Transaction terminal automatic order placement method and system, electronic device and storage medium

CN122550255APending Publication Date: 2026-08-11TIANLIANXIN (WUHAN) SEMICONDUCTOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-11
Publication Date
2026-08-11

AI Technical Summary

Technical Problem

[0003]然而,现有基于界面的自动下单方案在证券交易场景下仍存在以下不足:其一,交易终端页面可能存在异步刷新、风险提示弹窗、确认弹窗、登录失效及网络波动等情况,若未先判断当前页面状态是否稳定并限定当前状态下允许执行的动作,则容易在错误页面继续执行输入或提交操作;其二,现有方案通常更关注是否完成字段输入,而缺少对页面实际显示的证券代码、价格、数量、方向或账户信息的再次读取和校验,导致输入后的页面显示内容可能与原始交易意图不一致;其三,委托提交后,现有方案往往仅依据一次提示信息判断下单结果,缺少结合委托查询记录进行复核的闭环确认机制,因而难以可靠判断委托是否真正进入交易系统

Benefits of technology

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: Without relying on standard trading interfaces or brokerage open interfaces, this invention reduces erroneous operations under abnormal states such as page refresh, pop-ups, and login failures by stably determining the page state and executing state machine constraints; by reading back and verifying page fields and comparing them with transaction information, it helps to detect inconsistencies in account, securities code, trading direction, price, or quantity before order submission; through closed-loop verification of results after order submission, it reduces misjudgments caused by single prompt messages; and by suppressing duplicate submissions of orders with uncertain results, it reduces the possibility of duplicate orders of the same type, thus helping to improve the accuracy and security of automatic order placement in securities trading terminals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122550255A_ABST
    Figure CN122550255A_ABST
Patent Text Reader

Abstract

This invention discloses a method, system, electronic device, and storage medium for automatic order placement on a trading terminal. The method includes: receiving a structured trading intent and a target terminal adaptation configuration; determining the target trading information carried by the structured trading intent and the current page observation data; determining a stable page state matching the page observation data from multiple candidate page states; navigating from the stable page state to a target order placement page matching the target trading information; and performing automatic order placement processing on the structured trading intent based on valid trading information. The beneficial effects of this invention are: reducing erroneous operations under abnormal states such as page refresh, pop-ups, and login failures; by reading back and verifying page fields and comparing them with trading information, it helps to detect inconsistencies in account, securities code, trading direction, price, or quantity before order submission; reducing misjudgments caused by single prompt information; and improving the accuracy and security of automatic order placement on securities trading terminals.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of financial technology, specifically to a method, system, electronic device, and storage medium for automatic order placement on a trading terminal. Background Technology

[0002] In existing automated trading solutions, one type of solution submits orders by calling open interfaces of brokerage firms, quantitative trading terminals, or intermediate gateways. Another type uses desktop automation, web automation, or general RPA technology to simulate human clicks, inputs, and confirmations on securities trading terminals. However, in the context of individual investor tools, not all brokerage clients or web trading terminals provide ordinary users with a unified, stable, and directly callable standard trading interface. Therefore, automated order placement often relies solely on the trading terminal page itself.

[0003] However, existing interface-based automatic order placement solutions still have the following shortcomings in securities trading scenarios: First, the trading terminal page may experience asynchronous refreshes, risk warning pop-ups, confirmation pop-ups, login failures, and network fluctuations. If the stability of the current page state is not determined first and the actions allowed under the current state are not limited, it is easy to continue to perform input or submission operations on the error page. Second, existing solutions usually focus more on whether the field input is completed, and lack the re-reading and verification of the actual security code, price, quantity, direction, or account information displayed on the page, which may lead to the content displayed on the page after input being inconsistent with the original trading intention. Third, after the order is submitted, existing solutions often judge the order result based on only one prompt message, lacking a closed-loop confirmation mechanism that combines order query records for review, making it difficult to reliably determine whether the order has actually entered the trading system. Summary of the Invention

[0004] The purpose of this invention is to provide a method for automatically placing orders on a trading terminal to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a method for automatically placing orders on a securities trading terminal, comprising: Receive structured transaction intent and target terminal adaptation configuration, wherein the target terminal adaptation configuration includes page anchors, field areas, button areas and state transition rules; Determine the target transaction information carried by the structured transaction intent and the current page observation data, wherein the page observation data is status identification data collected from the transaction terminal page; Determine a stable page state that matches the page observation data from a variety of candidate page states, wherein the stable page state is a page state that is consistently identified multiple times. Navigate from the stable page to the target order page that matches the target transaction information; Fields on the target order page that have been verified and are consistent with the target transaction information are identified as valid transaction information. Automatic order placement is performed on the structured transaction intent based on valid transaction information.

