Information processing method and device

Abstract

The invention discloses an information processing method and device, and belongs to the technical field of information processing. The method comprises the following steps: receiving first request information input by a user; performing intention analysis on the first request information to obtain first intention information and first scene information; starting the first application according to the first scene information; the first intent information is processed in the first application.

Description

Technical Field

[0001] The present application belongs to the field of information processing technology, and specifically relates to an information processing method and device. Background Art

[0002] With the rapid development of artificial intelligence, intelligent voice assistants and intelligent agents are becoming increasingly popular. Current intelligent agent skills mainly rely on user-initiated request information and identify user intent by analyzing the semantics of the request information.

[0003] Currently, when users don't explicitly specify an application, determining and launching it takes a long time, resulting in a poor user experience. Furthermore, the system has poor adaptability to diverse user needs. For example, when a user request is ambiguous, it's difficult to accurately identify the user's intent.

[0004] Therefore, the current response efficiency to the user's request information is low. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide an information processing method and apparatus that can solve the current problem of low efficiency in responding to request information.

[0006] In a first aspect, an embodiment of the present application provides an information processing method, the method comprising:

[0007] receiving first request information input by a user;

[0008] Performing intent analysis on the first request information to obtain first intent information and first scenario information;

[0009] Starting a first application according to the first scenario information;

[0010] Process the first intent information in the first application.

[0011] In a second aspect, an embodiment of the present application provides an information processing device, the device comprising:

[0012] A receiving module, configured to receive first request information input by a user;

[0013] an analysis module, configured to perform intent analysis on the first request information to obtain first intent information and first scenario information;

[0014] A starting module, configured to start a first application according to the first scenario information;

[0015] A processing module is used to process the first intent information in the first application.

[0016] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.

[0017] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0018] In a fifth aspect, an embodiment of the present application provides a chip, which includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the method described in the first aspect.

[0019] In a sixth aspect, an embodiment of the present application provides a computer program product, which is stored in a storage medium and executed by at least one processor to implement the method described in the first aspect.

[0020] In an embodiment of the present application, intent analysis is performed on a first request message input by a user to obtain first intent information and first scenario information. Based on the first scenario information, a first application is launched. By directly analyzing the scenario information, compatible applications are quickly selected, significantly reducing the time required to launch the application. Processing the first intent information in the first application accurately executes the first intent information in the first request message, improving the efficiency of responding to the first request message. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flowchart of an information processing method provided by an embodiment of the present application;

[0022] Figure 2 is a flowchart of another information processing method provided by an embodiment of the present application;

[0023] Figure 3 This is a flowchart of another information processing method provided by an embodiment of the present application;

[0024] Figure 4 is a structural diagram of an information processing device provided in an embodiment of the present application;

[0025] Figure 5 This is one of the hardware structure diagrams of the electronic device according to the embodiment of the present application;

[0026] Figure 6 This is the second hardware structure diagram of the electronic device according to the embodiment of the present application. DETAILED DESCRIPTION

[0027] The following will be combined with the accompanying drawings of the embodiments of the present application to clearly describe the technical solutions of the embodiments of the present application. Obviously, the embodiments described are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of this application.

[0028] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0029] In response to the problems arising in the related art, the embodiments of the present application provide an information processing method and device, which can solve the problem of low efficiency in responding to request information in the related art.

[0030] The information processing method provided in the embodiments of the present application is described in detail below through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0031] Figure 1 A flowchart of an information processing method provided in an embodiment of the present application.

[0032] like Figure 1 As shown, the information processing method may include steps 110 to 140, and the method is applied to an information processing device, as shown below:

[0033] Step 110: receiving first request information input by the user;

[0034] First request information: This is the raw data a user inputs into the intelligent system of an electronic device, in the form of voice, text, etc. For example, in daily life, a user might say "Order a cup of coffee to go" to an intelligent voice assistant, or type this into the input box of a smart device. This is first request information.

[0035] Step 120: performing intent analysis on the first request information to obtain first intent information and first scenario information;

[0036] First Intent Information: The intelligent system deeply analyzes the first request information with complex natural language processing algorithms to dig out the most core demands of the user. Just like the sentence "Order a coffee for delivery", after removing redundant parts such as modifiers and modal particles, the core action "Buy coffee" is extracted, which is the first intent information. It represents the specific behavior that the user hopes to achieve.

[0037] First Scene Information: It is a classification of the environment where the user's request is located. By comprehensively analyzing keywords, context, etc. in the request information, its belonging field is judged. "Order a coffee for delivery" involves the behavior of commodity purchase, so it is classified as the "Shopping" scene. Common scenes also include travel, entertainment, learning, etc., and different scenes correspond to different types of application programs.

