Intelligent routing method and system of intra-day rotation strategy algorithm

Abstract

The invention provides an intelligent routing method and system for an intra-day rotation strategy algorithm, and belongs to the technical field of computers. Initial stock combination information selected by a target client through a client is received; calling a strategy routing table to match each stock with a corresponding strategy algorithm, and generating first stock combination strategy information; filtering strategies which do not meet conditions according to the use permission of the target customer to form second stock combination strategy information; calling stock blacklist screening to ensure that the recommended stocks are not in the blacklist, thereby generating third stock combination strategy information; calculating the first feature vector similarity and the second feature vector similarity between each stock and the strategy thereof and the customer portrait, and determining the third feature vector similarity by integrating the two features; and based on the third feature vector similarity, generating a recommendation priority score for each stock and strategy distribution to determine a final stock combination and strategy. According to the invention, the strategy distribution efficiency and the user experience are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an intelligent routing method and system for an intraday reversal strategy algorithm. Background Art

[0002] The intraday reversal strategy algorithm refers to a trading calculation algorithm that uses algorithmic strategies to sell high and buy low based on the client's original stock holdings to earn the stock's intraday volatility price difference and reduce the holding cost.

[0003] Traditionally, brokerage firms usually rely on strategy algorithms provided by a single supplier to conduct stock trading, and some of them can support a customer to choose from multiple strategies. However, they cannot actively manage and control strategy algorithms, and can only use passive means, that is, let customers manually select a strategy, and then assign a single algorithm strategy to the customer, resulting in low trading efficiency and strategy flexibility, and it is difficult to achieve effective active management and dynamic adjustment of different stocks and their corresponding strategy algorithms, resulting in poor user experience. Summary of the invention

[0004] The present invention provides an intelligent routing method and system for an intraday reversal strategy algorithm, which is used to solve at least one defect in the above-mentioned prior art.

[0005] In a first aspect, the present invention provides an intelligent routing method for an intraday reversal strategy algorithm, comprising: receiving initial stock combination information input by a target customer through a client; the initial stock combination information is generated based on a stock selection operation of the target customer on the client, including a plurality of stock information selected by the target customer; calling a strategy routing table to match a strategy algorithm for each stock information in the initial stock combination information, and generating first stock combination strategy information; the strategy routing table includes a plurality of stock information and a strategy algorithm matched to each stock information, and the first stock combination strategy information includes successfully matched stock information and corresponding strategy algorithm information; determining the target customer's strategy algorithm usage authority based on a user identity identifier of the target customer, and filtering out stocks that do not meet the strategy algorithm usage requirements from the first stock combination strategy information. Generate second stock combination strategy information using authorized stock information and corresponding strategy algorithm information; call the stock blacklist to filter out stock information and corresponding strategy algorithm information that are not in the stock blacklist from the second stock combination strategy information to generate third stock combination strategy information; calculate the first eigenvector similarity between the stock parameter information of each stock in the third stock combination strategy information and the customer portrait information of the target customer, as well as the second eigenvector similarity between the strategy algorithm information of each stock and the customer portrait information of the target customer; determine the third eigenvector similarity based on the first eigenvector similarity and the second eigenvector similarity, and generate recommendation priority information for the stock information and the corresponding strategy algorithm information in the third stock combination strategy information based on the third eigenvector similarity.

[0006] In a second aspect, the present invention also provides an intelligent routing system for an intraday reversal strategy algorithm, comprising: a client, a management terminal and a policy server; the policy server is respectively communicated with the client and the management terminal, and is used to implement the intelligent routing method for the intraday reversal strategy algorithm as described above.

[0007] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the intelligent routing method of any one of the above-mentioned intraday reversal strategy algorithms are implemented.

[0008] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of an intelligent routing method for an intraday reversal strategy algorithm as described in any one of the above.

[0009] The intelligent routing method and system of the intraday reversal strategy algorithm provided by the present invention have the following beneficial effects compared with the prior art: (1) The present invention realizes the aggregation of strategy algorithms of multiple suppliers, and through the automated strategy matching process, reduces manual intervention, speeds up the process from stock selection to transaction execution, and improves transaction efficiency; it can also effectively evaluate the degree of match between stocks and strategy algorithms and customer preferences, thereby providing more personalized and accurate solution recommendations and enhancing the user experience.

[0010] (2) The intelligent routing method of the intraday reversal strategy algorithm provided by the present invention relies on a large amount of real and objective historical transaction data, and obtains recommendation priority information through data analysis, rather than relying solely on theory or hypothesis, which makes the recommendation more reliable and practical; and, when forming the recommendation priority information, the present invention takes into account both the information of the customer's own characteristics (the third eigenvector similarity) and the data support from similar customers (priority score), so the recommendation is more reasonable, which greatly enhances the user experience.

