A menu navigation method, device, equipment, storage medium and product
By supporting natural language input and using large language models for menu navigation, the problems of complexity and inefficiency of menu navigation in the existing technology are solved, and a more efficient and accurate menu navigation experience is achieved.
Patent Information
- Application Number
- CN202410941077.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-15
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-07-15
AI Technical Summary
When the existing menu navigation method deals with complex menu structures and multi-level structures, users need to remember a large number of specific instructions, which increases learning cost and difficulty in use, and is huge in computing, resulting in low efficiency and accuracy.
Provide a menu navigation method that supports natural language input. By obtaining navigation text matching with menu node information, candidate menu nodes are determined, and pre-trained large language model is used for intent recognition, reducing computational complexity and time-consuming.
No need to rely on predefined keywords or fixed instruction patterns, which reduces user learning costs and difficulty in using, improves the efficiency and accuracy of menu navigation, and simplifies user interaction experience.
Smart Images

Figure CN118468890B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a menu navigation method, device, equipment, storage medium and product. Background Art
[0002] With the continuous development of information technology, various software, applications and systems are equipped with complex menu structures so that users can access a variety of functions and information. However, with the increasing complexity and hierarchy of menu structures, users often face difficulties in navigation, especially in applications with a large number of menu nodes and deep hierarchical structures. Users often need to switch repeatedly between multiple levels to find the required functions or information, which greatly reduces the user experience.
[0003] Existing menu navigation methods often have limitations when dealing with complex menu structures and multi-level menu navigation. For example, some menu navigation methods rely on predefined keywords or fixed command patterns. When the menu structure is complex, users need to remember a large number of specific commands to operate, which increases the user's learning cost and difficulty of use; some menu navigation methods rely solely on intent recognition. When the menu structure is complex and the number of menu nodes is large, a large number of menu nodes often need to be considered when performing intent recognition, resulting in huge computational complexity, time-consuming and inefficient. These problems seriously affect the user experience and reduce the efficiency and accuracy of menu navigation. Summary of the invention
[0004] In view of this, the embodiments of the present application are committed to providing a menu navigation method, apparatus, device, storage medium and product, which not only support the user's natural language input, but also reduce the computational complexity and time consumption, and improve the efficiency and accuracy of menu navigation.
[0005] According to a first aspect of an embodiment of the present application, a menu navigation method is provided, including:
[0006] Get the first navigation text;
[0007] Matching the first navigation text with node information of each menu node in the first menu node set, and determining a plurality of candidate menu nodes in the first menu node set according to the matching result;
[0008] Menu node intention recognition is performed on the first navigation text according to the node information corresponding to each candidate menu node, and a target menu node corresponding to the first navigation text is determined from each candidate menu node.
[0009] Optionally, the method further includes:
[0010] Acquire historical interaction information of the first navigation text;
[0011] The performing menu node intention recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determining the target menu node corresponding to the first navigation text in each candidate menu node, includes:
[0012] According to the node information corresponding to each candidate menu node and the historical interaction information, menu node intent recognition is performed on the first navigation text, and a target menu node corresponding to the first navigation text is determined in each candidate menu node.
[0013] Optionally, the performing menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node and the historical interaction information, and determining a target menu node corresponding to the first navigation text in each candidate menu node, includes:
[0014] Generate a task instruction according to the node information corresponding to each candidate menu node, the historical interaction information and the first navigation text, the task instruction being used to instruct the object receiving the task instruction to perform menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node and the historical interaction information, and determine the target menu node corresponding to the first navigation text in each candidate menu node;
[0015] The task instruction is input into a pre-trained large language model, and the pre-trained large language model executes the task instruction to determine a target menu node corresponding to the first navigation text among various candidate menu nodes.
[0016] Optionally, matching the first navigation text with node information of each menu node in the first menu node set, and determining a plurality of candidate menu nodes in the first menu node set according to the matching result, includes:
[0017] Matching the first navigation text with the node information of each menu node in the first menu node set, and determining the matching degree between the first navigation text and the node information of each menu node in the first menu node set;
[0018] According to the matching degree between the first navigation text and the node information of each menu node in the first menu node set, a preset number of candidate menu nodes are determined in the first menu node set.
[0019] Optionally, the method further includes:
[0020] If the target menu node corresponding to the first navigation text is not determined in each candidate menu node, generating prompt information, the prompt information is used to prompt the user to input supplementary navigation information;
[0021] and / or,
[0022] A target navigation path is determined according to the target menu node, wherein the target navigation path includes a navigation path from the current menu node to the target menu node.
[0023] Optionally, the matching includes keyword matching and / or semantic similarity matching, and the node information includes node name and / or node description information.
[0024] Optionally, the node information also includes entity information associated with the menu node and / or a navigation text example corresponding to the menu node, and the entity information includes at least one of an entity name, an entity alias, and an entity description information.
[0025] According to a second aspect of an embodiment of the present application, a menu navigation device is provided, including:
[0026] The first unit is used to obtain the first navigation text;
[0027] A second unit is used to match the first navigation text with the node information of each menu node in the first menu node set, and determine a plurality of candidate menu nodes in the first menu node set according to the matching result;
[0028] The third unit is used to perform menu node intention recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determine the target menu node corresponding to the first navigation text in each candidate menu node.
[0029] According to a third aspect of an embodiment of the present application, there is provided an electronic device, including a memory and a processor;
[0030] The memory is connected to the processor and is used to store programs;
[0031] The processor is used to implement the menu navigation method as described in any one of the first aspects of the embodiments of the present application by running the program in the memory.
[0032] According to a fourth aspect of an embodiment of the present application, a storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the menu navigation method as described in any one of the first aspects of the embodiments of the present application is implemented.
[0033] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, comprising computer program instructions, which, when executed by a processor, enable the processor to implement a menu navigation method as described in any one of the first aspects of the embodiments of the present application.
[0034] The menu navigation method provided in the present application first obtains a first navigation text; then matches the first navigation text with the node information of each menu node in a first menu node set, and determines a plurality of candidate menu nodes in the first menu node set according to the matching results; finally, performs menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determines a target menu node corresponding to the first navigation text in each candidate menu node.
