the first historical time period and the transaction information within the second historical time period. This can improve the diversity of the data and make the calculated target investment evaluation value more in line with the actual situation, that is, improve the accuracy of the determined target investment evaluation value.
[0124] In one embodiment of the present invention, see Figure 5 , Figure 5 This is a specific flow chart of determining the investment evaluation value in an embodiment of the present invention. The method may include the following steps:
[0125] S501: The transaction server calculates, for each target evaluation indicator, an indicator value of the target evaluation indicator based on first transaction information that meets the target evaluation indicator as a first indicator value.
[0126] S502: The transaction server calculates a weighted sum of the first indicator values of each target evaluation indicator as a first investment evaluation value.
[0127] S503: The transaction server calculates the indicator value of the target evaluation indicator based on the second transaction information that meets the target evaluation indicator as the second indicator value.
[0128] S504: The transaction server calculates a weighted sum of the second indicator values of each target evaluation indicator as a second investment evaluation value.
[0129] S505: The transaction server calculates the weighted sum of the first investment evaluation value and the second investment evaluation value to obtain a target investment evaluation value of the candidate investment strategy.
[0130] In one embodiment of the present invention, the target evaluation index includes at least one of the following: cumulative rate of return, annualized rate of return, cumulative Sharpe ratio, annualized Sharpe ratio, maximum drawdown rate of total equity, number of days required to recover from the maximum drawdown of total equity, annualized volatility, cumulative excess return, annualized excess rate of return, maximum drawdown rate of excess return, number of days required to recover from the maximum drawdown of excess return, information ratio, daily unilateral turnover rate, annualized unilateral turnover rate, daily bilateral turnover rate, and annualized bilateral turnover rate.
[0131] For each candidate investment strategy, when the target evaluation indicator is the cumulative rate of return, the transaction server can calculate the cumulative rate of return of investments made according to the candidate investment strategy in the first historical time period based on the first transaction information corresponding to the cumulative rate of return according to the following formula as the first evaluation indicator value.
[0132]
[0133] R represents the cumulative rate of return of investment in accordance with the candidate investment strategy during the first historical time period; A represents the total equity at the end of the first historical time period, which is the sum of the remaining funds at the end of the first historical time period and the holdings of the investment products held. The total equity at the end of the first historical time period is the remaining funds at the end of the first historical time period: the remaining funds at the end of the last transaction in the first historical time period; B represents the total equity at the beginning of the first historical time period, which is the initial funds allowed for investment at the beginning of the first historical time period.
[0134] When the target evaluation indicator is an annualized rate of return, the transaction server may calculate the annualized rate of return of investment according to the candidate investment strategy in the first historical time period according to the following formula based on the first transaction information corresponding to the annualized rate of return, as the first evaluation indicator value.
[0135]
[0136] P represents the annualized rate of return of investing in accordance with the candidate investment strategy during the first historical time period; R represents the cumulative rate of return of investing in accordance with the candidate investment strategy during the first historical time period; t represents the ratio of the number of trading days in a year to the number of trading days in the first historical time period.
[0137] When the target evaluation indicator is the cumulative Sharpe ratio, the trading server may calculate the Sharpe ratio of the investment based on the candidate investment strategy in the first historical time period according to the following formula based on the first transaction information corresponding to the cumulative Sharpe ratio as the first evaluation indicator value. The cumulative Sharpe ratio can measure the performance of the investment strategy relative to the risk-free portfolio, and is a measure of the risk premium obtained by the investment strategy. The risk-free portfolio may be a ten-year treasury bond.
[0138]
[0139] S represents the Sharpe ratio of investment based on the candidate investment strategy during the first historical time period; R represents the cumulative rate of return of investment based on the candidate investment strategy during the first historical time period; F represents the cumulative rate of return of the risk-free portfolio during the first historical time period; μ represents the variance of the daily rate of return of each trading day of investment based on the candidate investment strategy during the first historical time period.
[0140] When the target evaluation indicator is the maximum drawdown rate of total equity, the transaction server may calculate the maximum drawdown rate of total equity invested based on the candidate investment strategy in the first historical time period according to the following formula based on the first transaction information corresponding to the maximum drawdown rate of total equity as the first evaluation indicator value. The maximum drawdown rate of total equity is an important indicator for evaluating the extreme risk management capability of an investment strategy.
[0141]
[0142] D represents the maximum drawdown rate of the total equity invested based on the candidate investment strategy during the first historical time period; C1 represents the maximum value of the total equity of each trading day invested based on the candidate investment strategy during the first historical time period; C2 represents the minimum value of the total equity of each trading day invested based on the candidate investment strategy during the first historical time period.
[0143] When the target evaluation indicator is the annualized volatility, the transaction server may calculate the annualized volatility of the investment based on the candidate investment strategy in the first historical time period according to the following formula based on the first transaction information corresponding to the annualized volatility as the first evaluation indicator value. The annualized volatility is a commonly used risk measurement indicator. The larger the annualized volatility, the higher the risk of the investment strategy.
[0144]
[0145] σ represents the annualized volatility of investment based on the candidate investment strategy in the first historical period, n represents the number of trading days in the first historical period, r(i) represents the daily rate of return on the i-th trading day of investment based on the candidate investment strategy in the first historical period, It represents the average daily rate of return of each trading day of investment based on the candidate investment strategy in the first historical time period.
[0146] When the target evaluation indicator is the cumulative excess rate of return, the transaction server can calculate the cumulative excess rate of return of investments based on the candidate investment strategy in the first historical time period according to the following formula based on the first transaction information corresponding to the cumulative excess rate of return as the first evaluation indicator value. The calculation formula for the cumulative excess rate of return is the residual term in the expression of CAPM (Capital Asset Pricing Model). The cumulative excess rate of return can represent the part of the return corresponding to the investment strategy that is not related to the overall market return, and can be used to measure the ability of the investment strategy to select investment products. If the cumulative excess rate of return is a positive value, it indicates that the return of the investment product selected by the investment strategy is higher than the market benchmark portfolio; if the cumulative excess rate of return is a negative value, it indicates that the return of the investment product selected by the investment strategy is lower than the market benchmark portfolio, and the market benchmark portfolio is the Shanghai and Shenzhen 300 Index.