[0011] (3) The present invention supports the integration of multiple strategy algorithms and matches the optimal strategy algorithm for the customer's stock portfolio from multiple strategy algorithms, thus having strong strategy adaptability.

[0012] (4) The present invention introduces a blacklist mechanism to exclude specific stocks that are not suitable for strategy trading; sets customer algorithm permissions to ensure that each customer can only use strategies that match themselves, which is more targeted.

[0013] (5) The present invention can realize automatic matching of strategy algorithms without requiring customers to perform excessive operations, thereby improving customers' stock trading experience; securities dealers only need to regularly maintain the strategy routing table, customer algorithm permissions, and stock blacklists to achieve effective active management and dynamic adjustment of different stocks and their corresponding strategies, thereby improving work and management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a flow chart of the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention; Figure 2 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] It should be noted that, in the description of the embodiments of the present invention, 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 includes 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 "include one..." do not exclude the presence of other identical elements in the process, method, article or device including the elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0018] The execution subject of the method flow in the present invention may be a policy server (or a centralized algorithm management platform with computing and data storage functions). In the present invention, a client terminal used by a customer and a management terminal used by a brokerage firm may be provided to respectively realize the interaction between the customer and the brokerage firm and the policy server. Figure 1-Figure 2 The intelligent routing method and system of the intraday reversal strategy algorithm provided by the embodiment of the present invention are described.

[0019] Figure 1 It is a flow chart of the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps: Step 101: receiving initial stock portfolio information input by a target customer through a client; the initial stock portfolio information is generated based on the stock selection operation of the target customer on the client, and includes information of multiple stocks selected by the target customer.

[0020] It should be noted that the client here may be an application designed for customers to facilitate users to perform stock selection operations.

[0021] It is understandable that the client performs stock selection operations through the client provided by the brokerage firm to generate an initial stock portfolio information containing multiple stock information; this information may include stock codes, quantities, etc. Specifically, the initial stock portfolio information here may be generated based on the stock portfolio selected from the intraday revolving coupon pool.

[0022] Step 102: calling the policy routing table to match the policy algorithm for each stock information in the initial stock combination information, and generating the first stock combination policy information.

[0023] The strategy routing table includes multiple stock information and strategy algorithms matching each stock information, that is, the table defines the mapping relationship between different stocks and different trading strategies. For each stock, a matching strategy algorithm is found, and the information is packaged into the first stock combination strategy information, which includes the successfully matched stock information and the corresponding strategy algorithm information; Step 103: Determine the target customer's policy algorithm usage authority based on the target customer's user identity, filter out stock information and corresponding policy algorithm information that do not meet the policy algorithm usage authority from the first stock combination policy information, and generate second stock combination policy information.

[0024] The present invention can extract the user's unique identifier (such as user ID) from the session or request, which is usually generated by the client at the time of login, and determine the policy algorithm that the user is allowed to use based on the user's identity, for example, by searching the permission database for a list of policy algorithms that are accessible to the target customer.

[0025] Traverse each record in the first stock combination strategy information (i.e., each stock and the corresponding strategy algorithm), and for each strategy algorithm, check whether it is in the list of strategy algorithms allowed to be used by the target customer. If it is not in the list, remove this record from the collection to form a new stock combination strategy information that meets the user's permissions, i.e., the second stock combination strategy information.

[0026] Step 104: calling the stock blacklist to filter out the stock information and corresponding strategy algorithm information that are not in the stock blacklist from the second stock combination strategy information to generate the third stock combination strategy information.

[0027] Optionally, traverse each record in the second stock combination strategy information (i.e., each stock and the corresponding strategy algorithm). For each stock, check whether it exists in the blacklist. If it exists, remove this record from the second stock combination strategy information; otherwise, retain it to form a new stock combination strategy information that meets the blacklist screening criteria, i.e., the third stock combination strategy information.

[0028] Step 105: Calculate the first feature vector similarity between the stock parameter information of each stock in the third stock portfolio strategy information and the customer portrait information of the target customer, and the second feature vector similarity between the strategy algorithm information of each stock and the customer portrait information of the target customer.

[0029] Obtain stock parameter information of each stock (such as price-to-earnings ratio, volatility, trading volume, volatility, turnover rate and amplitude, etc.) from the database or API; customer profile information includes customer risk preference, investment style, historical trading behavior, etc.; strategy algorithm information includes: the risk level of the strategy algorithm, applicable market conditions and expected rate of return, etc.

[0030] The above stock parameter information, strategy algorithm information and customer profile information are mapped into feature vectors.

[0031] The first feature vector similarity is used to characterize the similarity between the feature vector of stock parameter information and the feature vector of customer profile information. Commonly used methods include cosine similarity, Euclidean distance, etc.

[0032] The second feature vector similarity is used to characterize the similarity between the strategy algorithm information feature vector and the customer profile information feature vector. The calculation method is the same as above.