[0006] Preferably, the structured trading intent includes at least the target account, security code, trading market, trading direction, target order price, and order quantity; The target terminal adaptation configuration is used to constrain the execution rules of page navigation, field filling, status recognition, and delegated submission.

[0007] Preferably, the page observation data includes one or more of the following: full-page screenshot, partial area screenshot, OCR text recognition results, control tree information, and DOM structure information; The stable state of the page is determined by continuously collecting data at a preset sampling interval and continuously identifying the same state a preset number of times.

[0008] Preferably, navigating to the target order page means determining the next action that can be executed based on the current stable state and the target transaction direction; When the page is in a stable state such as a risk warning pop-up, an unknown state, a page refresh state, or a login failure state, field entry and delegation submission are prohibited.

[0009] Preferably, the readback verification includes reading the account identifier, securities code, transaction direction, order price, and order quantity actually displayed on the page; The reading result is compared with the target transaction information in the structured transaction intent. If the comparison is consistent, the verification is deemed successful.

[0010] Preferably, the readback verification of the order price includes calculating the minimum bid unit difference between the readback price and the target order price; When the difference is not greater than the preset maximum number of errors, the commission price verification is deemed successful.

[0011] Preferably, the automatic order placement process includes submitting the order and performing closed-loop verification of the results by combining the prompt information with the order query record; When the verification result is uncertain, a similar delegate signature is generated, and duplicate submissions of delegates with the same signature are suppressed for a preset blocking period.

[0012] Secondly, embodiments of the present invention provide an automatic order placement system for a securities trading terminal, comprising: The transaction intent preprocessing module is used to receive structured transaction intents and target terminal adaptation configurations, wherein the target terminal adaptation configurations include page anchors, field areas, button areas, and state transition rules. The status determination and page navigation module is used to determine the current page observation data and the target transaction information carried by the structured transaction intent, wherein the page observation data is status identification data collected from the transaction terminal page; The field filling and readback verification module is used to determine the stable page state that matches the page observation data from a variety of candidate page states. The stable page state is the page state that is identified consistently multiple times. The target page navigation module is used to navigate from the stable state of the page to the target order page that matches the target transaction information. The submission verification and suppression module is used to determine the field information in the target order page that has been read back and verified and is consistent with the target transaction information as valid transaction information; The automatic order execution module is used to automatically place orders based on valid transaction information for the structured transaction intent.

[0013] Thirdly, embodiments of the present invention provide an electronic device including a memory, a processor, and a computer program stored in the memory, wherein when the processor executes the computer program, it implements the method described in any one of the first aspects.

[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any one of the first aspects of the present invention.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: Without relying on standard trading interfaces or brokerage open interfaces, this invention reduces erroneous operations under abnormal states such as page refresh, pop-ups, and login failures by stably determining the page state and executing state machine constraints; by reading back and verifying page fields and comparing them with transaction information, it helps to detect inconsistencies in account, securities code, trading direction, price, or quantity before order submission; through closed-loop verification of results after order submission, it reduces misjudgments caused by single prompt messages; and by suppressing duplicate submissions of orders with uncertain results, it reduces the possibility of duplicate orders of the same type, thus helping to improve the accuracy and security of automatic order placement in securities trading terminals. Attached Figure Description

[0016] Figure 1 This application embodiment illustrates the application environment of an automatic order placement system for a securities trading terminal. Figure 2 The overall flowchart of the automatic order placement method for securities trading terminals in this application is shown in the embodiment. Figure 3 Flowchart of field filling and readback verification in this application embodiment Figure 4 This application's embodiment of the securities trading terminal automatic order placement system framework diagram Figure 5 This application contains an embodiment of an electronic device structural block diagram. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please refer to Figure 1, which is a schematic diagram of an application environment for automatic order placement in a securities trading terminal according to an exemplary embodiment. As shown in Figure 1, the application environment may include an automatic order execution system 100, a securities trading terminal 200, and an external information source 300.