[0147] α=E[r(i)-[f+β(r(i)-f)]] (6)
[0148] α represents the cumulative excess return of investing based on the candidate investment strategy during the first historical time period; E represents the mathematical expectation; r(i) represents the daily return of the i-th trading day of investing based on the candidate investment strategy during the first historical time period; f represents the daily return of the risk-free portfolio during the first historical time period; β represents the preset coefficient in the CAPM model.
[0149] When the target evaluation indicator is the information ratio, the trading server may calculate the information ratio of the investment based on the candidate investment strategy in the first historical time period according to the following formula based on the first transaction information corresponding to the information ratio as the first evaluation indicator value. The information ratio is used to measure the return of the investment strategy relative to the market benchmark portfolio. The information ratio is generally used to evaluate pure long active trading investment strategies (for example, alpha investment strategies and benchmark timing investment strategies), and the information ratio is not suitable for evaluating long-short combined hedging investment strategies.
[0150]
[0151] I represents the information ratio of investment based on the candidate investment strategy in the first historical time period; R represents the cumulative rate of return of investment according to the candidate investment strategy in the first historical time period; F represents the cumulative rate of return of the risk-free portfolio in the first historical time period; ε represents the variance of the daily rate of return of the market benchmark portfolio in the first historical time period.
[0152] When the target evaluation indicator is the daily unilateral turnover rate, for each trading day in the first historical time period, the trading server can calculate the ratio of the amount of funds purchased in the first investment product on the trading day to the total equity on the trading day to obtain the daily unilateral turnover rate of the trading day, and calculate the average value of the daily unilateral turnover rates of each trading day in the first historical time period as the first indicator value.
[0153] In addition, the annualized Sharpe ratio can be calculated based on the cumulative Sharpe ratio. The number of days required to recover the maximum drawdown of total equity represents the number of days between the trading day with the minimum value of total equity and the trading day with the maximum value of total equity when investing according to the candidate investment strategy in the first historical time period. The annualized excess return rate can be calculated based on the cumulative excess return rate. The maximum drawdown rate of the excess return rate and the number of days required to recover the maximum drawdown of the excess return rate can both be calculated based on the excess return rate of each trading day in the first historical time period. The annualized unilateral turnover rate can be calculated based on the daily unilateral turnover rate. For each trading day in the first historical time period, the daily bilateral turnover rate of the trading day is the ratio of the total amount of funds bought and sold for the first investment product on the trading day to the total equity on the trading day. The trading server can calculate the average of the daily bilateral turnover rates of each trading day in the first historical time period as the first indicator value. The annualized bilateral turnover rate can be calculated based on the daily bilateral turnover rate.
[0154] Furthermore, the transaction server may calculate the weighted sum of the first indicator values of each target evaluation indicator to obtain the first investment evaluation value. Exemplarily, the transaction server may record the correspondence between the target evaluation indicator and the weight in the form of series values (data corresponding to the value and index). See Table 1, which is a correspondence table between a target evaluation indicator and a weight provided by an embodiment of the present invention. The sum of the weights of each target evaluation indicator is 1.
[0155] Table 1
[0156] Evaluation indicators Weight zb-a zb-a1 zb-b zb-b1 zb-c zb-c1 zb-d zb-d1 zb-e zb-e1 zb-f zb-f1 zb-g zb-g1
[0157] zb-a to zb-g represent different target evaluation indicators, and zb-a1 to zb-g1 represent weights of corresponding target evaluation indicators. The transaction server may calculate the indicator values of the evaluation indicators zb-a to zb-g based on the first transaction information corresponding to the first historical time period, and then the transaction server may calculate the weighted sum of the indicator values of the evaluation indicators zb-a to zb-g according to the weights zb-a1 to zb-g1 to obtain the first investment evaluation value.
[0158] The transaction server calculates the second investment evaluation value in a similar manner to the first investment evaluation value, and reference may be made to the relevant description of the aforementioned embodiment.
[0159] Based on the above processing, multiple different evaluation indicators can be combined to obtain the investment evaluation value of the candidate investment strategy. The investment evaluation value can reflect the characteristics of the investment strategy from multiple dimensions.
[0160] For each candidate investment strategy, the transaction server may calculate the weighted sum of the first investment evaluation value and the second investment evaluation value of the candidate investment strategy to obtain the target investment evaluation value of the candidate investment strategy. The weight of the first investment evaluation value is a, the weight of the second investment evaluation value is b, and the sum of a and b is 1. For example, the weight of the first investment evaluation value is 0.1, and the weight of the second investment evaluation value is 0.9. The transaction server may also normalize the target investment evaluation value to a value between 0 and 100 to obtain the final investment evaluation value of the candidate investment strategy.
[0161] Exemplarily, the target evaluation indicators include: cumulative rate of return, cumulative Sharpe ratio and maximum drawdown rate of total equity. The weight of the cumulative rate of return is 0.8, the weight of the cumulative Sharpe ratio is 0.1, and the weight of the maximum drawdown rate of total equity is 0.1. The first investment evaluation value is: the weighted sum of the cumulative rate of return, the cumulative Sharpe ratio and the maximum drawdown rate of total equity in the first historical time period. The second investment evaluation value is: the weighted sum of the cumulative rate of return, the cumulative Sharpe ratio and the maximum drawdown rate of total equity in the second historical time period. The target evaluation indicator is: 0.1×first investment evaluation value+0.9×second investment evaluation value. The target investment evaluation value is normalized to the range of 0 to 100 to obtain the final investment evaluation value of the candidate investment strategy.