[0038] First, perform lexical analysis on the first request information, split the sentence into individual words, and mark their词性, for example, "Order" is a verb and "coffee" is a noun. Then perform syntactic analysis to construct the syntactic structure tree of the sentence and clarify the relationship between words. Finally, through semantic analysis, combined with the knowledge base and context, the first intent information and the first scene information are extracted. For example, through the semantic analysis of "Order a coffee for delivery", it is determined that the user's intent is to buy coffee and the scene belongs to shopping.

[0039] Step 130, according to the first scene information, start the first application;

[0040] First Application: According to the first scene information, the system filters out the matching application program from the pre-set application list. When it is determined that the scene is "Shopping", combined with the user's needs, the takeaway application can meet the needs, so it is selected as the first application. This application will be responsible for executing the specific intent of the user.

[0041] The application scenario mapping table records the corresponding relationships between different scenes and adaptable applications. When the first scene information "Shopping" is obtained, the matching application can be searched in the mapping table. Since the takeaway application is closely related to the shopping scene and can meet the need to buy coffee, the system sends an instruction to the operating system to start this takeaway application. This process involves the process management of the operating system and the loading mechanism of the application program.

[0042] Step 140, process the first intent information in the first application.

[0043] Package the first intent information "Buy coffee" into a specific data format and pass it to the started takeaway application through the application programming interface. After receiving the intent information, the takeaway application performs operations according to its pre-set business logic. For example, search for coffee-related products in the database of the takeaway application, display the product list for the user to choose, and finally complete the order placement process.

[0044] By directly analyzing scenario information, it quickly selects compatible apps, significantly reducing the time it takes to find them. It also simplifies the process by automatically launching apps for users. In reality, user expressions can be imprecise. For example, "Give me a cup of coffee, delivered to my door," differs from the standard expression "Order a cup of coffee for takeout." However, with natural language processing technology, the system can still accurately identify the intent as "buy coffee" and the scenario as "shopping," launching the food delivery app and accurately fulfilling user needs. This improves the overall efficiency and accuracy of the system and provides users with a smoother user experience.

[0045] When a user says "Order a cup of coffee for takeout," the intelligent system's microphone receives the voice signal and converts it into digital data. Using natural language processing technology, this data is analyzed lexically, syntactically, and semantically, deriving the first intent information, "Buy coffee," and the first scenario information, "Shopping." Based on this scenario information, the food delivery app is matched in the application scenario mapping table, and a launch command is sent to the operating system. After the app starts, the "Buy coffee" intent information is passed to the food delivery app via an API. The food delivery app then searches its product database for coffee products, displaying a variety of coffee options for the user to choose from. The user selects their favorite coffee and places their order. The entire process is efficient and smooth, precisely meeting user needs.

[0046] The following combination Figure 2 To explain, such as Figure 2 As shown, first, the first request information is received; secondly, the first request information is analyzed and split into different intent information, scenario information and application information, and multiple groups of information may be obtained, such as the first group [first intent information, first scenario information, first application information] and the second group [second intent information, second scenario information, second application information].

[0047] Then, the corresponding applications are started according to the application information obtained by splitting. For example, the first application is started according to the first application information and associated with the first intent information; the second application is started according to the second application information and associated with the second intent information.

[0048] Then, for each launched application, the corresponding action execution agent is instantiated. For example, for the first application, action execution agent 1 is instantiated and associated with the first intent information; for the second application, action execution agent 2 is instantiated and associated with the second intent information.

[0049] Finally, the action execution agent generates a specific action. Action execution agent 1 generates the first action, category "Other," and associates the first intent information, agent 1, and the first action information. Action execution agent 2 generates the first action, category "Other," and associates the second intent information, agent 2, and the second action information.

[0050] In a possible embodiment, step 120 may specifically include the following steps:

[0051] Inputting the first request information into a first intent splitting model, and outputting the first intent information and the first scenario information;

[0052] The first intent splitting model is trained based on multiple groups of first training data, and each group of the first training data includes: first sample request information, first label intent information and first label scene information associated with the first sample request information.

[0053] The first intent splitting model is a model built based on machine learning or deep learning, specifically used to process natural language request information, and accurately split the original request information entered by the user into corresponding intent information and scenario information.

[0054] First training data: To ensure the first intent splitting model can accurately perform its tasks, it requires a large amount of labeled data for training. This data is called the first training data. It consists of multiple data groups, each of which consists of the first sample request information, closely related first label intent information, and first label scenario information. This labeled data acts as a learning sample for the model, helping it continuously optimize its processing logic.