[0019] The automatic order execution system 100 receives structured trading intentions and performs page status recognition, page navigation, field readback verification, and order submission processing on the securities trading terminal 200. The automatic order execution system 100 may include multiple sub-modules, including a trading intention preprocessing module 110, a status determination and page navigation module 120, a field filling and readback verification module 130, and a submission verification and suppression module 140. The trading intention preprocessing module 110 acquires and verifies structured trading intentions and loads the appropriate configuration for the current terminal; the status determination and page navigation module 120 collects page observation data, determines the page's stable state, and navigates to the target order page; the field filling and readback verification module 130 fills in fields on the target order page and performs readback verification of the actual displayed content to determine valid trading information; the submission verification and suppression module 140 completes order submission based on valid trading information and performs closed-loop verification of the order result, suppressing duplicate submissions of the same type of order when the result is uncertain.

[0020] External information source 300 provides auxiliary information related to automated order placement, such as transaction time verification, network connection status, account fund status, risk control rules information, and log retention information. External information source 300 may include time services, network status services, log storage services, and risk control rules services. Automated order placement execution system 100 can interact with external information source 300 to complete pre-transaction verification, anomaly handling, and process evidence retention.

[0021] The securities trading terminal 200 provides a trading interface and order submission portal. The securities trading terminal 200 can be a brokerage desktop client, a web-based trading terminal, or other terminal with a trading interface. The securities trading terminal 200 displays the order page, input fields, prompts, pop-up messages, and order query records. The automatic order execution system 100 does not directly call the brokerage's open interface; instead, it completes the automatic order placement process through interface observation and controlled operation of the securities trading terminal 200.

[0022] In addition, through the collaborative work of the status determination and page navigation module 120, the field filling and readback verification module 130, and the submission verification and suppression module 140, the page status can be stably determined, the field information can be consistently verified, the entrustment result can be closed-loop verified, and the repeated submissions in scenarios with uncertain results can be suppressed, which helps to improve the accuracy and security of the automatic order placement process.

[0023] In an optional embodiment, the automatic order execution system 100, the securities trading terminal 200, and the external information source 300 can be connected via a LAN (local area network), WAN (wide area network), or mobile data network to realize real-time transmission and feedback of page observation data, trading intent information, and order result information.

[0024] In practical applications, through the collaborative work of various modules in the automatic order execution system 100, and combined with auxiliary information provided by external information source 300, page navigation, field validation, order submission and result verification can be stably completed without relying on the standard open interface of the brokerage firm. This adapts to the interface differences of different brokerage terminals and the ever-changing operating environment, thereby improving the reliability of the automatic order placement process.

[0025] Furthermore, it should be noted that Figure 1 only shows one application environment of the automatic order placement method for securities trading terminals provided in this disclosure.

[0026] It should be noted that the following diagram illustrates one possible sequence of steps, and it is not strictly required to follow this order. Some steps can be executed in parallel without interdependence. The user information (including but not limited to user account information, device information, transaction behavior information, etc.) and data (including but not limited to transaction records, field readback data, page status data, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0027] Please see Figure 2 , Figure 2 A flowchart of an automatic order placement method for a securities trading terminal according to this application can be shown.

[0028] In step S1, the structured transaction intent and target terminal adaptation configuration are received.

[0029] In the embodiments described in this specification, the structured trading intent can be a trading instruction containing core trading information, generated by a strategy module, rule engine, or manual triggering module. The target terminal adaptation configuration can be a set of page adaptation rules that match the current brokerage client or web trading terminal. The target terminal adaptation configuration can associate page anchor areas, field input areas, button trigger areas, prompt information areas, and state transition rules, and include configuration items such as state determination thresholds, sampling intervals, and verification parameters.

[0030] In one possible implementation, structured trading intentions can be obtained from local strategy services, cloud trading services, or human-interactive interfaces. After receiving the structured trading intentions, the automated order execution system can parse them and load an adaptation configuration that matches the currently running securities trading terminal for subsequent page status recognition and navigation.

[0031] In another possible implementation, structured trading intentions can be received through real-time communication with the upper-level trading system. For example, a quantitative trading system or rule-triggered system pushes trading instructions to an automated order execution system via a network interface. Upon receiving the instructions, the system verifies their validity and loads the corresponding terminal adaptation rules to ensure consistency between the page operation and the terminal interface.