[0162] With respect to step S213 and step S214, the recommendation server may determine a preset number of investment strategies from each candidate investment strategy in descending order of the target investment evaluation value as the target investment strategy, for example, from each candidate investment strategy, determine the investment strategy with the largest corresponding target investment evaluation value as the target investment strategy. Furthermore, the recommendation server may send recommendation information including the target investment strategy and the corresponding target investment evaluation value to the client.
[0163] For step S215, the client can receive the recommendation information sent by the recommendation server and display the recommendation information to the target user, that is, display the target investment strategy and the corresponding target investment evaluation value for the user to browse. Subsequently, the user can invest according to the target investment strategy.
[0164] In addition, the recommendation server can also sort the candidate investment strategies in descending order of the target investment evaluation value, use the sorting result as recommendation information, and send the recommendation information to the client, that is, send the sorting result to the client. The client can display the received recommendation information, that is, display the candidate investment strategies and the corresponding target investment evaluation values in descending order of the target investment evaluation value. The user can determine the target investment strategy based on the displayed target investment evaluation values of each candidate investment strategy.
[0165] In one embodiment, for each investment strategy to be recommended, the recommendation server may obtain the target investment evaluation value of the investment strategy to be recommended and the similarity between the investment strategy to be recommended and the target user. Then, the recommendation server may calculate the weighted sum of the target investment evaluation value of the investment strategy to be recommended and the corresponding similarity as the final investment evaluation value of the investment strategy to be recommended. Then, the recommendation server may determine the first preset number of investment strategies from each investment strategy to be recommended in descending order of the final investment evaluation values as the target investment strategy.
[0166] In one embodiment, for each investment strategy to be recommended, the recommendation server may obtain the target investment evaluation value of the investment strategy to be recommended, and determine the investment strategy whose investment evaluation value meets the preset recommendation condition from the investment strategies to be recommended as the candidate investment strategy. Then, for each candidate investment strategy, obtain the similarity between the candidate investment strategy and the target user. Then, the recommendation server may determine the investment strategy whose corresponding similarity meets the preset selection condition from the investment strategies to be recommended as the target investment strategy.
[0167] In one embodiment, the transaction server may also obtain a target evaluation index, and then calculate a target investment evaluation value of the candidate investment strategy according to the target evaluation index, the first transaction information, and the second transaction information.
[0168] In one implementation, the configuration information also includes: multiple target evaluation indicators determined by the target user through the configuration page; the investment evaluation value acquisition request includes multiple target evaluation indicators, so that the transaction server determines the target investment evaluation value of the candidate investment strategy based on the target evaluation indicators, the first transaction information and the second transaction information.
[0169] The target user can input an indicator configuration instruction for obtaining the evaluation indicator to the client. When the client receives the indicator configuration instruction, it can display a configuration page containing multiple preset evaluation indicators. The user can then select the target evaluation indicator for evaluating the investment strategy from multiple evaluation indicators to input the indicator selection instruction to the client. For example, the configuration page displays multiple evaluation indicators such as cumulative rate of return, annualized rate of return, cumulative Sharpe ratio, information ratio and annualized volatility. The target user can select the target evaluation indicator as: annualized rate of return, information ratio and annualized volatility. Then, the client can send an information recommendation request containing the target evaluation indicator to the recommendation server. Subsequently, the recommendation server can send an investment evaluation value acquisition request containing the target evaluation indicator to the transaction server, and the transaction server can determine the target investment evaluation value of each candidate investment strategy based on the target evaluation indicator, the first transaction information and the second transaction information.
[0170] Based on the above processing, the client can flexibly configure the evaluation indicators according to the user's instructions to obtain the target evaluation indicators that meet the user's needs, which can meet the user's personalized needs for configuring the evaluation indicators and improve the user experience.
[0171] In another implementation, the investment evaluation value acquisition request includes a target user identifier, so that the transaction server determines an evaluation indicator that matches the user characteristics of the target user from the preset evaluation indicators as the target evaluation indicator; based on the target evaluation indicator, the first transaction information and the second transaction information, the target investment evaluation value of the candidate investment strategy is determined.
[0172] Exemplarily, the user characteristics of the target user include: when the risk type is conservative, the trading server can determine an evaluation indicator that can represent the risk level of the investment strategy (for example, annualized volatility, maximum drawdown rate of total equity, Sharpe ratio, etc.) as the target evaluation indicator. Alternatively, the user characteristics of the target user include: when the return is higher than 1%, the trading server can determine an evaluation indicator that can represent the return corresponding to the investment strategy (for example, cumulative rate of return, annualized rate of return, excess rate of return, etc.) as the target evaluation indicator.
[0173] Based on the above processing, the target evaluation index is an evaluation index that matches the user characteristics of the target user. The target evaluation index meets the needs of the target user, and the evaluation index can be personalized for the target user to improve the user experience.
[0174] See also Figure 6 , Figure 6Schematic diagram of the configuration page of the client in an embodiment of the present invention. The user can configure the investment strategy and evaluation indicators in the configuration page. The strategy configuration items for configuring the investment strategy include: stock selection settings (for setting the screening conditions of stocks), trading model (i.e., the trading method for investment) and risk control (i.e., the risk type of stocks).
[0175] Selecting the stock pool as the whole market means that users can select all stocks in the investment market. Selecting the sector as all means that users can select stocks from any sector (for example, military industry and semiconductors, etc.). Selecting the industry as all means that users can select stocks from any industry (for example, medical and new energy, etc.). Filtering ST means filtering ST type stocks, which refers to stocks whose companies have abnormal financial conditions or other abnormal conditions. Users can enter the required stock selection index or evaluation index in the search factor input box to quickly select the required stock selection index or evaluation index.