[0033] It should be noted that the characteristic contents in the customer profile information, strategy algorithm information, and stock parameter information in the present invention can be manually selected and adjusted. The above setting method is only an optional embodiment and does not limit the protection scope of the present invention. For example, in another embodiment, the customer's occupation information can also be added to the customer profile information, which will not be repeated here.

[0034] The following is a brief description of an implementation process for vectorizing feature content: For information whose feature content has been determined and is to be vectorized (i.e., stock parameter information, customer profile information, or strategy algorithm information), the information can be vectorized using encoding processing to form a corresponding feature vector.

[0035] In the field of machine learning, there is an important link called "feature engineering", which is used to extract features from raw data and convert them into a format suitable for machine learning. The present invention can use the technology here to realize the vectorization of information. For example, the embedding processing method is adopted.

[0036] Based on the above content, assume that the eigenvector of the customer portrait information after quantization is [0.75, 0.36, 0.08, 0.2, 0.45], the eigenvector of the stock D parameter information is [0.60, 0.15, 0.20, 0.5, 0.7], and the eigenvector corresponding to the strategy X is [0.4, 0.1, 0.5, 0.6, 0.3]; by calculating the cosine similarity of [0.75, 0.36, 0.08, 0.2, 0.45] and [0.60, 0.15, 0.20, 0.5, 0.7], it can be obtained that the similarity of the first eigenvector is about 0.893; by calculating the cosine similarity of [0.75, 0.36, 0.08, 0.2, 0.45] and [0.4, 0.1, 0.5, 0.6, 0.3], it can be obtained that the similarity of the second eigenvector is about 0.705.

[0037] Step 106: Determine the third feature vector similarity based on the first feature vector similarity and the second feature vector similarity, and generate recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock combination strategy information based on the third feature vector similarity to determine the final target stock combination and the corresponding strategy algorithm.

[0038] Optionally, the third feature vector similarity can be the arithmetic mean of the first feature vector similarity and the second feature vector similarity. For the above example of stock D taking strategy X, the first feature vector similarity is about 0.893, the second feature vector similarity is about 0.705, and the third feature vector similarity is 0.799.

[0039] It should be noted that in the actual operation process, researchers can make feedback adjustments to the specific method and process of vectorization of each information based on whether the final calculation result of the similarity of the third eigenvector is consistent with the actual situation, so as to make the vectorization representation and similarity calculation of the information more accurate.

[0040] The above method can be used to calculate the third eigenvector similarities between all stocks and the corresponding strategy algorithms, and arrange all the third eigenvector similarities in descending order to sort the stocks and the corresponding strategy algorithms; based on the sorting results, a recommendation list including stock codes, strategy names and their similarity scores (generally set to the third eigenvector similarity in the present invention), i.e., recommendation priority information, can be generated.

[0041] It should be noted that after steps 101 to 104, the matching of the strategy algorithm of the stock has been completed. In the process of priority sorting, the stock and the corresponding strategy algorithm are "bound" together for sorting, and the subsequent process no longer involves the matching process of the strategy algorithm.

[0042] It should be noted that there are multiple ways to use the recommendation priority information here. The present invention can select a suitable usage method according to actual needs. The following introduces several common usage methods: (1) Sending the recommendation priority information to the management end so that the management end can understand the priority of the stocks and the corresponding strategy algorithms (i.e., the matching degree with the customers).

[0043] The management end here can be an application designed for securities companies, which is convenient for securities companies to manage configuration information such as policy routing tables and customer information. For example, the management end can issue update operation instructions to dynamically update one or more of the policy routing tables, target customers' policy algorithm usage permissions, and stock blacklists.

[0044] Assume that after step 104, according to the third stock combination strategy information, the strategy matching situation can be determined as follows: Stock 1+Strategy Algorithm A; Stock 2+ Strategy Algorithm B; Stock 3+ Strategy Algorithm C.

[0045] After steps 105 and 106, the generated recommendation priority information may be: 1. Stock 2+ strategy algorithm B, third eigenvector similarity: 0.7; 2. Stock 3+ strategy algorithm C, third eigenvector similarity: 0.6; 3. Stock 1+ strategy algorithm A. Similarity of the third eigenvector: 0.5.

[0046] (2) Send the recommendation priority information to the server so that the algorithm can determine the final target stock portfolio and the corresponding strategy algorithm based on the recommendation priority information.

[0047] For example, the recommendation priority information (which may only include stock information and corresponding similarity scores, but does not include strategy information corresponding to the stock) is: 1. Stock 2. Similarity of the third eigenvector: 0.7; 2. Stock 3. Similarity of the third eigenvector: 0.6; 3. Stock 1. Similarity of the third eigenvector: 0.5.

[0048] The user may select the top two stocks and inform the policy server of the selection result, so that the policy server can perform stock transactions of the top two stocks based on the policy algorithm determined in step 104 .