[0032] In practical applications, by receiving structured transaction intents and loading target terminal adaptation configurations, subsequent page observation, status determination, field filling and submission operations can conform to the interface structure and interaction logic of the current terminal, which helps to improve the stability and adaptability of the automatic order placement process.

[0033] In one possible implementation, a structured trading intent may include at least the target account, security code, trading market, trading direction, target order price, and order quantity, and may further include fields such as order type, validity period, risk control parameters, and trading intent generation time.

[0034] In one optional implementation, the signature for similar orders can consist of a target account identifier, security code, trading direction, order quantity, trading market, normalized price index, and time window identifier. The normalized price index can be discretized based on the target order price and the minimum price unit corresponding to the target market, enabling the system to identify semantically identical or highly similar duplicate orders when the outcome is uncertain. If the same order signature is found in the duplicate submission suppression table, and its status is still pending confirmation or the outcome is uncertain, then submitting similar orders again is prohibited for a preset blocking period.

[0035] In step S2, the current page observation data and the target transaction information carried by the structured transaction intent are determined. The page observation data is status identification data collected from the transaction terminal page.

[0036] In the embodiments of this specification, the current page observation data can be various types of data collected from the securities trading terminal page for status recognition, including one or more of the following: full-page screenshot, partial area screenshot, OCR text recognition results, control tree information, and DOM structure information. The target transaction information can be transaction-related information extracted from structured trading intentions, including the target account, security code, trading market, trading direction, order price, and order quantity.

[0037] In one possible implementation, the automated order execution system continuously collects data from the transaction terminal page at preset sampling intervals to obtain page observation data, and identifies candidate page states based on page anchor points, text features, and interface structure. When the same candidate state appears consecutively a preset number of times, the candidate state is determined as a stable page state.

[0038] In practical applications, by determining the page observation data and target transaction information, a data foundation can be provided for determining the stability of the page state, avoiding erroneous operations when the page is refreshed, pop-ups appear, or loading is incomplete, and improving the reliability of page navigation.

[0039] In step S3, a stable page state that matches the page observation data is determined from multiple candidate page states, and the system navigates to the target order page that matches the target transaction information.

[0040] In the embodiments of this specification, a stable page state refers to an operable page state where the identification results are consistent after multiple consecutive identifications and the interface no longer changes. This state is used to exclude unstable situations such as asynchronous refresh, pop-up obstruction, loading, and login failure. The target order page can be a buy page or a sell page that matches the transaction direction.

[0041] In one possible implementation, the automatic order execution system judges the current stable state based on state machine constraints. When the page state is a risk warning pop-up, an unknown state, a page refresh state, or a login failure state, it prohibits field filling and order submission. When the page state is a stable and tradable state, it navigates to the target order page consistent with the trading direction.

[0042] In one alternative implementation, the stable page state that matches the page observation data from multiple candidate page states can be obtained using a round-robin verification method.

[0043] Furthermore, the system prompts the user to enter the security code on the target order page. After completion, the system reads at least one or more of the following: the security code displayed on the page, the security name displayed on the page, the trading market identifier displayed on the page, and the tradable direction prompt on the page, as the basis for the first round of verification. Only when the read results match the target security code and the target market will the system proceed to the subsequent stage of filling in the price and quantity fields.

[0044] After the security code verification passes, the system sequentially fills in the order price and order quantity. Once completed, the field readback verification module reads one or more of the following currently displayed parameters: security code, trading direction, target account identifier, order price, order quantity, and order type, and performs a second round of verification. Specifically, the security code must match the target security code, the trading direction must match the target trading direction, the target account identifier must match the target account in the trading intent, and the order quantity must match the target order quantity. If the page has an order type field, the order type must also match the target order type.

[0045] Regarding the order price, this application allows for the use of the minimum quotation unit corresponding to the target trading market. Establish rules for determining price consistency. Let the price for page reload be... The target order price is The number of minimum price unit differences between the two is When the number of errors is not greater than the preset maximum number. When the price is verified, it can be determined that the price verification has passed. The formula for determining price consistency is:

[0046] When the following conditions are met: At that time, the page price verification is deemed successful.

[0047] in, This indicates the price of the page request obtained from the readback. Indicates the target order price in the trading intent; This represents the minimum price unit corresponding to the target market m.