[0035] The menu navigation method provided by the present application, on the one hand, does not need to rely on predefined keywords or fixed command modes, supports users' natural language input, and reduces users' learning costs and usage difficulties; on the other hand, by performing preliminary and rapid matching of the first navigation text with the menu nodes in the first menu node set, multiple candidate menu nodes are identified, which greatly narrows the scope of subsequent menu node intention recognition. When performing menu node intention recognition, only the candidate menus determined by the preliminary match need to be considered, which not only greatly reduces the computational complexity and time consumption, improves navigation efficiency, but also improves the accuracy of menu navigation. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0037] Figure 1 A structural diagram of a menu navigation system provided in an embodiment of the present application.
[0038] Figure 2 A flowchart of the first menu navigation method provided in an embodiment of the present application.
[0039] Figure 3 A flowchart of a second menu navigation method provided in an embodiment of the present application.
[0040] Figure 4 A schematic diagram of a processing flow for determining a target menu node corresponding to a first navigation text from among various candidate menu nodes provided in an embodiment of the present application.
[0041] Figure 5 A flowchart of the third menu navigation method provided in an embodiment of the present application.
[0042] Figure 6 A structural schematic diagram of a menu navigation device provided in an embodiment of the present application.
[0043] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0044] The technical solution of the embodiment of the present application is suitable for application in various scenarios that require menu navigation, especially application scenarios with complex, multi-level menu systems, such as mobile banking, smart home, car systems, self-service customer service scenarios, etc. In these scenarios, users usually need to quickly and accurately locate the required functions or information. The technical solution of the embodiment of the present application can reduce the computational complexity and time consumption, significantly improve the efficiency and accuracy of menu navigation, and bring users a smoother and more convenient interactive experience.
[0045] The technical solution provided in the embodiments of the present application can be exemplarily applied to hardware devices such as processors, electronic devices, servers (including cloud servers), or packaged into software programs to be run. When the hardware device executes the processing of the technical solution of the embodiments of the present application, or the above-mentioned software program is run, the automatic splitting of the target task and the automatic calling of the application program interface required for the task can be achieved to complete the purpose of the target task. The embodiments of the present application only exemplarily introduce the specific processing of the technical solution of the present application, and do not limit the specific implementation form of the technical solution of the present application. Any technical implementation form that can execute the processing of the technical solution of the present application can be adopted by the embodiments of the present application.
[0046] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0047] Before introducing this application solution, the relevant technologies are first introduced:
[0048] With the continuous development of information technology, various software, applications and systems are equipped with complex menu structures so that users can access a variety of functions and information. However, with the increasing complexity and hierarchy of menu structures, users often face difficulties in navigation, especially in applications with a large number of menu nodes and deep hierarchical structures. Users often need to switch repeatedly between multiple levels to find the required functions or information, which greatly reduces the user experience.
[0049] Existing menu navigation methods often have limitations when dealing with complex menu structures and multi-level menu navigation. For example, some menu navigation methods rely on predefined keywords or fixed command patterns. When the menu structure is complex, users need to remember a large number of specific commands to operate, which increases the user's learning cost and difficulty of use; some menu navigation methods rely on intent recognition. When the menu structure is complex and there are many menu nodes, a large number of menu nodes need to be considered when performing intent recognition, resulting in huge computational complexity, time-consuming and inefficient. These problems seriously affect the user experience and reduce the efficiency and accuracy of menu navigation.
[0050] In view of this, the embodiments of the present application provide a menu navigation method, apparatus, device, storage medium and product, which not only support the user's natural language input, but also reduce the computational complexity and time consumption, and improve the efficiency and accuracy of menu navigation, which are described in detail one by one in the following embodiments.
[0051] Exemplary Systems
[0052] For ease of understanding, firstly, an exemplary introduction to the implementation environment of the menu navigation method provided in the embodiment of the present application is given. Figure 1 , Figure 1 Schematic diagram of the structure of a menu navigation system. The menu navigation method provided in the present application can be exemplarily applied to the menu navigation system.
[0053] like Figure 1 As shown, the menu navigation system includes an instruction acquisition module 110, a processor 120 and an output module 130, and the processor 120 is communicatively connected with the instruction acquisition module 110 and the output module 130 respectively.
[0054] The instruction acquisition module 110 is used to receive navigation instruction input from the user, and send the received navigation instruction to the processor 120 via the communication network.
[0055] The instruction acquisition module 110 can be a voice input component or a text input component. The voice input component can be any functional component that can support the user to perform voice input, such as a microphone, a pickup, etc., and the specific processing process of voice collection can refer to the implementation process of conventional technology. The text input component can be any functional component that can support the user to perform text input, such as a touch screen, a keyboard, etc., and the specific processing process of text collection can refer to the implementation process of conventional technology.
[0056] The processor 120 is used to execute the menu navigation method provided by the embodiment of the present application after obtaining the navigation instruction sent by the instruction acquisition module 110. Specifically, first obtain the first navigation text; then match the first navigation text with the node information of each menu node in the first menu node set, and determine multiple candidate menu nodes in the first menu node set according to the matching results; finally, perform menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determine the target menu node corresponding to the first navigation text in each candidate menu node. As for other specific implementations of the menu navigation method, they will be specifically expanded in the subsequent content and will not be described in detail here.
[0057] It should be noted that when the instruction acquisition module 110 is a text input component, the navigation instruction received by the processor 120 is a text instruction, which is the first navigation text; when the instruction acquisition module 110 is a voice input component, the navigation instruction received by the processor 120 is a voice instruction, and the first navigation text is the transcribed text of the voice instruction.
[0058] The processor 120 can be various types of processing devices, the specific type of which is determined according to the scale, application scenario and performance requirements of the menu navigation system. It can be an independent server, a server cluster consisting of multiple physical servers, or a cloud server providing cloud computing services.
[0059] The output module 130 is used to display information corresponding to the target menu node determined by the processor 120 to the user, or to perform corresponding operations according to the target menu node, such as jumping to the corresponding function interface, playing related multimedia content, etc. At the same time, the output module 130 can also be used to provide feedback to the user, such as voice prompts, text prompts, or graphical interface prompts, etc., to enhance the user interaction experience.
[0060] Exemplarily, the output module 130 may be an output device such as a display screen or a speaker, which is used to transmit information to the user in a visual or auditory form.
[0061] Exemplary Methods
[0062] Figure 2 The flowchart of the first voice navigation method provided in the embodiment of the present application is shown in FIG. Figure 2 As shown, the voice navigation method provided in this embodiment includes steps S201-S203:
[0063] S201: Obtain a first navigation text.