[0049] (3) The server directly uses the top stocks in the recommendation priority information as the target stock combination for stock trading. For example, the target stock combination is a few stocks ranked above the middle, or of course it can be all stocks. The specific situation can be set as needed to ensure flexibility, but the recommendation priority information is not sent to the client to improve the convenience of product use. There is no need to perform too much consideration and operation. The target user only needs to select a certain type of stock from the intraday revolving coupon pool at the beginning, and no excessive operations are required afterwards.

[0050] The intelligent routing method and system for the intraday reversal strategy algorithm provided by the present invention not only improves transaction efficiency, but also can effectively evaluate the degree of match between stocks and strategy algorithms and customer preferences, thereby providing more personalized and accurate solution recommendations and enhancing the user experience.

[0051] Based on the content of the above embodiment, as an optional embodiment, the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention generates recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock combination strategy information based on the third feature vector similarity, including: (1) Calculate the similarity between the target customer and other customers’ profile information, and determine the customer set whose similarity meets the preset conditions.

[0052] The present invention can extract portrait information of all customers (including target customers) from the client's database or other data sources. This information may include risk preference, investment period, historical trading behavior, etc.; use a suitable similarity measurement method (such as cosine similarity, Euclidean distance, etc.) to calculate the similarity between the target customer and all other customers; set a threshold and only retain those customers whose similarity is higher than this threshold as the customer set.

[0053] This step is to find a set of other customers who have similar investment preferences as the target customer.

[0054] (2) Statistically analyzing the historical transaction data of all customers in the customer set to prioritize each stock information and corresponding strategy algorithm information in the third stock portfolio strategy information; wherein the historical transaction data includes historical stock portfolio information and corresponding historical strategy algorithm information.

[0055] Extract historical transaction data of all customers in the customer set from the database or relevant data source, and ensure that the acquired data includes the time of each transaction, the stock code involved, the buying and selling actions, and the strategy algorithm information used.

[0056] Traversing the historical transaction data of all customers in the customer set, obtaining the number of times each stock and its corresponding strategy algorithm in the third stock portfolio strategy information has been used in the past period of time; According to the number of times it is used (of course, it can also be combined with other performance indicators, such as turnover rate, average return rate, etc.), a priority score is given to each stock information and the corresponding strategy algorithm information in the third stock portfolio strategy information.

[0057] (3) Generate recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock portfolio strategy information according to the priority score and the third eigenvector similarity.

[0058] The score obtained from the analysis of historical trading data is combined with the similarity of the third eigenvector, the total score of each stock and strategy is recalculated, and all stocks and strategies are sorted according to the total score to form a final recommendation priority list (recommendation priority information); the recommendation priority list includes stock codes, strategy names and their comprehensive scores.

[0059] It is understandable that the present invention can set different weights for these two factors to adjust their relative importance in the final score. In addition, in order to coordinate the problem of a large difference between the priority score and the third feature vector similarity value, all priority scores can be normalized first.

[0060] The intelligent routing method of the intraday reversal strategy algorithm provided by the present invention relies on a large amount of real and objective historical transaction data, and obtains recommendation priority information through statistical analysis, rather than relying solely on theory or hypothesis, which makes the recommendation more reliable and practical; when forming the recommendation priority information, the present invention takes into account both the information of the customer's own characteristics (the third eigenvector similarity) and the data support from similar customers (priority score), which makes the recommendation more reasonable and scientific, and enhances the user experience.

[0061] Optionally, the present invention can also send the recommended authorization quantity for each stock to the client to assist the target customer in making a decision and improve the user experience.

[0062] Optionally, the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention further includes: (1) Obtaining historical behavior data of each customer in the customer set, as well as its corresponding historical stock portfolio information and corresponding historical strategy algorithm information; the historical behavior data includes the number of times the user clicks on each stock in the client, and the length of time the user stays on each stock interface, etc.

[0063] (2) Take historical behavior data as input, and use its corresponding historical stock information and corresponding historical strategy algorithm information as output to train the customer decision model.

[0064] Specifically, the historical behavior data collected in the above step (1) is used as input data, and the corresponding historical stock information and strategy algorithm information is used as output labels to construct a training data set; according to business needs, a suitable machine learning or deep learning model is selected as the customer decision model to be trained.

[0065] The selected model is trained using the above training data set, and the model parameters are optimized to minimize the prediction error. The training goal is to enable the model to accurately predict the corresponding stock information and strategy algorithm based on the given behavioral data.

[0066] (3) The current behavior data of the target customers is input into the trained customer decision model, and the stock recommendation information and the corresponding strategy algorithm recommendation information are output.

[0067] (4) Collect statistics on all stock recommendation information and corresponding strategy algorithm recommendation information output by the customer decision model over a period of time.

[0068] (5) When the stock recommendation information and the corresponding strategy algorithm recommendation information exist in the third stock combination strategy information, the stock recommendation information and the corresponding strategy algorithm recommendation information are identified when generating the recommendation priority information.