[0176] The stock selection indicators include: market conditions (i.e., the market conditions indicator factors in the aforementioned embodiment), technical indicators (i.e., the quantity and price indicator factors in the aforementioned embodiment), financial indicators and financial data (i.e., the financial indicator factors in the aforementioned embodiment), and morphological indicators (i.e., the morphological indicator factors in the aforementioned embodiment).
[0177] The market conditions include: the stock's opening price, closing price, highest price, lowest price, trading volume, turnover, turnover rate and days on the market, etc. For example, users can choose the opening price as a stock selection indicator. Technical indicators include: the stock's mean, MACD (Moving Average Convergence / Divergence), Aroon indicator, KDJ (Stochastic Index), ROC (Price Rate of Change), MFI (Money Flow Index), Bollinger Bands, TRIX (Triple Exponentially Smoothed Average), etc. Financial indicators include: the valuation, profitability, per share indicators and debt repayment capacity of the company to which the stock belongs.
[0178] My Indicators indicates the number of stock picking indicators that the user has selected, for example, Figure 6 In the example, the user selected 4 stock selection indicators: opening price, price-to-book ratio, operating profit and two crows. The opening price range is greater than 1.01 yuan, the price-to-book ratio is greater than 2 times, and the operating profit is greater than 10 million. The delete button "×" is used to delete the selected stock selection indicator. "Add custom rules" is used to add the stock selection conditions set by the user.
[0179] The trading models include: regular rotation mode and conditional trigger mode. Regular rotation mode means that users set the cycle of adjusting the holding funds of investment products and the maximum holding stock data, etc. The conditional trigger mode means that users set the buying and selling conditions of stocks. Figure 6 The trading model that the user has selected is the periodic rotation method.
[0180] After setting the parameter values and evaluation indicators of each strategy configuration item, the user can click the "Save" button to input the configuration information including the parameter values and target evaluation indicators of each strategy configuration item to the client. Based on the above processing, the user can configure the investment strategy and evaluation indicators on the configuration page, which can meet the user's personalized configuration needs and improve the user experience.
[0181] See also Figure 7 , Figure 7 A schematic diagram of the principle of an information recommendation method provided by an embodiment of the present invention.
[0182] Market conditions and factor sources include: Shanghai and Shenzhen Stock Exchanges, China Financial Futures Exchange and Dalian Commodity Exchange, etc. The market conditions factor access device is used to obtain product information such as market conditions index factors, financial index factors, morphological index factors, volume and price index factors, public opinion index factors and characteristic data index factors of each investment product in the investment market within a specified time period (such as the first historical time period and the second historical time period) from the market conditions and factor sources. The market conditions factor access device can be implemented by the transaction server.
[0183] The quantitative strategy making device is used to receive the investment strategy configured by the user. The quantitative strategy making device provides the user with a uniform amount of virtual funds (for example, 1 million virtual funds) as the initial amount of funds allowed for investment. The quantitative strategy making device includes a stock selection module, a trading model and a risk control model. The user sets the stock selection conditions (i.e., the screening conditions for stocks) through the stock selection module of the quantitative strategy making device. The stock selection conditions configured by the stock selection module include: market conditions (i.e., the market index factors in the aforementioned embodiment), technical indicators (i.e., the quantity and price index factors in the aforementioned embodiment), financial indicators and financial data (i.e., the financial index factors in the aforementioned embodiment), morphological indicators (i.e., the morphological index factors in the aforementioned embodiment), etc.
[0184] The user sets the trading mode of investment through the trading model of the quantitative strategy making device. The trading modes include: regular rotation mode and conditional trigger mode. The regular rotation mode means that the user sets the cycle of adjusting the holding funds of the investment product and the maximum holding stock data. The conditional trigger mode means that the user sets the buying conditions and selling conditions of the stock. The quantitative strategy making device can be implemented by the recommendation server.
[0185] Furthermore, when the product information of the stock meets the buying condition, the quantitative strategy operation device will buy the stock according to the investment strategy to conduct a stock buying transaction. When the product information of the stock meets the selling condition, the quantitative strategy operation device will sell the held stock according to the investment strategy to conduct a stock selling transaction. The quantitative strategy operation device uniformly sets the start and end time of the investment strategy (that is, determines the start and end time of the historical time period), and runs different investment strategies in the form of programs according to the product information obtained by the market factor access device. When the product information of the investment product meets the buying condition, the quantitative strategy operation device will buy the investment product according to the investment strategy. When the product information of the investment product meets the selling condition, the quantitative strategy operation device will sell the held investment product according to the investment strategy. The quantitative strategy operation device performs a position change operation based on the product information such as the investment strategy and the market index factor to obtain the position market value of the investment product. The position market value is the product of the position price of the investment product and the position quantity. In addition, the quantitative strategy operation device can also record the position status (i.e., transaction information) of the investment strategy operation. The position status includes: the remaining funds, total equity, the code and number of investment products held, the real-time market net value, and the average transaction price of the investment products purchased, etc.
[0186] In addition, the quantitative strategy operation device is also configured with a transaction fee deduction standard. When performing a position change operation, the corresponding transaction fee is deducted when making an investment according to the transaction fee deduction standard. For example, the minimum transaction fee is 5 yuan, and the two-way transaction fee is 0.02% of the transaction amount. The quantitative strategy operation device can be implemented by the transaction server.
[0187] The quantitative strategy evaluation device is used to calculate the target investment evaluation value of the candidate investment strategy based on the transaction information of the investment according to the candidate investment strategy, and determine the recommendation information including the target investment strategy recommended to the target user. The quantitative strategy evaluation device can be implemented by the transaction server.