[0064] The first navigation text may be understood as a text instruction for navigating in a menu navigation system, and generally includes a user's description of a desired menu item or function, and can reflect the user's intention.
[0065] Optionally, the first navigation text may include but is not limited to the following: menu node name, function description or requirement, keyword or phrase, natural language instruction, etc. For example, "I lost my savings card and need to report the loss as soon as possible", "Please open the device menu", "How to change the password", "I want to view account information", "Please take me to the shopping cart", and "How do I change my delivery address" etc.
[0066] Optionally, obtaining the first navigation text includes: a text input component of the menu navigation system receives the first navigation text input by the user, and sends the first navigation text to a processor, and the processor obtains the first navigation text.
[0067] Optionally, obtaining the first navigation text includes: a voice input component of the menu navigation system receives a first navigation voice input by a user, and sends the first navigation voice to a processor, and after receiving the first navigation voice, the processor performs voice recognition on the first navigation voice to obtain a transcribed text of the first navigation voice, i.e., the first navigation text. As for the specific content of voice recognition, it can be implemented with reference to relevant existing technologies, and this application does not limit this.
[0068] Optionally, obtaining the first navigation text includes: the processor receives the first navigation text sent by other applications or services.
[0069] S202: Match the first navigation text with node information of each menu node in the first menu node set, and determine a plurality of candidate menu nodes in the first menu node set according to the matching result.
[0070] A menu node can be understood as an independent unit in the menu navigation, usually represented by a clickable item or link, which can be a top-level menu node or a sub-menu node nested under a top-level menu node. For example, in the navigation menu of a financial system, debit card charges and loss reports, debit card loss reports, electronic payments, payment settlements, account management, account details, corporate information changes, help center, contact customer service, etc. can all be considered menu nodes.
[0071] The first menu node set can be understood as a set of menu nodes pre-defined in the menu navigation system and used to construct a navigation structure, and each menu node represents a different functional area, information classification or operation entry.
[0072] Optionally, the first menu node set includes all menu nodes in the menu navigation system. Exemplarily, in a menu navigation system of an e-commerce website, the "first menu node set" includes top-level menu nodes such as "Home", "Product Category", "Shopping Cart", and "My Order"; each top-level menu node may contain multiple sub-menu nodes, which also belong to the first menu node set, for example, "Product Category" may contain sub-menu nodes such as "Electronic Products", "Household Supplies", "Clothing, Shoes and Hats"; further, the sub-menu nodes may also contain more detailed sub-menu nodes, which also belong to the first menu node set, for example, "Electronic Products" may also contain more detailed sub-nodes such as "Mobile Phones", "Computers", and "Cameras".
[0073] The node information of a menu node can be understood as information associated with the menu node, which usually includes element information used to describe and identify the characteristics of the menu node, and can help understand the content and function of the menu node.
[0074] Optionally, the node information of the menu node includes a node name and / or node description information.
[0075] The node name is a direct identifier of the menu node, which can help quickly identify the function and content of the menu node, such as reporting a lost debit card, swallowing a card, adjusting accounts, and changing corporate information.
[0076] The node description information provides a more detailed description of the menu node function, which helps to further explain the node name, and may include the specific content of the node, the category to which it belongs, the usage scenario, etc. For example, the description information of the "debit card loss report" node may be "report the loss of a bank card at the counter or by phone", the description information of the "card swallowing and account adjustment" node may be "involving POS or ATM device card swallowing and account adjustment processing", and the description information of the "enterprise information change" node may be "services for enterprises to update bank account information due to information changes".
[0077] Optionally, the node information of the menu node may also include entity information associated with the menu node, node ID, parent node ID, navigation path, etc., which can help locate the menu node, and this application does not limit this.
[0078] It should be noted that before executing the menu navigation method provided by the present application, the first menu node set and the node information of the menu nodes in the first menu node set have been set and stored in advance. When executing the menu navigation method provided by the present application, the processor can directly call the preset first menu node set and the node information of each menu node in the first menu node set.
[0079] Optionally, matching the first navigation text with node information of each menu node in the first menu node set, and determining a plurality of candidate menu nodes in the first menu node set according to the matching result, comprises steps A1-A2:
[0080] A1. Match the first navigation text with node information of each menu node in the first menu node set to determine a matching degree between the first navigation text and node information of each menu node in the first menu node set.
[0081] Specifically, after obtaining the first navigation text, the processor calls the node information of each menu node in the pre-set first menu node set, and performs a first match between the first navigation text and the node information of each menu node in the first menu node set, so as to obtain the matching degree between the first navigation text and the node information of each menu node in the first menu node set.
[0082] The matching degree can be understood as the similarity or association between the first navigation text and the node information of each menu node in the first menu node set, and is used to quantify the degree of fit between the first navigation text and each menu node.
[0083] Optionally, the first matching may adopt a single matching algorithm or a combination of multiple matching algorithms. When the first matching adopts a single matching algorithm, the matching degree calculated by the matching algorithm is the matching degree of the first matching. When the first matching adopts a combination of multiple matching algorithms, the matching degree calculated by each matching algorithm is subjected to weighted calculation and other mathematical operations to obtain the matching degree of the first matching.
[0084] Optionally, the first matching may be any one of keyword matching and semantic similarity matching, or a combination of the two.
[0085] Keyword matching can be understood as comparing the occurrence of keywords in the first navigation text and the node information of each menu node. According to the comparison result, the matching degree between the first navigation text and the node information of each menu node can be determined.
[0086] Semantic similarity matching can be understood as comparing the similarity in meaning between the first navigation text and the node information of each menu node. According to the comparison result, the matching degree between the first navigation text and the node information of each menu node can be determined.
[0087] Combining keyword matching with semantic similarity matching can take into account the efficiency, flexibility and robustness of the first matching, making the matching result of the first matching more accurate.
[0088] As for the specific implementation methods of keyword matching, semantic similarity matching and weighted calculation, they can all be implemented by referring to relevant existing technologies, and this application will not provide detailed descriptions for this.
[0089] A2. Determine a preset number of candidate menu nodes in the first menu node set according to a matching degree between the first navigation text and node information of each menu node in the first menu node set.