[0048] When strict consistency is required, It can be 0; however, when the brokerage terminal has display format processing, rounding, or price correction rules, Alternatively, you can choose 1. This setting balances project feasibility with transaction semantic consistency, preventing system misjudgments caused by slight deviations in the display layer.

[0049] In step S4, the field information in the target order page that has been read back and verified and is consistent with the target transaction information is determined as valid transaction information, and automatic order placement is performed on the structured transaction intent based on the valid transaction information.

[0050] In the embodiments of this specification, readback verification can refer to reading the account identifier, security code, trading direction, order price, and order quantity actually displayed on the page, and comparing them with the target trading information for consistency. Valid trading information can refer to the set of assignable fields that have all fields verified and are consistent with the trading intent.

[0051] In one possible implementation, the automatic order execution system reads the actual content displayed on the page after the fields are filled in, performs consistency checks on the account, code, direction, and quantity, and performs difference checks on the order price according to the minimum quotation unit. All fields that pass the checks are determined as valid transaction information.

[0052] In practical applications, verifying valid transaction information through readback can identify inconsistencies in fields before order submission. Combined with closed-loop verification of results after submission and suppression of duplicate submissions, this helps improve the accuracy and security of automated order placement.

[0053] In one optional implementation, after the action execution module performs the submission action, if a confirmation pop-up, risk warning pop-up, or other blocking page appears before or after submission, the system will handle it accordingly based on the pop-up type. For pop-ups defined in the adaptation configuration and allowing automatic confirmation, the system performs a confirmation action; for pop-ups defined in the adaptation configuration but requiring manual confirmation, the system transfers the process to manual control; for unidentified or unconfigured pop-ups, the system stops the automatic submission process and records evidence. This approach balances automation efficiency with avoiding continued operation on unknown high-risk pop-ups.

[0054] In one optional implementation, a closed-loop verification mechanism can be used for readback verification. Specifically, the system can comprehensively consider at least one piece of evidence, such as the real-time prompt text on the page, the order number returned on the page, the newly added order record in the order query page, the security code, trading direction, order price, order quantity in the newly added order record, and whether the record generation time falls within a preset time window, to judge the submission result. If a clear failure prompt is detected and there is no corresponding order record in the order query page, the order is determined to have failed; if a success prompt is detected and there is a unique matching newly added order record in the order query page, the order is determined to have succeeded; if no clear success or failure prompt is detected, or no unique matching record can be found in the order query page, the result is determined to be uncertain.

[0055] When the system determines the result to be uncertain, it initiates a duplicate submission suppression process. Specifically, the system can automatically access the entrustment query page for a second review and refresh the entrustment records according to the review interval within the review time window. If a unique matching entrustment record is found in a subsequent refresh, the result can be updated to "order placed successfully." If confirmation is still not possible after the review, the entrustment signature corresponding to that entrustment is written to the duplicate submission suppression table, and entrustments with the same entrustment signature are prohibited from being submitted again during the blocking period. At the same time, screenshots of the entire process, field values, status values, and reasons for the exception are recorded, and manual intervention or abnormal termination is triggered. This can effectively reduce the risk of duplicate orders.

[0056] Furthermore, in an optional embodiment of the present invention, the target terminal is a desktop brokerage client. The page observation and acquisition module obtains a full-page screenshot through the operating system screenshot interface and performs partial cropping of the securities code field area, price field area, quantity field area, account area, and prompt area. The current status recognition module determines whether the current page is the logged-in homepage, the buy page, the risk warning pop-up page, or the order query page by combining page anchor matching and OCR text recognition. Subsequently, the system sequentially executes pre-trade verification, page navigation, securities code entry and first-round verification, order price and order quantity entry and second-round verification, order submission, and result closed-loop verification. If only a click action is detected but it cannot be confirmed whether the actual entry into the trading system has occurred, the process proceeds to the result uncertainty handling flow.

[0057] In an optional embodiment of the present invention, the target terminal is a web-based trading terminal. The page observation and acquisition module prioritizes reading the available DOM structure information and combines it with page screenshots for status identification. For example, the system can determine whether it is currently on the buy page based on the buy page label, securities code input box node, price input box node, quantity input box node, and submit button node on the webpage. It should be noted that even if the DOM field values ​​are readable, the system still prefers to perform a readback verification of the visible content on the page before submission, rather than directly replacing the actual displayed values ​​on the page with the cached DOM values. In other words, only when both the DOM field values ​​and the page OCR recognition values ​​match the target trading intention... Figure 1 The system only performs the submission action upon receiving the confirmation message. This reduces the risks associated with front-end cache not being refreshed, input not actually taking effect, or temporary inconsistencies between the displayed page value and the node value.