[0188] The quantitative strategy evaluation output device is used to send recommendation information to the client, so that the client displays the recommendation information to the target user. The quantitative strategy evaluation output device can be implemented by the recommendation server.
[0189] Based on the information recommendation method provided by the embodiment of the present invention, an investment strategy evaluation platform is provided for users, and the investment strategy can be evaluated in combination with the first historical time period and the second historical time period to obtain a target investment evaluation value that conforms to the actual situation. The investment strategy is run based on the real-time market index factors corresponding to the historical time period, and the funds for running the investment strategy are virtual funds, which can facilitate simulation before the investment strategy is actually run. Users can flexibly configure investment strategies and evaluation indicators, and the configuration method is simple and easy to operate, which can meet the personalized needs of users and improve user experience.
[0190] and Figure 1 Corresponding to the method embodiment of Figure 8 , Figure 8 A structural diagram of a recommendation server provided in an embodiment of the present invention, wherein the recommendation server includes:
[0191] An information recommendation request receiving module 801 is used to receive an information recommendation request sent by the client; wherein the information recommendation request includes a target user identifier and configuration information of the information to be recommended; a module for obtaining investment strategies to be recommended 802 is used to obtain multiple investment strategies to be recommended based on the configuration information when the information to be recommended is determined to be an investment strategy based on the configuration information; a user feature obtaining module 803 is used to send a corresponding user feature obtaining request to the transaction server based on the target user identifier; so that the transaction server obtains the user features of the target user based on the transaction data of the target user's historical investments, and returns the same to the recommendation server; a product feature obtaining module 804 is used to send a corresponding product feature obtaining request to the transaction server for each investment strategy to be recommended; so that the transaction server obtains the product features of the investment product corresponding to the investment strategy to be recommended, and returns the same to the recommendation server; a similarity calculation module 805 is used to respectively calculate the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user; a candidate investment strategy obtaining module 806 is used to obtain the product features of the investment product corresponding to the target user. The investment strategies to be recommended whose user similarity meets the preset selection conditions are used as candidate investment strategies; an investment evaluation value acquisition module 807 is used to send a corresponding investment evaluation value acquisition request to the transaction server for each candidate investment strategy; so that the transaction server obtains historical transaction scenario information within a first historical time period in response to the investment evaluation value acquisition request, and calculates the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtains the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and based on the first transaction information and the second transaction information, calculates the target investment evaluation value of the candidate investment strategy and returns it to the recommendation server; a target investment strategy acquisition module 808 is used to determine, from each candidate investment strategy, a target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions; a recommendation information output module 809 is used to send the target investment strategy as recommendation information to the client, so that the client displays the recommendation information.
[0192] Optionally, the configuration information includes: configuration parameter values of various strategy configuration items used to determine the investment strategy; the configuration parameter values are input by the target user through a configuration page; and a plurality of strategy configuration items are displayed on the configuration page;
[0193] The investment strategy acquisition module 802 is specifically used to generate multiple investment strategies to be recommended based on the configuration parameter values of each strategy configuration item input by the target user; wherein each investment strategy to be recommended includes a screening condition and a transaction method.
[0194] Optionally, the configuration information further includes: a plurality of target evaluation indicators determined by the target user through the configuration page; the investment evaluation value acquisition request includes the plurality of target evaluation indicators, so that the transaction server determines the target investment evaluation value of the candidate investment strategy based on the target evaluation indicators, the first transaction information and the second transaction information;
[0195] or,
[0196] The investment evaluation value acquisition request includes the target user identifier, so that the transaction server determines an evaluation indicator that matches the user characteristics of the target user from the preset evaluation indicators as the target evaluation indicator; based on the target evaluation indicator, the first transaction information and the second transaction information, the target investment evaluation value of the candidate investment strategy is determined.
[0197] Optionally, the configuration information includes: a strategy identifier of an investment strategy;
[0198] The investment strategy acquisition module 802 is specifically used to determine the investment strategy to which the strategy identifier belongs from the locally recorded investment strategies, and obtain multiple investment strategies to be recommended; wherein each investment strategy to be recommended includes a screening condition and a transaction method.
[0199] Optionally, the similarity calculation module 805 is specifically used to, for each investment strategy to be recommended, perform mapping processing on the product features corresponding to the investment strategy to be recommended, to obtain a first feature vector corresponding to the investment strategy to be recommended; perform mapping processing on the user features of the target user, to obtain a second feature vector corresponding to the target user; calculate the similarity between the first feature vector and the second feature vector, to obtain the similarity between the product features corresponding to the investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user;
[0200] The candidate investment strategy acquisition module 806 is specifically used to determine a preset number of investment strategies to be recommended from the investment strategies to be recommended as candidate investment strategies in descending order of similarity with the target user;
[0201] or,
[0202] From among the investment strategies to be recommended, an investment strategy to be recommended whose similarity with the target user is greater than a preset threshold is determined as a candidate investment strategy.
[0203] Based on the recommendation server provided in the embodiment of the present invention, based on the user characteristics of the target user, the candidate investment strategies whose similarity with the user characteristics of the user meets the preset selection conditions are determined, and the target investment strategy recommended to the user is determined from the candidate investment strategies. The target investment strategy is an investment strategy that meets the real needs of the target user, and personalized recommendation services can be provided to the target user. In addition, the candidate investment strategy is determined from the investment strategies to be recommended, which can reduce the amount of data to be processed subsequently and improve the efficiency of information recommendation.