[0090] Optionally, after determining the degree of match between the first navigation text and the node information of each menu node in the first menu node set, a preset number of menu nodes with the highest degree of match are selected from the first menu node set as candidate menu nodes in descending order of the degree of match. Exemplarily, the menu navigation system includes a total of 100 menu nodes, that is, the first menu node set includes 100 menu nodes. When the preset number is set to 10, the candidate menu nodes finally determined are the 10 menu nodes with the highest degree of match.
[0091] Optionally, after determining the matching degree between the first navigation text and the node information of each menu node in the first menu node set, all menu nodes having a matching degree greater than a preset matching degree threshold are selected as candidate menu nodes.
[0092] According to the foregoing content, it can be determined that the higher the matching degree between the first navigation text and the node information of the menu node, the higher the degree of consistency between the first navigation text and the menu node, and the greater the possibility that the target menu node corresponding to the first navigation text is the menu node. Therefore, the candidate menu nodes include multiple menu nodes in the first menu node set that are most consistent with the target menu node corresponding to the first navigation text.
[0093] It should be noted that the number of candidate menu nodes and the specific value of the preset matching threshold are determined according to the actual application scenario, and this application does not limit this.
[0094] Optionally, after determining the candidate menu nodes, the candidate menu nodes may be optimized, for example, by adjusting and optimizing the ranking of the candidate menus through a user feedback loop to make them more in line with the actual needs and preferences of the user.
[0095] S203: performing menu node intent recognition on the first navigation text according to node information corresponding to each candidate menu node, and determining a target menu node corresponding to the first navigation text from among each candidate menu node.
[0096] The target menu node may be understood as the menu node that the user wishes to access through this input.
[0097] Performing menu node intention recognition on the first navigation text can be understood as analyzing the first navigation text to determine the access intention of the first navigation text to the menu node.
[0098] Optionally, conventional intent recognition technology can be used to perform intent recognition on the first navigation text, determine the recognition result of the intent, and then deeply match the result of the intent recognition with the node information corresponding to each candidate menu node to determine the matching degree corresponding to each candidate menu node, and select the candidate menu node with the highest matching degree as the target menu node.
[0099] Furthermore, when performing intent recognition on the first navigation text, not only the first navigation text itself is considered, but also the node information of each candidate menu node is comprehensively considered. When performing intent recognition on the first navigation text, the node information of the candidate menu nodes can provide semantic and contextual support for the first navigation text, helping the processor to better understand the specific meaning of the first navigation text in a specific scenario, which is conducive to accurately understanding the intent of the user's query.
[0100] As an optional implementation manner, the node information of the menu node may further include entity information associated with the menu node, and the entity information includes at least one of an entity name, an entity alias, and entity description information.
[0101] The entity information associated with the menu node can be understood as the relevant information of the specific business object or data model represented in the menu node.
[0102] For example, the menu node is "Report Lost Debit Card", the entity name of the entity associated with the menu node is "Debit Card", the entity alias is "Bank Card, Savings Card, Magnetic Stripe Card, Chip Card, etc.", and the entity description information is "A bank card issued by the issuing bank to the cardholder, with no credit limit, in which the cardholder deposits money first and then uses it."
[0103] When performing menu node intent recognition on the first navigation text, the entity information associated with the candidate menu nodes can provide richer semantic and contextual support for the first navigation text. On the one hand, the entity alias provides additional entity keywords. For example, if a specific product or service name is mentioned in the first navigation text, and the name happens to match the associated entity name or alias of a menu node, then the system can more confidently determine that the user's intention is to access the menu node. On the other hand, the entity description information provides the system with a detailed description of the entity's functions and features. For example, if the first navigation text's description of the entity is vague and does not use the preset entity name and entity alias, but the description of the entity matches the preset entity description information, then the system can determine the entity based on the entity description information, and then determine the user's intention.
[0104] As an optional implementation manner, the node information of the menu node may further include a navigation text example corresponding to the menu node.
[0105] The navigation text example corresponding to the menu node can be understood as a navigation text sample whose target menu node is the menu node, which is usually closely related to the function or content of the menu node and can be regarded as a typical navigation text that a user may input.
[0106] For example, if the menu node is "Report Lost Debit Card", the navigation text example corresponding to the menu node may be "I accidentally lost my card and need to report it lost"; if the menu node is "Corporate Information Change", the navigation text example corresponding to the menu node may be "Change my company's bank account information".
[0107] When performing menu node intent recognition on the first navigation text, the navigation text examples corresponding to the candidate menu nodes will provide additional references for the intent recognition process. For example, by comparing the first navigation text with the navigation text examples of each candidate menu node, an additional and more direct reference point can be provided to the processor, which helps reduce misjudgment and ambiguity and helps the processor better determine the user's menu node intent.
[0108] Specifically, after determining multiple candidate menu nodes with the highest matching degree with the first navigation text, the processor executes the first strategy and analyzes the first navigation text using the node information corresponding to the candidate menu nodes to determine the user's true intention, that is, to determine the target menu node that the user wants to access and operate.
[0109] There are multiple ways to implement the first strategy. Optionally, the first strategy is implemented based on the processing concept of the model. An intention recognition model is pre-trained, and the node information corresponding to each candidate menu node and the first navigation text are input into the intention recognition model to obtain the target menu node. Among them, the trained intention recognition model can perform menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determine the target menu node corresponding to the first navigation text in each candidate menu node.
[0110] Furthermore, a pre-trained large language model can be used to implement the first strategy. The powerful language processing and matching capabilities of the large language model can further improve the accuracy of intent recognition. This will be elaborated in subsequent content and will not be explained in detail here.
[0111] The menu navigation method provided by the present application, on the one hand, does not need to rely on predefined keywords or fixed command modes, supports users' natural language input, and reduces users' learning costs and usage difficulties; on the other hand, by performing preliminary and rapid matching of the first navigation text with the menu nodes in the first menu node set, multiple candidate menu nodes are identified, which greatly narrows the scope of subsequent menu node intention recognition. When performing menu node intention recognition, only the candidate menus determined by the preliminary match need to be considered, which not only greatly reduces the computational complexity and time consumption, improves navigation efficiency, but also improves the accuracy of menu navigation.
[0112] Figure 3 A flowchart of a second menu navigation method provided in an embodiment of the present application is shown as follows: Figure 3 As shown, the menu navigation method provided in this embodiment includes steps S301-S304:
[0113] S301: Obtain a first navigation text.