[0055] First sample request information: As part of the first training data, it is a representative example of user requests collected from real-world scenarios or simulated. For example, "Order a pizza delivery" or "Book a flight to Beijing tomorrow" are examples. These sample request information allows the model to learn user intent and context characteristics under different expression methods.

[0056] First-label intent information: For each first sample request, the user's core desire is manually annotated. For example, for the sample request "Order a pizza for delivery," its first-label intent information might be "Buy a pizza." This clarifies the specific behavior the user wants to complete and serves as an important reference for model learning.

[0057] First-label scenario information: This is also a manually annotated representation of the first sample request information, used to indicate the scenario category of the request. For example, the first-label scenario information for "Order a pizza delivery" is "Shopping." This helps the model understand the broad scenario domains to which different requests belong, facilitating subsequent accurate classification.

[0058] During training, the model inputs the first sample request information. Using its complex internal neural network structure and algorithms, the model attempts to output the corresponding intent and scenario information. The model's output is then compared with the pre-labeled first-label intent information and first-label scenario information, and the error between the two is calculated. Based on this error, the model automatically adjusts its internal parameter weights using optimization algorithms such as backpropagation, continuously reducing the error and gradually bringing the model output closer to the labeled result. Through repeated training on large amounts of data, the model gradually learns the inherent connections and patterns between request information, intent, and scenario.

[0059] When a user enters a first request, it is fed into the trained first intent segmentation model. The model analyzes and processes the input request based on the patterns and features learned during training. It understands the request from multiple perspectives, including vocabulary, grammatical structure, and semantics. Through complex internal computational logic, it ultimately outputs the corresponding first intent and first scenario. For example, if a user enters "order a cup of coffee to go," the model, based on the knowledge acquired during training, identifies "buy coffee" as the first intent and "shopping" as the first scenario.

[0060] By training the model with a large amount of annotated first training data, it can learn a rich variety of request expressions and their corresponding intents and scenarios. Compared to traditional intent recognition methods that rely on simple rule matching or manual settings, this model can handle more complex and varied user requests, greatly improving the accuracy of intent and scenario recognition. For example, even for some vague or personalized requests, the model can accurately determine the intent and scenario based on its training experience.

[0061] As the primary training data continues to expand and update, the primary intent splitting model can continuously learn new request patterns and scenario categories. This enables the entire intelligent system to adapt to new application scenarios and changes in user needs. For example, when new shopping patterns or emerging service areas emerge, simply adding the relevant new data to the training dataset and retraining the model can quickly adapt to these changes and accurately identify user intent in new scenarios without requiring large-scale modifications to the underlying system logic, thereby enhancing the system's adaptability and scalability.

[0062] In a possible embodiment, step 130 may specifically include the following steps:

[0063] Determining, based on a scenario database, at least one application information matching the first scenario information, the scenario database comprising a plurality of intent information and at least one application information associated with each of the intent information;

[0064] determining first application information from the at least one application information;

[0065] Start the first application indicated by the first application information.

[0066] The scenario database is a collection of information specifically used to associate different intents with corresponding applications. Like an intelligent system's "information warehouse," it covers the correspondence between user intent and applicable applications in a variety of common scenarios, providing data support for the system when processing user requests. For example, in a shopping scenario, it records information about various e-commerce and food delivery applications corresponding to different items purchased.

[0067] Application information: This refers to descriptive data about an application, including its name, features, applicable scenarios, and the instructions required to launch the application. This information allows the system to accurately identify and invoke the corresponding application. For example, for a food delivery app, its application information might include the app name "XX Takeaway," its primary function being to provide food delivery services, its applicable scenario being food purchases during shopping, and the operating system instructions required to launch the app.

[0068] First application information: application information that best meets the user's current needs, determined through specific screening rules from multiple application information matched according to the first scenario information.

[0069] The first scenario information obtained from the first intent splitting model is compared with the data in the scenario database. A large amount of different intent information and its associated application information are pre-stored in the scenario database. The system uses a search algorithm to find intent information that matches the first scenario information, and then obtains at least one application information associated with it. For example, if the first scenario information is "shopping", the system will find all intent information related to shopping scenarios in the scenario database, such as "buying food" and "buying clothes", and then obtain the application information associated with each of these intent information, such as food delivery applications, clothing e-commerce applications, etc.

[0070] After acquiring at least one piece of app information, the system will determine the first piece of app information from this information based on a specific filtering strategy. This filtering strategy can be based on various factors, such as app usage frequency, app quality of service rating, and the app's relevance to the user's current location. By comprehensively considering these factors, the system selects the app information that best meets the user's current needs as the first piece of app information.