[0204] and Figure 1 Corresponding to the method embodiment of Fig. 9 , Fig. 9 A structural diagram of a transaction server provided in an embodiment of the present invention, wherein the transaction server comprises:
[0205] A user feature acquisition module 901 is used to obtain the user features of the target user based on the historical investment data of the target user and return the user features to the recommendation server when receiving a user feature acquisition request sent by the recommendation server based on the target user identifier included in the information recommendation request; wherein the information recommendation request is sent by the client to the recommendation server, and the information recommendation request includes the target user identifier and the configuration information of the information to be recommended; a product feature acquisition module 902 is used to obtain the product features of the investment product corresponding to each investment strategy to be recommended and return the product features to the recommendation server when receiving a product feature acquisition request sent by the recommendation server; so that the recommendation server calculates the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user respectively, as the similarity between the investment strategy to be recommended and the target user; obtain the investment strategy to be recommended whose similarity with the target user meets the preset selection condition as a candidate investment strategy; wherein the investment strategy to be recommended is: the recommendation server determines that the information to be recommended is an investment strategy based on the configuration information In the case of omission, it is obtained based on the configuration information; a transaction information acquisition module 903 is used for, for each candidate investment strategy, when receiving an investment evaluation value acquisition request sent by the recommendation server, for the investment evaluation value acquisition request, obtaining historical transaction scenario information within a first historical time period, and calculating the first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein, the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtaining the second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy; wherein, the second historical time period is a time period for simulated transactions based on the candidate investment strategy; an investment evaluation value acquisition module 904 is used to calculate the target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information, and return it to the recommendation server; so that the recommendation server determines the target investment strategy whose corresponding target investment evaluation value meets the preset recommendation conditions from each candidate investment strategy; and sends the target investment strategy as recommendation information to the client, so that the client displays the recommendation information.
[0206] Optionally, the transaction information acquisition module 903 is specifically used to acquire product information of each investment product in the historical transaction scenario within the first historical time period as the first product information; determine each investment product that matches the screening condition in the candidate investment strategy from the investment products in the historical transaction scenario within the first historical time period as the first investment product; based on the transaction method in the candidate investment strategy and the first product information of the first investment product, calculate the transaction information that meets the target evaluation indicator when investing according to the candidate investment strategy within the first historical time period as the first transaction information;
[0207] The transaction information acquisition module 903 is specifically used to obtain the product information of each investment product in the historical transaction scenarios within the second historical time period as the second product information; from the various investment products in the historical transaction scenarios within the second historical time period, determine the investment products that match the screening conditions in the candidate investment strategy as the second investment products; within the second historical time period, simulate transactions based on the transaction method in the candidate investment strategy and the second product information of the second investment product to obtain transaction information that meets the target evaluation indicators as the second transaction information.
[0208] Optionally, the investment evaluation value acquisition request carries multiple target evaluation indicators;
[0209] The investment evaluation value acquisition module 904 is specifically used to calculate, for each target evaluation indicator, the indicator value of the target evaluation indicator based on the first transaction information that meets the target evaluation indicator, as the first indicator value; calculate the weighted sum of the first indicator values of each target evaluation indicator as the first investment evaluation value; calculate the indicator value of the target evaluation indicator based on the second transaction information that meets the target evaluation indicator as the second indicator value; calculate the weighted sum of the second indicator values of each target evaluation indicator as the second investment evaluation value; calculate the weighted sum of the first investment evaluation value and the second investment evaluation value to obtain the target investment evaluation value of the candidate investment strategy.
[0210] Based on the transaction server provided by the embodiment of the present invention, based on the user characteristics of the target user, the candidate investment strategies whose similarity with the user characteristics of the user meets the preset selection conditions are determined, and the target investment strategy recommended to the user is determined from the candidate investment strategies. The target investment strategy is an investment strategy that meets the real needs of the target user, and personalized recommendation services can be provided to the target user. In addition, the candidate investment strategy is determined from the investment strategies to be recommended, which can reduce the amount of data to be processed subsequently and improve the efficiency of information recommendation.
[0211] In another embodiment provided by the present invention, a computer-readable storage medium is further provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of the information recommendation method applied to the recommendation server described in any one of the embodiments, or the steps of the information recommendation method applied to the transaction server described in any one of the above embodiments are implemented.
[0212] In another embodiment provided by the present invention, a computer program product comprising instructions is also provided. When the computer is run on a computer, the computer executes the information recommendation method applied to the recommendation server described in any of the above embodiments, or the steps of the information recommendation method applied to the transaction server described in any of the above embodiments.
[0213] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state hard disk Solid State Disk (SSD)), etc.
[0214] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0215] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system, recommendation server, transaction server, computer-readable storage medium, and computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.
[0216] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.
Claims
1. An information recommendation method, characterized in that: The method is applied to a recommendation server in a recommendation system, the recommendation system further comprising a client; the recommendation server is communicatively connected with a transaction server, and the method comprises: receiving an information recommendation request sent by the client; wherein the information recommendation request includes a target user identifier and configuration information of the information to be recommended; wherein the configuration information includes: configuration parameter values of each strategy configuration item for determining the investment strategy; the configuration parameter values are input by the target user through a configuration page; and the configuration page displays a plurality of strategy configuration items; In the case where it is determined based on the configuration information that the information to be recommended is an investment strategy, obtaining a plurality of investment strategies to be recommended based on the configuration information; Sending a corresponding user feature acquisition request to the transaction server based on the target user identifier; so that the transaction server obtains the user features of the target user based on the historical investment data of the target user and returns them to the recommendation server; For each investment strategy to be recommended, a corresponding product feature acquisition request is sent to the transaction server; so that the transaction server obtains the product features of the investment product corresponding to the investment strategy to be recommended, and returns the product features to the recommendation server; Calculating the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user respectively as the similarity between the investment strategy to be recommended and the target user; Obtaining a to-be-recommended investment strategy whose similarity with the target user satisfies a preset selection condition as a candidate investment strategy; For each candidate investment strategy, a corresponding investment evaluation value acquisition request is sent to the transaction server; so that the transaction server obtains historical transaction scenario information within a first historical time period in response to the investment evaluation value acquisition request, and calculates first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtains second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and calculates a target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information, and returns the result to the recommendation server; From each candidate investment strategy, determine a target investment strategy whose corresponding target investment evaluation value meets a preset recommendation condition; sending the target investment strategy as recommendation information to the client, so that the client displays the recommendation information; The respectively calculating the similarity between the product feature corresponding to each investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user includes: For each investment strategy to be recommended, mapping processing is performed on the product features corresponding to the investment strategy to be recommended to obtain a first feature vector corresponding to the investment strategy to be recommended; Mapping the user features of the target user to obtain a second feature vector corresponding to the target user; The similarity between the first feature vector and the second feature vector is calculated to obtain the similarity between the product feature corresponding to the investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user.