[0069] Optionally, the identification method of the recommended information may be one or more of the following methods: highlight prompt, text prompt, etc.

[0070] Optionally, for the identified recommendation information, a higher ranking is given when generating the recommendation priority or its recommendation level is directly increased.

[0071] As an optional embodiment, in the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention, after determining the final target stock combination and the corresponding strategy algorithm based on the priority sorting information, the following steps are included: (1) Extract the semantic feature vector that describes the stock composition of the target stock portfolio.

[0072] The present invention can select key features that can accurately describe the stock portfolio status, and map the key features to vectors as the key features.

[0073] Preferably, the present invention can directly select the stock code as the key feature.

[0074] (2) Based on the semantic feature vector, at least one group of historical stock transaction data with the same stock composition status is retrieved from a pre-constructed historical stock transaction database.

[0075] Among them, each historical stock transaction data in the historical stock transaction database has corresponding stock composition information, authorized fund ratio information of each stock in the stock portfolio, and transaction performance information.

[0076] Specifically, the present invention can pre-build a database containing a large amount of historical stock trading data. Each historical record (historical stock trading data) should at least include the following information: stock portfolio information, that is, a list of specific stocks that constitute the portfolio; information on the percentage of authorized funds for each stock; and trading performance information (such as rate of return, etc.).

[0077] The present invention can use a similarity matching algorithm (such as cosine similarity, Euclidean distance) to search for historical transaction records with the same stock composition in the historical stock transaction database. It should be noted that the retrieval vector of the historical stock transaction data (the vector for similarity matching with the semantic feature vector) is obtained based on its stock combination information. The acquisition method can refer to the acquisition method of the semantic feature vector, which will not be repeated here.

[0078] It should be noted that after performing a retrieval based on semantic feature vectors, the present invention can also perform multi-dimensional secondary screening based on factors such as time windows and market conditions to ensure that the retrieved historical data is as close as possible to the current market environment and the needs of target customers.

[0079] (3) Determine the recommended information on the authorized capital ratio of the target stock portfolio based on the authorized capital ratio information retrieved from the historical stock transaction data.

[0080] Optionally, the current authorized funds ratio is determined based on an average value of the authorized funds ratio information in the retrieved historical stock transaction data.

[0081] Optionally, the present invention conducts in-depth analysis and evaluation of the retrieved historical stock trading data, particularly focusing on data that has performed well in the past and has controllable risks.

[0082] By combining the authorized funds ratio information in historical data and the preferences of individual customers, the authorized funds ratio of each stock in the target stock portfolio is ultimately determined to generate authorized funds ratio recommendation information.

[0083] Optionally, the authorized funds ratio recommendation information is fed back to the client and / or management end.

[0084] As an optional embodiment, the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention, the method for determining the authorized fund ratio of each stock in the target stock portfolio, includes: calculating the market value ratio of the market value of each stock in the target stock portfolio to the total market value of the target stock portfolio; generating the authorized fund ratio recommendation information of each stock according to the market value ratio of each stock. Optionally, the authorized fund ratio recommendation information is fed back to the client and / or management end.

[0085] For example, assuming that the authorized funds of the target customer are 500,000, and the target stock portfolio includes: stock G1, stock G2, stock G3 and stock G4, and the market value of stocks G1, G2, G3 and G4 in the total market value of the target stock portfolio accounts for 20%, 20%, 30% and 30% respectively. The authorized funds ratio recommendation information can be "The authorized funds ratios of stocks G1, G2, G3 and G4 are 20%, 20%, 30% and 30% respectively, and accordingly, the authorized funds are 100,000, 100,000, 150,000 and 150,000 respectively.".

[0086] The method for determining the authorized funds ratio of each stock provided by the present invention is particularly suitable for scenarios where rapid response to market changes is required or the decision-making process is desired to be simplified, thereby improving transaction efficiency and user experience.

[0087] Based on the contents of the above embodiments, as an optional embodiment, the present invention can also determine the current authorized fund ratio of each stock based on the technical analysis historical data of each stock; wherein the technical analysis historical data includes but is not limited to the historical authorized fund ratio, the overnight position ratio of individual stocks and the volatility of individual stocks.

[0088] Based on the content of the above embodiment, as an optional embodiment, the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention, the steps of constructing the strategy routing table include but are not limited to the following steps: (1) Receive the technical analysis historical data of any stock imported through the management terminal; the technical analysis historical data includes but is not limited to the historical transaction amount, volatility and turnover rate of the stock; Among them, the historical transaction amount is used to record the transaction amount data of the stock in different time periods; the volatility is used to characterize the volatility of stock prices, which is a measure of the uncertainty of stock returns and is used to reflect the risk level of financial assets; the turnover rate refers to the frequency of stock trading in the market within a certain period of time, which is one of the indicators reflecting the liquidity of stocks.