[0114] S302: Match the first navigation text with the node information of each menu node in the first menu node set, and determine a plurality of candidate menu nodes in the first menu node set according to the matching result.
[0115] The content of step S301 corresponds to the content of step S201 in the above embodiment, and the content of step S302 corresponds to the content of step S202 in the above embodiment. The specific contents of steps S301 and S302 can be found in Figure 2 The contents of the illustrated embodiments will not be repeated here.
[0116] S303: Acquire historical interaction information of the first navigation text.
[0117] The historical interaction information can be understood as the interaction information between the user and the menu navigation system before the current dialogue. Optionally, the historical interaction information includes data records generated when the user interacts with the menu navigation system in the past preset time period. Optionally, the historical interaction information includes a preset number of interaction data records between the user and the menu navigation system before the current dialogue and with the shortest time interval from the current dialogue.
[0118] Optionally, the historical interaction information includes multiple rounds of historical dialogue information of a current session between the user and the menu navigation system.
[0119] The multiple rounds of historical conversations in the current session can provide the first navigation text with semantic content, contextual information, and contextual information associated with the first navigation text, which can help the processor better understand the content expressed by the first navigation text and more accurately identify the menu node intention corresponding to the first navigation text. In addition, in the face of the diversity of user input, the multiple rounds of historical conversation information provide the system with additional data points, allowing the processor to handle various input situations more robustly. For example, if the first navigation text is "how to modify", if the user mentions in the historical conversation of the current session that he wants to check the order status, then the processor can infer from the historical conversation that the user may want to know how to modify the order, rather than other unrelated operations.
[0120] Optionally, the historical interaction information includes historical behaviors of the user, and the historical behaviors include but are not limited to a combination of one or more of the menu nodes visited by the user, dwell time, operation path, search query, and click behaviors.
[0121] The user's historical behavior can reflect the user's behavioral habits and preferences. By analyzing these historical behaviors, the processor can learn and memorize the user's behavior patterns and preferences, so that when the first navigation text is used to identify the menu node intent, the user's behavior patterns and preferences can be used as a basis for auxiliary judgment. For example, when the first navigation text is relatively vague, contains multiple intents, and is not enough to clearly point to a candidate menu node, the processor can combine the user's behavior patterns and preferences to assist in judgment and determine the most likely target menu node, thereby providing users with more personalized and accurate services.
[0122] In summary, historical interaction information can help the processor understand user needs more comprehensively, thereby more accurately identifying the user's navigation intentions and improving the user experience.
[0123] S304: According to the node information corresponding to each candidate menu node and the historical interaction information, menu node intent recognition is performed on the first navigation text, and a target menu node corresponding to the first navigation text is determined from each candidate menu node.
[0124] Specifically, after the processor determines multiple candidate menu nodes in the first menu node set and obtains the historical interaction information of the first navigation text, it calls the pre-set node information corresponding to each candidate menu node, and performs a comprehensive analysis of the historical interaction information of the first navigation text, the node information corresponding to each candidate menu node, and the first navigation text. Based on the three factors of the historical interaction information of the first navigation text, the node information corresponding to each candidate menu node, and the first navigation text, the menu node intention of the first navigation text is jointly identified, and a candidate menu node that best matches the intention of the first navigation text is determined from multiple candidate menu nodes as the target menu node.
[0125] Furthermore, a pre-trained large language model can be used to implement menu node intent recognition for the first navigation text. As an optional implementation method, Figure 4 As shown, step S304 includes steps S401-S402:
[0126] S401, generating a task instruction according to the node information corresponding to each candidate menu node, the historical interaction information and the first navigation text, wherein the task instruction is used to instruct the object receiving the task instruction to perform menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node and the historical interaction information, and determine the target menu node corresponding to the first navigation text in each candidate menu node.
[0127] The task instruction may be a sentence, a question or a combination of a series of instructions. The present application does not limit the specific content and format of the task instruction.
[0128] Optionally, the task instruction includes node information corresponding to each candidate menu node, historical interaction information, and a first navigation text.
[0129] Optionally, the task instruction includes node information corresponding to each candidate menu node, historical interaction information, a first navigation text, and a task description text. The task description text can be understood as text content that describes the task so that the object receiving the task instruction can clearly understand the task.
[0130] Further, the contents included in the task instruction are combined according to a preset task instruction format to obtain the task instruction. The task instruction format is used to specify the structure and representation method of the task instruction, which is pre-set and stored on the controller or server side and can be directly retrieved and used when needed. The task instruction format is determined according to the actual application scenario and requirements, and this application does not limit this.
[0131] For example, the specific content of a task instruction is as follows:
[0132] Prompt: Assuming you are an intelligent interactive assistant, please determine the target menu node of the first navigation text in the candidate menu node based on the node information and historical interaction information of the given candidate menu node. Requirements: 1. The answer is output in the json format of [{"target menu node"}]; 2. If there is no suitable target menu node in the candidate menu node, the predicted target menu node is "other"
[0133] ##First navigation text: Please report the loss as soon as possible
[0134] ##Node information of candidate menu nodes: {"Debit card loss report": "desc: After a bank card is lost, report it at the counter or by phone; example: I accidentally lost my card and need to report it lost", "Company information change": "desc: A service where a company needs to update its bank account information due to information changes; example: Change my company's bank account information", "...": "..."}
[0135] ##Node information of candidate menu node: {alias: savings card, desc: bank card used by the cardholder to deposit money first and then use it}
[0136] ##Historical interaction information: {User: "Hello, my savings card seems to be lost, what should I do?"; Customer Service: "Hello, I'm sorry to hear that your savings card is lost. In order to protect the safety of your funds, it is recommended that you report the loss as soon as possible"}
[0137] S402: Input the task instruction into a pre-trained large language model, and execute the task instruction by the pre-trained large language model to determine a target menu node corresponding to the first navigation text from among the candidate menu nodes.
[0138] A large language model (LLM) refers to a generative deep neural network model based on a Transformer structure, which has a powerful natural language processing capability. It should be noted that any large language model can be used in the embodiments of the present application.
[0139] The embodiment of the present application requires pre-training of a large language model. Specifically, a data set consisting of multiple groups of task instructions and target menu nodes is used as a training sample to pre-train the existing large language model, so that the trained large language model can execute the task instruction, that is, execute the task corresponding to the task instruction, perform menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node and the historical interaction information, and determine the target menu node corresponding to the first navigation text in each candidate menu node.