2. The method according to claim 1, characterized in that In the case where it is determined based on the configuration information that the information to be recommended is an investment strategy, obtaining a plurality of investment strategies to be recommended based on the configuration information includes: Based on the configuration parameter values of each strategy configuration item input by the target user, a plurality of investment strategies to be recommended are generated; wherein each investment strategy to be recommended includes a screening condition and a transaction method.
3. The method according to claim 2, characterized in that The configuration information also includes: a plurality of target evaluation indicators determined by the target user through the configuration page; the investment evaluation value acquisition request includes the plurality of target evaluation indicators, so that the transaction server determines the target investment evaluation value of the candidate investment strategy based on the target evaluation indicators, the first transaction information and the second transaction information; or, The investment evaluation value acquisition request includes the target user identifier, so that the transaction server determines an evaluation indicator that matches the user characteristics of the target user from the preset evaluation indicators as the target evaluation indicator; based on the target evaluation indicator, the first transaction information and the second transaction information, the target investment evaluation value of the candidate investment strategy is determined.
4. The method according to claim 1, characterized in that: The configuration information includes: a strategy identifier of an investment strategy; In the case where it is determined based on the configuration information that the information to be recommended is an investment strategy, obtaining a plurality of investment strategies to be recommended based on the configuration information includes: From the locally recorded investment strategies, the investment strategy to which the strategy identifier belongs is determined, and a plurality of investment strategies to be recommended are obtained; wherein each investment strategy to be recommended includes a screening condition and a transaction method.
5. The method according to claim 1, characterized in that The step of obtaining the investment strategy to be recommended, whose similarity with the target user satisfies the preset selection condition, as the candidate investment strategy includes: Determining a preset number of investment strategies to be recommended from among the investment strategies to be recommended as candidate investment strategies in descending order of similarity with the target user; or, From among the investment strategies to be recommended, an investment strategy to be recommended whose similarity with the target user is greater than a preset threshold is determined as a candidate investment strategy.
6. An information recommendation method, characterized in that: The method is applied to a transaction server, the transaction server is communicatively connected with a recommendation server in a recommendation system; the recommendation system also includes a client; the method includes: Upon receiving a user feature acquisition request sent by the recommendation server based on the target user identifier included in the information recommendation request, the user features of the target user are obtained based on the historical investment data of the target user, and the information is returned to the recommendation server; wherein the information recommendation request is sent by the client to the recommendation server, and the information recommendation request includes the target user identifier and configuration information of the information to be recommended; the configuration information includes: configuration parameter values of each policy configuration item for determining the investment strategy; the configuration parameter values are input by the target user through a configuration page; and a plurality of policy configuration items are displayed on the configuration page; For each investment strategy to be recommended, upon receiving a product feature acquisition request sent by the recommendation server, the product features of the investment product corresponding to the investment strategy to be recommended are obtained and returned to the recommendation server; so that the recommendation server respectively calculates the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user as the similarity between the investment strategy to be recommended and the target user; obtains the investment strategy to be recommended whose similarity with the target user meets the preset selection condition as a candidate investment strategy; wherein the investment strategy to be recommended is: obtained based on the configuration information when the recommendation server determines that the information to be recommended is an investment strategy based on the configuration information; For each candidate investment strategy, upon receiving an investment evaluation value acquisition request sent by the recommendation server, obtain historical transaction scenario information within a first historical time period for the investment evaluation value acquisition request, and calculate first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtain second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; Based on the first transaction information and the second transaction information, a target investment evaluation value of the candidate investment strategy is calculated and returned to the recommendation server; so that the recommendation server determines a target investment strategy whose corresponding target investment evaluation value satisfies a preset recommendation condition from each candidate investment strategy; and sends the target investment strategy as recommendation information to the client, so that the client displays the recommendation information; The respectively calculating the similarity between the product feature corresponding to each investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user includes: For each investment strategy to be recommended, mapping processing is performed on the product features corresponding to the investment strategy to be recommended to obtain a first feature vector corresponding to the investment strategy to be recommended; Mapping the user features of the target user to obtain a second feature vector corresponding to the target user; The similarity between the first feature vector and the second feature vector is calculated to obtain the similarity between the product feature corresponding to the investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user.
7. The method according to claim 6, characterized in that The obtaining of historical transaction scenario information within a first historical time period, and calculating first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information, includes: Acquire product information of each investment product in the historical transaction scenario within the first historical time period as first product information; Determine, from among the investment products in the historical transaction scenarios within the first historical time period, investment products that match the screening conditions in the candidate investment strategy as first investment products; Based on the transaction method in the candidate investment strategy and the first product information of the first investment product, calculate the transaction information that meets the target evaluation index when investing according to the candidate investment strategy during the first historical time period as the first transaction information; The obtaining of second transaction information corresponding to the second historical time period corresponding to the candidate investment strategy includes: Acquire product information of each investment product in the historical transaction scenario within the second historical time period as second product information; Determine, from among the investment products in the historical transaction scenarios within the second historical time period, investment products that match the screening conditions in the candidate investment strategy as second investment products; During the second historical time period, simulated trading is performed based on the trading method in the candidate investment strategy and the second product information of the second investment product to obtain trading information that meets the target evaluation index as the second trading information.