[0089] Of course, the technical analysis historical data in the present invention is not limited to the above data and can be adjusted according to needs.

[0090] (2) Call the stock backtesting tool through the API interface, and backtest each strategy algorithm based on technical analysis historical data to obtain the backtesting results.

[0091] Among them, stock backtesting means setting certain stock index combinations, based on the real market data that has occurred in the past, starting from a certain point in history, strictly selecting stocks according to the set combinations, and simulating the rules of real financial market transactions to perform model buying and model selling, and obtain data such as profit rate and maximum drawdown rate within a period of time. Stock backtesting tools, as the name suggests, are tools for stock backtesting, such as a stock backtesting software.

[0092] The present invention can develop an API interface for a stock backtesting tool, pass the acquired technical analysis historical data as input parameters to the API interface, start the backtesting process through the API call, and simulate the performance of each strategy algorithm under past market conditions.

[0093] Receive backtesting results from the backtesting tool, which usually include but are not limited to the following indicators: return analysis, risk assessment.

[0094] Taking all the above indicators into consideration, a comprehensive analysis of the performance of each strategy is conducted to identify which strategies perform well under specific conditions and which perform poorly.

[0095] (3) Based on the backtest results of each strategy algorithm, determine the strategy algorithm that must be adopted for any stock to construct a strategy routing table.

[0096] For example, for any stock A, three strategy algorithms are tested: strategy algorithm 1, strategy algorithm 2, and strategy algorithm 3; and three backtest results are obtained. Among them, strategy algorithm 2 is more in line with the customer profile of the target customer, so strategy algorithm 2 is determined to be the strategy algorithm that stock A must adopt.

[0097] Using the above method, multiple stocks are processed, and the final selected strategy algorithm is matched with the corresponding stock and recorded in the strategy routing table. This strategy routing table contains at least the following information: Stock code: identifies a specific stock; Recommended strategy: The best strategy algorithm corresponding to this stock.

[0098] Through this method, the present invention not only improves the scientificity and rationality of strategy selection, but also enhances the flexibility and adaptability of the system, which can better meet the personalized needs of different customers, while improving the overall transaction efficiency and user experience.

[0099] Based on the contents of the above embodiments, as an optional embodiment, the intelligent routing method of the intraday turnaround strategy algorithm provided by the present invention also includes: after determining that the stocks in the target stock portfolio selected by the target customer have not completed all execution cycles of the original strategy algorithm during the intraday turnaround transaction, and have been successfully re-matched to a new strategy algorithm, the new successfully matched strategy algorithm is run instead of the original strategy algorithm.

[0100] Specifically, if a stock in the target stock portfolio has an exposure during the intraday reversal transaction (the entire execution cycle of the strategy algorithm has not been completed), but the stock has been removed from the new securities pool, this scenario is not covered by the algorithm routing function, and the exposure monitoring function of the strategy algorithm management platform (server) will make a unified record. However, if the stock only changes the strategy algorithm, the customer's exposure data will be passed to the new strategy algorithm specified later and processed by the new strategy algorithm later.

[0101] Through the above dynamic strategy switching mechanism, the intelligent routing method of the intraday reversal strategy algorithm provided by the present invention can not only respond quickly to changes in market changes, but also ensure that all operations are carried out under the premise of safety and compliance, providing investors with a more scientific and reasonable trading solution. This method not only improves the intelligence level of the system, but also improves user experience and service quality.

[0102] The present invention also provides an intelligent routing system for an intraday reversal strategy algorithm, the system comprising: a client, a management end and a policy server; the policy server is communicated with the client and the management end respectively, and is used to implement the intelligent routing method for the intraday reversal strategy algorithm as described above.

[0103] The following is a brief description: 1. Some of the functions that can be realized by the client are as follows: 1) The present invention provides an intuitive and easy-to-use graphical user interface (GUI), which enables customers to easily view recommended security pools, select stocks, adjust fund allocation (of course, it can also be automatically allocated by the policy server), etc.

[0104] 2) Respond to user operation requests and pass these instructions to the policy server.

[0105] 3) Provide users with detailed transaction status updates, strategy switching notifications, and recommendation information to maintain transparency and user right to know.

[0106] 2. Some of the functions that can be realized by the management end are as follows: 1) Dynamically update one or more of the policy routing table, the target customer's policy algorithm usage rights, and the stock blacklist.

[0107] 2) Realize the management and maintenance of customer information.