[0058] In an optional embodiment of the present invention, the system is applicable to scenarios where multiple securities accounts can be switched on the same terminal. Before entering the buy page, the system first reads the current account identifier area to determine whether the currently logged-in account is consistent with the target account in the trading intent. If they are inconsistent, the system does not continue to fill in the securities code, price, and quantity, but instead performs an account switch or terminates the transaction directly. After the information is filled in, the system reads the account identifier area again for verification before submission to prevent account switching errors from occurring during page redirection, page refresh, or pop-up processing. By verifying the account identifier before and after submission, the risk of placing orders with the wrong account in multi-account scenarios can be reduced.

[0059] In an optional embodiment of the present invention, the system sets processing strategies for various abnormal situations, including login failure, network interruption or page unresponsiveness, failure to load the buy page, field not found, incorrect securities code display, incorrect order price display, incorrect order quantity display, inconsistent target account, unidentified pop-up, and uncertain results. Correspondingly, the system can perform one or more actions such as re-collecting the page and re-judging the status, returning to the homepage and re-navigating, refreshing the page, terminating the current process, triggering manual intervention, and recording the abnormal code and evidence data. By pre-setting abnormal handling branches, the method can maintain high engineering stability even in complex terminal environments.

[0060] In another implementation, the system retains the original transaction intent, terminal identifier and adapted configuration version, screenshots of key steps, page OCR text results, field readback values, current status identification results, actual action sequence, submission time, result classification, exception codes, and manual takeover records. This evidentiary data can be used for post-event tracing and dispute localization, as well as for adapted configuration optimization and rule iteration.

[0061] Please see Figure 3 , Figure 3 This diagram illustrates the structural block of an automatic order placement system for a securities trading terminal according to this application. The transaction intent preprocessing module 10 is used to receive structured transaction intents and target terminal adaptation configurations, wherein the target terminal adaptation configurations include page anchors, field areas, button areas, and state transition rules. The status determination and page navigation module 20 is used to determine the current page observation data and the target transaction information carried by the structured transaction intent, wherein the page observation data is status identification data collected from the transaction terminal page; The field filling and readback verification module 30 is used to determine the stable state of the page that matches the page observation data from a variety of candidate page states. The stable state of the page is the page state that is identified consistently multiple times. The stable state of the page is used to navigate to the target order page that matches the target transaction information. The submission verification and suppression module 40 is used to determine the field information in the target order page that has been read back and verified and is consistent with the target transaction information as valid transaction information, and to perform automatic order placement processing on the structured transaction intent based on the valid transaction information.

[0062] In one possible implementation, the target transaction information includes the target account, security code, trading market, trading direction, order price, and order quantity; Transaction intent preprocessing module 10 includes: The configuration loading unit is used to load page adaptation rules that match the current trading terminal, so as to adapt to the interface layout and interaction methods of different brokerage terminals.

[0063] In one possible implementation, the page observation data includes one or more of the following: full-page screenshot, partial area screenshot, OCR text recognition results, control tree information, and DOM structure information; The status determination and page navigation module 20 includes: The data acquisition unit is used to continuously collect page status data according to a preset sampling interval and extract target transaction information from the transaction intent.

[0064] In one possible implementation, the field filling and readback verification module 103 includes: The continuous recognition unit is used to determine the page as a stable state when the same candidate page state appears continuously for a preset number of times.

[0065] The state machine constraint unit is used to determine the allowed operations based on the current stable state of the page and the target transaction information, and to prohibit field filling and delegation submission in abnormal states.

[0066] In one possible implementation, the submission verification and suppression module 104 includes: The readback verification unit is used to read the account identifier, securities code, transaction direction, order price and order quantity actually displayed on the page, and compare them with the target transaction information. If the comparison is consistent, it is determined to be valid transaction information.

[0067] The order submission unit is used to submit orders based on valid transaction information; The closed-loop verification unit is used to verify the results by combining the prompt information with the entrusted query record; The duplicate suppression unit is used to suppress duplicate submissions of the same type of request when the verification result is uncertain.