[0071] After obtaining the first application information, the application launch instruction information contained therein is extracted. This instruction information is a specific code that interacts with the operating system and is used to trigger the launch of the corresponding application program. For example, in a mobile operating system, the corresponding API interface is called according to the instruction information to launch the first application indicated by the first application information, such as opening the "XX Takeaway" application.

[0072] By building a scenario database, the system can quickly and accurately find applications that match user requests based on a wealth of pre-stored data. This avoids blind searches or the random launch of irrelevant applications, significantly improving the accuracy of application selection. For example, when a user makes a travel-related request, the system can accurately locate the travel booking app among a large number of available applications, rather than mistakenly selecting other unrelated ones.

[0073] Determining the first app from multiple matching apps based on different screening strategies can adapt to the personalized preferences and diverse needs of different users. Different users have different app usage habits and factors they care about, and this step comprehensively considers these factors to provide each user with app options that better suit their needs. For example, a user who prioritizes price may be prioritized for apps with promotions, while a user who prioritizes delivery speed may be recommended apps with high-quality delivery services.

[0074] This allows the entire process from scene information matching to app launch to be completed in a fraction of the time. Compared to manually searching for apps, this significantly reduces the time users have to wait for apps to launch, improves the overall system response efficiency, and provides users with a smoother user experience.

[0075] The above-mentioned step of determining the first application information from the at least one application information may specifically include the following steps:

[0076] displaying the at least one piece of application information, wherein the at least one piece of application information is arranged according to the recommended order information;

[0077] In response to a selection input of the first application information among the at least one application information, the first application information is determined.

[0078] Recommendation order information: This is the data used to rank multiple app information. It's generated based on a variety of factors, aiming to prioritize the apps most likely to meet user needs. These factors include, but are not limited to, app usage frequency, user reviews, and relevance to the current scenario and user history. For example, if a user frequently uses a food delivery platform to purchase lunch, the platform's app information may be ranked higher in the recommendation order when it relates to food purchases in a shopping scenario.

[0079] Selected input: When a user interacts with an intelligent system, it refers to a clear selection operation made for at least one application information presented by the system, which can be achieved through various means, such as clicking on the icon or name of a certain application in a graphical interface, or saying the name of a specific application in voice interaction.

[0080] After obtaining at least one application information that matches the first scenario information, these application information are sorted according to the recommended order information. The generation of the recommended order information involves complex algorithms and comprehensively considers factors in multiple dimensions. For example, for the intention of buying coffee in a shopping scenario, the system will analyze the applications used by the user when purchasing coffee in the past, the user reviews of various coffee delivery applications, and the applications with fast delivery speed around the current user's location, etc. Then, these factors are calculated according to the set weights to obtain the comprehensive scores of each application information, and they are sorted according to the scores. After the sorting is completed, the system presents these application information to the user in a specific interface form.

[0081] The user views the list of application information presented by the system and, according to their own needs and preferences, performs a selected input operation on a certain application information. When the selected input of the user is detected, the application information corresponding to this operation is identified and determined as the first application information. For example, when the user clicks on the icon of the "XX Coffee Delivery" application on the mobile phone screen, after the system receives this click event, it determines that the user has selected the information of the "XX Coffee Delivery" application through information such as the coordinate position, and thus determines it as the first application information.

[0082] By presenting the application information in the recommended order, the user does not need to blindly search among numerous applications and can quickly focus on the applications that meet their own needs.

[0083] In a possible embodiment, in step 120, it may specifically include the following steps:

[0084] Perform intention analysis on the first request information to obtain the first application information, the first intention information, and the first scenario information;

[0085] This embodiment may further include the following steps:

[0086] Start the first application indicated by the first application information.

[0087] Conduct a comprehensive analysis of the first request information. First, perform lexical analysis, break the sentence into individual words, and mark their词性, such as "order" is a verb and "coffee" is a noun. Then, carry out syntactic analysis to construct the syntactic structure tree of the sentence and clarify the modifying, subject-predicate-object, etc. relationships between the words. Then, enter the semantic analysis stage, and in combination with a large semantic knowledge base and the user's historical interaction data, deeply understand the meaning of the user's request.

[0088] During the semantic analysis process, the system not only extracts the user's primary intent, such as "buy coffee," but also determines the primary scenario information to which the request belongs, such as "shopping." More importantly, based on these analysis results and the system's pre-built application association database, the primary application information is directly derived. This database stores the correspondence between various types of intent information, scenario information, and application information. For example, in the "shopping" scenario, the "buy coffee" intent corresponds to application information such as "XX Takeout" and "XX Coffee Official App." The system uses matching rules to filter out the primary application information that best meets the user's needs.