[0140] Specifically, after generating the task instructions, the generated task instructions are input into the pre-trained large language model. The large language model executes the task instructions, performs menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node and the historical interaction information, and outputs the target menu node corresponding to the first navigation text.
[0141] In this way, relying on the excellent natural language processing and matching capabilities of the pre-trained large language model, it can comprehensively and accurately understand the node information, historical interaction information and first navigation text corresponding to each candidate menu node, and at the same time deeply understand the task description text. On this basis, it can comprehensively consider multiple factors to perform more accurate menu node intent recognition on the first navigation text.
[0142] Moreover, this solution is applicable to any scenario. It only needs to obtain the first navigation text, the historical interaction information of the first navigation text, and the node information corresponding to each candidate menu node. The target menu node corresponding to the first navigation text can be determined with the help of the powerful capability of the large language model, which has higher universality.
[0143] Figure 5 A flowchart of a third menu navigation method provided in an embodiment of the present application is shown as follows: Figure 5 As shown, the menu navigation method provided in this embodiment includes steps S501-S505:
[0144] S501: Obtain a first navigation text.
[0145] S502: Match the first navigation text with the node information of each menu node in the first menu node set, and determine a plurality of candidate menu nodes in the first menu node set according to the matching result.
[0146] S503: Perform menu node intent recognition on the first navigation text according to node information corresponding to each candidate menu node, and determine whether a target menu node corresponding to the first navigation text is determined in each candidate menu node.
[0147] The content of step S501 corresponds to the content of step S201 in the above embodiment, the content of step S502 corresponds to the content of step S202 in the above embodiment, and the content of step S503 corresponds to the content of step S203 in the above embodiment. The specific contents of steps S501-S503 can be found in Figure 2 The contents of the illustrated embodiments will not be repeated here.
[0148] In step S503, if the target menu node corresponding to the first navigation text is determined in each candidate menu node, step S504 is executed; if the target menu node corresponding to the first navigation text is not determined in each candidate menu node, step S505 is executed.
[0149] S504: Determine a target navigation path according to the target menu node, where the target navigation path includes a navigation path from the current menu node to the target menu node.
[0150] Optionally, the navigation path of each menu node is pre-set and stored. After the target menu node is determined, the navigation path from the current menu node to the target menu node, i.e., the target navigation path, can be calculated based on the preset navigation path of the target menu node and the navigation path of the current menu node.
[0151] Optionally, the hierarchical structure of the navigation menu and the parent-child relationship of each menu node are pre-set and stored. After the target menu node is determined, the navigation path from the current menu node to the target menu node, i.e., the target navigation path, can be automatically calculated based on the preset hierarchical structure of the navigation menu and the parent-child relationship of each menu node.
[0152] Optionally, when calculating the target navigation path, the minimization of the number of user clicks and / or shortcuts (such as direct voice commands or shortcut menu options) are considered to determine the optimal target navigation path to directly guide the user to the target menu node.
[0153] Optionally, the user's operating habits are determined based on the user's historical interaction information, and when determining the target navigation path, the user's operating habits are taken into consideration to generate a navigation path that is more familiar and satisfactory to the user.
[0154] Further, after determining the target navigation path, the menu navigation system will generate a reply message to the first navigation text according to the navigation path. The reply message may be a voice message or a text message, including a prompt for the user's next operation, and may clearly guide the user on how to go to the target menu node through voice input or text input or click operation. For example, the reply message may be "I have helped you locate the report loss service, please say 'go to report loss' or click here to continue".
[0155] Furthermore, when the user encounters difficulties during navigation and asks for help, the menu navigation system can provide corresponding support. For example, if the user says "I don't know how to operate", the menu navigation system can provide more detailed step-by-step instructions or directly contact customer support.
[0156] Based on the above, the menu navigation system not only ensures that users can efficiently reach the target service, but also improves the overall user experience through dynamic feedback and continuous optimization. This intelligent navigation method uses big data and machine learning technology to adapt to the behavior patterns of different users and continuously improves itself to ensure the best user support and service.
[0157] S505: If a target menu node corresponding to the first navigation text is not determined among the candidate menu nodes, generate prompt information, where the prompt information is used to prompt the user to input supplementary navigation information.
[0158] When the first navigation text content is too vague, the first navigation text content is seriously incomplete, there are multiple interpretations of the first navigation text, the first navigation text is input incorrectly, or the menu structure is updated, etc., the processor may not be able to determine the target menu node of the first navigation text from the candidate menu nodes. In this case, in order to improve the user experience and help the user complete the target operation, the menu navigation system will generate a prompt message to guide the user to enter supplementary navigation information so that the system can determine the user's target menu node based on the supplementary navigation information entered by the user.
[0159] For example, the prompt information may be "Your input is vague, please provide more specific navigation instructions, such as 'view user information' or 'browse order history'", "'Settings' may involve multiple aspects, please specify the specific content you want to set, such as 'network settings', 'account settings' or 'privacy settings', etc.", "It seems that your input is incorrect, please check the spelling or grammar, and re-enter your navigation instructions" and other information. Through the above prompt information, users can be guided to gradually clarify their own navigation needs, and finally input more easily identifiable supplementary navigation information.
[0160] Exemplary Devices
[0161] Corresponding to the above-mentioned menu navigation method, an embodiment of the present application also provides a menu navigation device. Figure 6 is a structural diagram of a menu navigation device provided in an embodiment of the present application, such as Figure 6 As shown, the menu navigation device provided in the embodiment of the present application includes:
[0162] The first unit 601 is used to obtain a first navigation text;
[0163] The second unit 602 is used to match the first navigation text with the node information of each menu node in the first menu node set, and determine a plurality of candidate menu nodes in the first menu node set according to the matching result;
[0164] The third unit 603 is used to perform menu node intention recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determine the target menu node corresponding to the first navigation text in each candidate menu node.
[0165] The menu navigation device provided by the present application, on the one hand, does not need to rely on predefined keywords or fixed command modes, supports users' natural language input, and reduces users' learning costs and usage difficulties; on the other hand, by performing preliminary and rapid matching of the first navigation text with the menu nodes in the first menu node set, multiple candidate menu nodes are identified, which greatly narrows the scope of subsequent menu node intention recognition. When performing menu node intention recognition, only the candidate menus determined by the preliminary match need to be considered, which not only greatly reduces the computational complexity and time consumption, improves navigation efficiency, but also improves the accuracy of menu navigation.