[0108] 3. Some functional modules of the policy server are as follows: 1) A first processing module, configured to receive initial stock portfolio information input by a target customer through a client; the initial stock portfolio information is generated based on the stock selection operation of the target customer on the client, and includes information of multiple stocks selected by the target customer; 2) A second processing module is used to call the strategy routing table to match the strategy algorithm for each stock information in the initial stock combination information to generate the first stock combination strategy information; the strategy routing table includes multiple stock information and the strategy algorithm matched by each stock information, and the first stock combination strategy information includes the successfully matched stock information and the corresponding strategy algorithm information; 3) A third processing module, configured to determine the target customer's policy algorithm usage authority based on the target customer's user identity, filter out the stock information and corresponding policy algorithm information that do not meet the policy algorithm usage authority from the first stock combination policy information, and generate the second stock combination policy information; 4) A fourth processing module, configured to call the stock blacklist to filter out the stock information and corresponding strategy algorithm information not in the stock blacklist from the second stock combination strategy information, so as to generate the third stock combination strategy information; 5) A fifth processing module, used to calculate a first feature vector similarity between the stock parameter information of each stock in the third stock portfolio strategy information and the customer profile information of the target customer, and a second feature vector similarity between the strategy algorithm information of each stock and the customer profile information of the target customer; 6) A sixth processing module, configured to determine a third feature vector similarity according to the first feature vector similarity and the second feature vector similarity, and generate recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock combination strategy information based on the third feature vector similarity.

[0109] In summary, the intelligent routing method and system of the intraday reversal strategy algorithm provided by the present invention has the following beneficial effects compared with the prior art: (1) The present invention realizes the aggregation of strategy algorithms of multiple suppliers, and through the automated strategy matching process, reduces manual intervention, speeds up the process from stock selection to transaction execution, and improves transaction efficiency; it can also effectively evaluate the degree of match between stocks and strategy algorithms and customer preferences, thereby providing more personalized and accurate solution recommendations and enhancing the user experience.

[0110] (2) The intelligent routing method of the intraday reversal strategy algorithm provided by the present invention relies on a large amount of real and objective historical transaction data, and obtains recommendation priority information through data analysis, rather than relying solely on theory or hypothesis, which makes the recommendation more reliable and practical; and, when forming the recommendation priority information, the present invention takes into account both the information of the customer's own characteristics (the third eigenvector similarity) and the data support from similar customers (priority score), so the recommendation is more reasonable, which greatly enhances the user experience.

[0111] (3) The present invention supports the integration of multiple strategy algorithms and matches the optimal strategy algorithm for the customer's stock portfolio from multiple strategy algorithms, thus having strong strategy adaptability.

[0112] (4) The present invention introduces a blacklist mechanism to exclude specific stocks that are not suitable for strategy trading; sets customer algorithm permissions to ensure that each customer can only use strategies that match themselves, which is more targeted.

[0113] (5) The present invention can realize automatic matching of strategy algorithms without requiring customers to perform excessive operations, thereby improving customers' stock trading experience; securities dealers only need to regularly maintain the strategy routing table, customer algorithm permissions, and stock blacklists to achieve effective active management and dynamic adjustment of different stocks and their corresponding strategies, thereby improving work and management efficiency.

[0114] Figure 2 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 2 As shown, the electronic device may include: a processor 210, a communication interface 220, a memory 230 and a communication bus 240, wherein the processor 210, the communication interface 220 and the memory 230 communicate with each other through the communication bus 240. The processor 210 may call the logic instructions in the memory 230 to execute the intelligent routing method of the intraday reversal strategy algorithm, the method comprising: determining the stock portfolio selected by the target customer in the target intraday reversal coupon pool; matching the algorithm strategy for the stocks in the stock portfolio based on the preset strategy routing table, the customer algorithm authority and the stock blacklist; removing the stocks that failed to match the algorithm strategy from the stock portfolio, retaining the stocks that successfully matched the algorithm strategy to construct the target stock portfolio; determining the authorized fund ratio of each stock in the target stock portfolio, so as to run the strategy algorithm that successfully matched based on the authorized fund ratio of each stock.

[0115] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the intelligent routing method of the intraday reversal strategy algorithm provided in the above-mentioned embodiments, and the method includes: determining the stock portfolio selected by the target customer in the target intraday reversal coupon pool; matching the algorithm strategy for the stocks in the stock portfolio based on a preset strategy routing table, customer algorithm permissions and a stock blacklist; eliminating stocks that fail to match the algorithm strategy from the stock portfolio, and retaining stocks that successfully match the algorithm strategy to construct a target stock portfolio; determining the authorized fund ratio of each stock in the target stock portfolio, and running the successfully matched strategy algorithm based on the authorized fund ratio of each stock.

[0116] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an intelligent routing method for executing the intraday reversal strategy algorithm provided in the above-mentioned embodiments, the method comprising: determining the stock portfolio selected by the target customer in the target intraday reversal coupon pool; matching the algorithm strategy for the stocks in the stock portfolio based on a preset strategy routing table, customer algorithm permissions, and a stock blacklist; removing stocks that fail to match the algorithm strategy from the stock portfolio, and retaining stocks that successfully match the algorithm strategy to construct a target stock portfolio; determining the authorized fund ratio of each stock in the target stock portfolio, and running the successfully matched strategy algorithm based on the authorized fund ratio of each stock.