8. The method according to claim 6, characterized in that The investment evaluation value acquisition request carries multiple target evaluation indicators; The step of calculating the target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information includes: For each target evaluation indicator, based on the first transaction information that meets the target evaluation indicator, calculate an indicator value of the target evaluation indicator as a first indicator value; Calculate the weighted sum of the first indicator values of each target evaluation indicator as the first investment evaluation value; Calculating an index value of the target evaluation index based on the second transaction information that meets the target evaluation index as a second index value; Calculate the weighted sum of the second indicator values of each target evaluation indicator as the second investment evaluation value; A weighted sum of the first investment evaluation value and the second investment evaluation value is calculated to obtain a target investment evaluation value of the candidate investment strategy.
9. An information recommendation system, characterized in that: The recommendation system comprises: a client and a recommendation server; the recommendation server is communicatively connected with the transaction server; The client is used to send an information recommendation request including a target user identifier and configuration information of the information to be recommended to the recommendation server; wherein the configuration information includes: configuration parameter values of various strategy configuration items for determining the investment strategy; the configuration parameter values are input by the target user through a configuration page; and the configuration page displays multiple strategy configuration items; The recommendation server is configured to receive the information recommendation request; if it is determined based on the configuration information that the information to be recommended is an investment strategy, obtain a plurality of investment strategies to be recommended based on the configuration information; and send a corresponding user feature acquisition request to the transaction server based on the target user identifier; The transaction server is used to obtain user characteristics of the target user based on the historical investment data of the target user, and return the user characteristics to the recommendation server; The recommendation server is further configured to send a corresponding product feature acquisition request to the transaction server for each investment strategy to be recommended; The transaction server is further configured to obtain product characteristics of the investment product corresponding to the investment strategy to be recommended, and return the characteristics to the recommendation server; The recommendation server is further configured to calculate the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user; obtain the investment strategy to be recommended whose similarity with the target user satisfies the preset selection condition as the candidate investment strategy; and send a corresponding investment evaluation value acquisition request to the transaction server for each candidate investment strategy; The transaction server is further configured to obtain historical transaction scenario information within a first historical time period in response to an investment evaluation value acquisition request, and calculate first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtain second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and calculate a target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information, and return the result to the recommendation server; The recommendation server is further configured to determine, from among the candidate investment strategies, a target investment strategy whose corresponding target investment evaluation value satisfies a preset recommendation condition; and send the target investment strategy as recommendation information to the client; The client is further used to display the recommendation information; The recommendation server is specifically used for: For each investment strategy to be recommended, mapping processing is performed on the product features corresponding to the investment strategy to be recommended to obtain a first feature vector corresponding to the investment strategy to be recommended; Mapping the user features of the target user to obtain a second feature vector corresponding to the target user; The similarity between the first feature vector and the second feature vector is calculated to obtain the similarity between the product feature corresponding to the investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user.
10. A recommendation server, characterized in that: The recommendation server comprises: An information recommendation request receiving module, used for receiving an information recommendation request sent by a client; wherein the information recommendation request includes a target user identifier and configuration information of the information to be recommended; the configuration information includes: configuration parameter values of various strategy configuration items for determining the investment strategy; the configuration parameter values are input by the target user through a configuration page; the configuration page displays multiple strategy configuration items; A module for obtaining investment strategies to be recommended, configured to obtain a plurality of investment strategies to be recommended based on the configuration information when it is determined that the information to be recommended is an investment strategy based on the configuration information; A user feature acquisition module, configured to send a corresponding user feature acquisition request to a transaction server based on the target user identifier, so that the transaction server obtains the user features of the target user based on the historical investment transaction data of the target user, and returns the user features to the recommendation server; A product feature acquisition module, for sending a corresponding product feature acquisition request to the transaction server for each investment strategy to be recommended, so that the transaction server obtains the product features of the investment product corresponding to the investment strategy to be recommended, and returns the product features to the recommendation server; A similarity calculation module, used to calculate the similarity between the product features corresponding to each investment strategy to be recommended and the user features of the target user, as the similarity between the investment strategy to be recommended and the target user; A candidate investment strategy acquisition module is used to obtain a to-be-recommended investment strategy whose similarity with the target user meets a preset selection condition as a candidate investment strategy; An investment evaluation value acquisition module is used to send a corresponding investment evaluation value acquisition request to the transaction server for each candidate investment strategy; so that the transaction server obtains historical transaction scenario information within a first historical time period in response to the investment evaluation value acquisition request, and calculates first transaction information corresponding to the candidate investment strategy based on the historical transaction scenario information; wherein the first historical time period is a time period earlier than the moment when the information recommendation request is received; and obtains second transaction information corresponding to a second historical time period corresponding to the candidate investment strategy; wherein the second historical time period is a time period for simulated transactions based on the candidate investment strategy; and calculates a target investment evaluation value of the candidate investment strategy based on the first transaction information and the second transaction information, and returns it to the recommendation server; A target investment strategy acquisition module is used to determine, from among the candidate investment strategies, a target investment strategy whose corresponding target investment evaluation value meets a preset recommendation condition; A recommendation information output module, used for sending the target investment strategy as recommendation information to the client, so that the client displays the recommendation information; The similarity calculation module is specifically used for: For each investment strategy to be recommended, mapping processing is performed on the product features corresponding to the investment strategy to be recommended to obtain a first feature vector corresponding to the investment strategy to be recommended; Mapping the user features of the target user to obtain a second feature vector corresponding to the target user; The similarity between the first feature vector and the second feature vector is calculated to obtain the similarity between the product feature corresponding to the investment strategy to be recommended and the user feature of the target user as the similarity between the investment strategy to be recommended and the target user.