[0166] Optionally, the device further comprises:
[0167] A fourth unit is used to obtain historical interaction information of the first navigation text;
[0168] The third unit 603 can be specifically used for:
[0169] According to the node information corresponding to each candidate menu node and the historical interaction information, menu node intent recognition is performed on the first navigation text, and a target menu node corresponding to the first navigation text is determined in each candidate menu node.
[0170] Optionally, the third unit 603 may be specifically used for:
[0171] Generate a task instruction according to the node information corresponding to each candidate menu node, the historical interaction information and the first navigation text, the task instruction being used to instruct the object receiving the task instruction to perform menu node intent recognition on the first navigation text according to the node information corresponding to each candidate menu node and the historical interaction information, and determine the target menu node corresponding to the first navigation text in each candidate menu node;
[0172] The task instruction is input into a pre-trained large language model, and the pre-trained large language model executes the task instruction to determine a target menu node corresponding to the first navigation text among various candidate menu nodes.
[0173] Optionally, the second unit 602 may be specifically used for:
[0174] Matching the first navigation text with the node information of each menu node in the first menu node set, and determining the matching degree between the first navigation text and the node information of each menu node in the first menu node set;
[0175] According to the matching degree between the first navigation text and the node information of each menu node in the first menu node set, a preset number of candidate menu nodes are determined in the first menu node set.
[0176] Optionally, the device further comprises:
[0177] A fifth unit is used to generate prompt information when the target menu node corresponding to the first navigation text is not determined in each candidate menu node, and the prompt information is used to prompt the user to input supplementary navigation information;
[0178] and / or,
[0179] The sixth unit is used to determine a target navigation path according to the target menu node, wherein the target navigation path includes a navigation path from the current menu node to the target menu node.
[0180] Optionally, the matching includes keyword matching and / or semantic similarity matching, and the node information includes node name and / or node description information.
[0181] Optionally, the node information also includes entity information associated with the menu node and / or a navigation text example corresponding to the menu node, and the entity information includes at least one of an entity name, an entity alias, and an entity description information.
[0182] The menu navigation device provided in this embodiment belongs to the same application concept as the menu navigation method provided in the above embodiments of this application, can execute the menu navigation method provided in any of the above embodiments of this application, and has the corresponding functional modules and beneficial effects of executing the menu navigation method. For technical details not fully described in this embodiment, please refer to the specific processing content of the menu navigation method provided in the above embodiments of this application, and will not be repeated here.
[0183] The functions implemented by the above first unit 601, the second unit 602 and the third unit 603 may be implemented by the same or different processors, respectively, and the embodiment of the present application is not limited thereto.
[0184] It should be understood that the units in the above devices can be implemented in the form of a processor calling software. For example, the device includes a processor, the processor is connected to a memory, and instructions are stored in the memory. The processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of each unit of the device, wherein the processor can be a general-purpose processor, such as a CPU or a microprocessor, etc., and the memory can be a memory in the device or a memory outside the device. Alternatively, the units in the device can be implemented in the form of hardware circuits, and the functions of some or all units can be realized by designing the hardware circuits. The hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the above units are realized by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit can be implemented by PLD, taking FPGA as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by the configuration file, so as to realize the functions of some or all of the above units. All units of the above devices can be implemented in the form of a processor calling software, or in the form of hardware circuits, or in part by a processor calling software, and the remaining part is implemented in the form of hardware circuits.
[0185] In an embodiment of the present application, a processor is a circuit with the ability to process signals. In one implementation, the processor may be a circuit with the ability to read and run instructions, such as a CPU, a microprocessor, a GPU, or a DSP; in another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, DPU, etc.
[0186] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.
[0187] In addition, all or part of the units in the above device can be integrated together, or can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a SOC. The SOC may include at least one processor for implementing any of the above methods or implementing the functions of each unit of the device. The type of the at least one processor may be different, for example, including a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.
[0188] An embodiment of the present application further provides a control device, which includes a processor and an interface circuit. The processor in the control device is connected to an instruction input component through the interface circuit of the control device.
[0189] The instruction acquisition module 110 can be a voice input component or a text input component. The voice input component can be any functional component that can support the user to perform voice input, such as a microphone, a pickup, etc., and its specific voice collection and voice conversion processing process can refer to the implementation process of conventional technology. The text input component can be any functional component that can support the user to perform text input, such as a touch screen, a keyboard light, and its specific text collection processing process can refer to the implementation process of conventional technology.
[0190] The command input component specifically refers to a functional component that can sense the user's command input operation, which can be a voice input component or a text input component. The voice input component can be any functional component that can support the user's voice input, such as a microphone, a pickup, etc.; the text input component can be any functional component that can support the user's text input, such as a touch screen, a keyboard light, etc.
[0191] The above-mentioned interface circuit can be any interface circuit that can realize the data communication function, for example, it can be a USB interface circuit, a Type-C interface circuit, a serial port circuit, a PCIE circuit, etc.
[0192] The processor in the control device is also a circuit with signal processing capability, which performs menu navigation according to the navigation voice or navigation text input by the instruction input component by executing any one of the menu navigation methods described in the above embodiments. The specific implementation of the processor can refer to the above processor implementation, and the embodiments of this application are not strictly limited.
[0193] When the control device is applied to an intelligent device, the command input component of the control device may be the voice input component or text input component of the intelligent device. Meanwhile, the processor of the control device may be a CPU or GPU of the intelligent device, and the interface circuit of the control device may be an interface circuit between the command input component of the intelligent device and a processor such as a CPU or GPU.
[0194] Exemplary Electronic Devices
[0195] Another embodiment of the present application also provides an electronic device, see Figure 7 As shown, the device includes:
[0196] Memory 200 and processor 210;
[0197] The memory 200 is connected to the processor 210 and is used to store programs;
[0198] The processor 210 is used to implement the menu navigation method disclosed in any of the above embodiments by running the program stored in the memory 200.
[0199] Specifically, the electronic device may further include: a bus, a communication interface 220 , an input device 230 and an output device 240 .
[0200] The processor 210, the memory 200, the communication interface 220, the input device 230 and the output device 240 are connected to each other via a bus.
[0201] A bus may include a pathway that transfers information between components of a computer system.
[0202] The processor 210 may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the scheme of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.