[0068] It should be understood that the aforementioned modules can be implemented in software, hardware circuits, or a combination of both. For the specific execution process of each module in the device embodiment, please refer to the relevant descriptions in the aforementioned method embodiments; they will not be repeated here.

[0069] Please see Figure 5 , Figure 5 A schematic diagram of the structure of an electronic device according to this application is shown. The electronic device may include a processor, a memory, a network interface, a display device, and an input device. The processor executes a computer program stored in the memory to implement any of the aforementioned method embodiments; the network interface is used for data communication with a securities trading terminal, a log service, or other external devices; the display device may be a liquid crystal display (LCD); and the input device may be a keyboard, mouse, touchpad, or touch layer. This electronic device can be a local computer, a server, or other devices with computing and control capabilities.

[0070] It should be understood that the illustrated electronic device is merely an example and does not constitute a limitation of this application. In other embodiments, the electronic device may include more components, reduce the number of components, combine some components, or adopt different component arrangements. As long as the foregoing method steps can be performed, they should fall within the protection scope of this application.

[0071] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is used to implement any of the foregoing method embodiments. The storage medium may be a read-only memory, random access memory, magnetic disk, optical disk, flash memory, or other medium capable of storing program code.

[0072] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for automatically placing orders for a securities trading terminal, characterized by, include: Receive structured transaction intent and target terminal adaptation configuration, wherein the target terminal adaptation configuration includes page anchors, field areas, button areas and state transition rules; Determine the target transaction information carried by the structured transaction intent and the current page observation data, wherein the page observation data is status identification data collected from the transaction terminal page; Determine the stable page state that matches the page observation data from multiple candidate page states, and navigate from the stable page state to the target order page that matches the target transaction information; Fields on the target order page that have been verified and are consistent with the target transaction information are identified as valid transaction information. Based on the valid transaction information, automatic order placement is performed on the structured transaction intent.

2. The method of claim 1, wherein, The structured trading intent includes at least the target account, security code, trading market, trading direction, target order price, and order quantity; The target terminal adaptation configuration is used to constrain the execution rules of page navigation, field filling, status recognition, and delegated submission.

3. The method of claim 2, wherein, The page observation data includes one or more of the following: full-page screenshots, partial area screenshots, OCR text recognition results, control tree information, and DOM structure information; The stable state of the page is determined by continuously collecting data at a preset sampling interval and continuously identifying the same state a preset number of times.

4. The method of claim 3, wherein, Navigating to the target order page means determining the next action that can be performed based on the current stable state and the target transaction direction; When the page is in a stable state such as a risk warning pop-up, an unknown state, a page refresh state, or a login failure state, field entry and delegation submission are prohibited.

5. The method of claim 4, wherein, The readback verification includes reading the account identifier, securities code, trading direction, order price, and order quantity actually displayed on the page; The reading result is compared with the target transaction information in the structured transaction intent. If the comparison is consistent, the verification is deemed successful.

6. The method of claim 5, wherein, The readback verification of the order price includes calculating the minimum bid unit difference between the readback price and the target order price; When the difference is not greater than the preset maximum number of errors, the commission price verification is deemed successful.

7. The method of claim 6, wherein, The automatic order placement process includes submitting the order and performing closed-loop verification of the results by combining the prompt information with the order query record; When the verification result is uncertain, a similar delegate signature is generated, and duplicate submissions of delegates with the same signature are suppressed for a preset blocking period.

8. A system for automatically placing orders at a securities trading terminal, characterized by include: The transaction intent preprocessing module is used to receive structured transaction intents and target terminal adaptation configurations, wherein the target terminal adaptation configurations include page anchors, field areas, button areas, and state transition rules. The status determination and page navigation module is used to determine the current page observation data and the target transaction information carried by the structured transaction intent, wherein the page observation data is status identification data collected from the transaction terminal page; The field filling and readback verification module is used to determine the stable page state that matches the page observation data from a variety of candidate page states. The stable page state is the page state that is identified consistently multiple times. The module navigates from the stable page state to the target order page that matches the target transaction information. The submission verification and suppression module is used to determine the field information in the target order page that has been read back and verified and is consistent with the target transaction information as valid transaction information, and to perform automatic order placement processing on the structured transaction intent based on the valid transaction information.

9. An electronic device, comprising: The method includes a memory, a processor, and a computer program stored in the memory, wherein when the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.