[0117] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, i.e., may be located in one place, or may be distributed to multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art may understand and implement the solution without creative effort.

[0118] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent routing method for an intraday reversal strategy algorithm, characterized in that: include: Receiving initial stock portfolio information input by the target customer through the client; the initial stock portfolio information is generated based on the stock selection operation of the target customer on the client, including information of multiple stocks selected by the target customer; Calling the strategy routing table to match the strategy algorithm for each stock information in the initial stock combination information to generate first stock combination strategy information; the strategy routing table includes multiple stock information and the strategy algorithm matched by each stock information, and the first stock combination strategy information includes successfully matched stock information and corresponding strategy algorithm information; Determine the target customer's policy algorithm usage authority based on the target customer's user identity, filter out the stock information and corresponding policy algorithm information that do not meet the policy algorithm usage authority from the first stock combination policy information, and generate the second stock combination policy information; Calling the stock blacklist to filter out the stock information and corresponding strategy algorithm information not in the stock blacklist from the second stock combination strategy information to generate the third stock combination strategy information; Calculate the first feature vector similarity between the stock parameter information of each stock in the third stock portfolio strategy information and the customer profile information of the target customer, and the second feature vector similarity between the strategy algorithm information of each stock and the customer profile information of the target customer; A third feature vector similarity is determined according to the first feature vector similarity and the second feature vector similarity, and recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock combination strategy information is generated based on the third feature vector similarity.

2. The intelligent routing method of the intraday reversal strategy algorithm according to claim 1, characterized in that: Based on the third feature vector similarity, recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock combination strategy information is generated, including: Calculate the similarity between the target customer and other customers’ profile information, and determine the customer set whose similarity meets the preset conditions; Performing statistical analysis on the historical transaction data of all customers in the customer set to prioritize each stock information and the corresponding strategy algorithm information in the third stock combination strategy information; wherein the historical transaction data includes historical stock combination information and the corresponding historical strategy algorithm information; According to the priority score and the third eigenvector similarity, recommendation priority information of the stock information and the corresponding strategy algorithm information in the third stock combination strategy information is generated.

3. The intelligent routing method of the intraday reversal strategy algorithm according to claim 1, characterized in that: After determining the final target stock portfolio and the corresponding strategy algorithm based on the priority ranking information, it also includes: Calculate the market value share of each stock in the target stock portfolio in the total market value of the target stock portfolio; Based on the market value ratio of each stock, the recommended information of the authorized fund ratio of each stock is generated.

4. The intelligent routing method of the intraday reversal strategy algorithm according to claim 1, characterized in that: After determining the final target stock portfolio and the corresponding strategy algorithm based on the priority ranking information, it also includes: Extracting a semantic feature vector describing the stock portfolio status of the target stock portfolio; Based on the semantic feature vector, at least one group of historical stock trading data with the same stock composition is retrieved from a pre-built historical stock trading database; wherein each historical stock trading data in the historical stock trading database has corresponding stock combination information and authorized fund ratio information of each stock in the stock combination; According to the authorized funds ratio information in the retrieved historical stock transaction data, the authorized funds ratio recommendation information of the target stock portfolio is determined.

5. The intelligent routing method of the intraday reversal strategy algorithm according to claim 1, characterized in that: The steps to construct a policy routing table include: Receive technical analysis historical data of any stock; technical analysis historical data includes: historical transaction amount, volatility and turnover rate of the stock; Call the stock backtesting tool through the API interface, and backtest each strategy algorithm based on technical analysis historical data to obtain backtesting results; Based on the backtest results of each strategy algorithm, determine the strategy algorithm that must be adopted for any stock to build a strategy routing table.

6. The intelligent routing method of the intraday reversal strategy algorithm according to claim 1, characterized in that: Also includes: In response to the update operation instruction, one or more of the policy routing table, the target customer's policy algorithm usage authority, and the stock blacklist are dynamically updated.

7. The intelligent routing method of the intraday reversal strategy algorithm according to claim 1, characterized in that: Also includes: When it is determined that the stocks in the target stock portfolio selected by the target customer have not completed all execution cycles of the original strategy algorithm during intraday turnaround transactions and have been successfully re-matched to a new strategy algorithm, the new successfully matched strategy algorithm will be run instead of the original strategy algorithm.

8. An intelligent routing system for an intraday reversal strategy algorithm, characterized in that: include: A client, a management terminal and a policy server; the policy server is respectively connected to the client and the management terminal for implementing the intelligent routing method of the intraday reversal policy algorithm as described in any one of claims 1 to 7.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the intelligent routing method of the intraday reversal strategy algorithm as described in any one of claims 1 to 7 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent routing method of the intraday reversal strategy algorithm as described in any one of claims 1 to 7 are implemented.

Images