[0203] The processor 210 may include a main processor, and may also include a baseband chip, a modem, and the like.
[0204] The memory 200 stores a program for executing the technical solution of the present invention, and may also store an operating system and other key services. Specifically, the program may include a program code, and the program code includes a computer operation instruction. More specifically, the memory 200 may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk storage, a flash, and the like.
[0205] The input device 230 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.
[0206] Output device 240 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.
[0207] The communication interface 220 may include any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.
[0208] The processor 210 executes the program stored in the memory 200 and calls other devices, which can be used to implement each step of any menu navigation method provided in the above embodiments of the present application.
[0209] An embodiment of the present application also proposes a chip, which includes a processor and a data interface. The processor reads and runs a program stored in a memory through the data interface to execute the menu navigation method introduced in any of the above embodiments. The specific processing process and its beneficial effects can be found in the introduction to the embodiment of the above menu navigation method.
[0210] Exemplary computer program products and storage media
[0211] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the menu navigation method according to various embodiments of the present application described in any of the above embodiments of this specification.
[0212] The computer program product may be written in any combination of one or more programming languages to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.
[0213] In addition, the embodiment of the present application may also be a storage medium, on which a computer program is stored, and the computer program is executed by a processor to execute the steps of the menu navigation method according to various embodiments of the present application described in any of the above embodiments of this specification, and specifically the following steps may be implemented:
[0214] Step S201, obtaining a first navigation text;
[0215] Step S202: matching the first navigation text with the node information of each menu node in the first menu node set, and determining a plurality of candidate menu nodes in the first menu node set according to the matching result;
[0216] Step S203: performing menu node intention recognition on the first navigation text according to the node information corresponding to each candidate menu node, and determining a target menu node corresponding to the first navigation text in each candidate menu node.
[0217] For the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0218] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0219] The steps in the methods of each embodiment of the present application can be adjusted in order, combined and deleted according to actual needs, and the technical features recorded in each embodiment can be replaced or combined.
[0220] The modules and sub-modules in the devices and terminals of the various embodiments of the present application can be combined, divided and deleted according to actual needs.
[0221] In the several embodiments provided in the present application, it should be understood that the disclosed terminals, devices and methods can be implemented in other ways. For example, the terminal embodiments described above are only schematic, for example, the division of modules or submodules is only a logical function division, and there may be other division methods in actual implementation, for example, multiple submodules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.
[0222] The modules or submodules described as separate components may or may not be physically separated, and the components of the modules or submodules may or may not be physical modules or submodules, that is, they may be located in one place, or they may be distributed on multiple network modules or submodules. Some or all of the modules or submodules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0223] In addition, each functional module or submodule in each embodiment of the present application may be integrated into one processing module, or each module or submodule may exist physically separately, or two or more modules or submodules may be integrated into one module. The above-mentioned integrated modules or submodules may be implemented in the form of hardware or in the form of software functional modules or submodules.
[0224] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0225] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly by hardware, software units executed by a processor, or a combination of the two. The software units may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
[0226] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.
[0227] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A menu navigation method, characterized in that: include: Acquire a first navigation text and first information representing user preferences and behavior habits; Matching the first navigation text with node information of each menu node in the first menu node set, and determining a plurality of candidate menu nodes in the first menu node set according to the matching result, wherein the matching includes keyword matching and / or semantic similarity matching, the node information includes entity information associated with the menu node and a navigation text example corresponding to the menu node, and the entity information includes an entity name, an entity alias, and entity description information; Generate a task instruction according to the node information corresponding to each candidate menu node, the first information and the first navigation text, the task instruction being used to instruct the object receiving the task instruction to perform menu node intention recognition on the first navigation text according to the node information corresponding to each candidate menu node and the user preferences and behavior habits represented by the first information, and determine the target menu node corresponding to the first navigation text in each candidate menu node; The task instruction is input into a pre-trained large language model, and the pre-trained large language model executes the task instruction to determine a target menu node corresponding to the first navigation text from among the candidate menu nodes.
2. The method according to claim 1, characterized in that The matching of the first navigation text with the node information of each menu node in the first menu node set, and determining a plurality of candidate menu nodes in the first menu node set according to the matching result, includes: Matching the first navigation text with the node information of each menu node in the first menu node set, and determining the matching degree between the first navigation text and the node information of each menu node in the first menu node set; According to the matching degree between the first navigation text and the node information of each menu node in the first menu node set, a preset number of candidate menu nodes are determined in the first menu node set.
3. The method according to claim 1, characterized in that: The method further comprises: If the target menu node corresponding to the first navigation text is not determined in each candidate menu node, generating prompt information, wherein the prompt information is used to prompt the user to input supplementary navigation information; and / or, A target navigation path is determined according to the target menu node, wherein the target navigation path includes a navigation path from a current menu node to the target menu node.
4. The method according to any one of claims 1 to 3, characterized in that The node information includes a node name and / or node description information.
5. A menu navigation device, characterized in that: include: A first unit is used to obtain a first navigation text and first information representing user preferences and behavior habits; a second unit, configured to match the first navigation text with node information of each menu node in the first menu node set, and determine a plurality of candidate menu nodes in the first menu node set according to the matching result, wherein the matching includes keyword matching and / or semantic similarity matching, the node information includes entity information associated with the menu node and a navigation text example corresponding to the menu node, and the entity information includes an entity name, an entity alias, and entity description information; A third unit is used to generate a task instruction according to the node information corresponding to each candidate menu node, the first information and the first navigation text, wherein the task instruction is used to instruct the object receiving the task instruction to perform menu node intention recognition on the first navigation text according to the node information corresponding to each candidate menu node and the user preference and behavior habit represented by the first information, and determine the target menu node corresponding to the first navigation text in each candidate menu node; The fourth unit is used to input the task instruction into a pre-trained large language model, and the pre-trained large language model executes the task instruction to determine the target menu node corresponding to the first navigation text from each candidate menu node.
6. An electronic device, characterized in that: including memory and processor; The memory is connected to the processor and is used to store programs; The processor is used to implement the menu navigation method according to any one of claims 1 to 4 by running the program in the memory.
7. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the menu navigation method according to any one of claims 1 to 4 is implemented.
8. A computer program product, characterized in that The method comprises computer program instructions, which, when executed by a processor, enable the processor to implement the menu navigation method according to any one of claims 1 to 4.