[0089] After obtaining the first application information, the instructions for launching the application are extracted. These instructions vary depending on the operating system and device type. In mobile operating systems, they usually call a specific API interface and pass the application's unique identifier or launch parameters.

[0090] After receiving the startup instruction, the operating system finds the corresponding application file based on the system's process management mechanism, loads the resources required by the application, including code, images, configuration files, etc., and finally starts the application and presents it to the user.

[0091] Traditional approaches require first determining intent and scenario, then matching application information from a wide range of applications based on the scenario. This embodiment, however, directly analyzes the request information to derive the first application information, eliminating the intermediate matching steps and significantly shortening the time from user request to application launch. For example, if a user requests "book a flight to Shanghai tomorrow," the system can quickly determine the first application information and directly launch it, eliminating the need for the user to wait through multiple steps, thus improving overall interaction efficiency.

[0092] During the analysis process, the system comprehensively considers intent, scenario, and its internal database of connections to more accurately identify the top application that best matches the user's needs. Compared to matching applications based solely on scenario, incorporating intent information can further narrow the application selection range, avoiding the selection of applications that are irrelevant or fail to meet the user's specific needs. For example, in the "shopping" scenario, for the "purchase office supplies" intent, it can accurately find specialized office supply procurement applications rather than general e-commerce applications, improving the accuracy and specificity of application selection.

[0093] The simplified process and improved accuracy allow users to access applications that meet their needs more quickly, reducing operation steps and waiting time, thereby significantly improving the user experience. Users will experience a smoother and more convenient experience when using intelligent systems, which increases their favorability and willingness to use the system.

[0094] In a possible embodiment, step 120 may specifically include the following steps:

[0095] Inputting the first request information into a second intent splitting model, and outputting the first application information, the first intent information, and the first scenario information;

[0096] The second intent splitting model is trained based on multiple sets of second training data, each set of the second training data including: second sample request information, label application information associated with the second sample request information, second label intent information, and second label scenario information;

[0097] This embodiment may also include the following steps:

[0098] Start the first application indicated by the first application information.

[0099] The second intent splitting model is a specialized model built on machine learning or deep learning algorithms. Its function is to analyze and process the user's first request information and directly output the first application information, the first intent information, and the first scenario information. By learning from a large amount of labeled training data, it grasps the inherent connections and pattern recognition capabilities between request information and applications, intents, and scenarios.

[0100] Secondary training data: A series of data sets prepared for training the secondary intent segmentation model. Each data set contains secondary sample request information, along with closely associated label application information, secondary label intent information, and secondary label scenario information. This data serves as "teaching material" for the model to learn from. By repeatedly learning from it, the model optimizes its parameters and improves the accuracy of its analysis of various request information.

[0101] Second sample request information: Representative user request examples collected from actual application scenarios or simulated and used to train the second intent segmentation model. These requests cover a variety of expressions and language habits to help the model learn diverse user expressions. Examples include "Book me a hotel in Beijing tomorrow" and "Find me a nearby barber shop."

[0102] Tagged application information: For each second sample request, manually annotated information is provided about the corresponding application that can meet the user's needs. This information includes key data such as the application name, the instructions required to launch the application, and the application's storage path on the device. It serves as an important reference for the model to learn how to directly link the request information to the application. For example, for the second sample request for "Order a pizza delivery," the tagged application information may point to the "XX Takeaway" application and include the application's launch code and related configuration information in the mobile phone system.

[0103] Second label intent information: This is also manually annotated content for the second sample request information, representing the user's core desire in the request. For example, for the request "Book a train ticket to Guangzhou the day after tomorrow," the second label intent information is "Purchase train ticket." This helps the model understand the behavioral purpose behind the user's request.

[0104] Second label scenario information: This labels the environment or domain of the second sample request. For example, "Book a train ticket to Guangzhou the day after tomorrow" falls under the "travel" scenario. By learning these scenario labels, the model can quickly determine the scenario category of the request when processing the actual request, thereby outputting relevant information more accurately.

[0105] The second intent segmentation model is trained using multiple sets of second training data. During the training process, the second sample request information is input into the model. The model uses its complex internal neural network structure and algorithms to perform lexical, syntactic, and semantic analysis on the input request information.

[0106] The model attempts to output the corresponding first application information, first intent information, and first scenario information. The model's output is then compared with the pre-labeled application information, second intent information, and second scenario information, using a loss function to calculate the error between the two. Based on this error, the model automatically adjusts its internal parameter weights using a backpropagation algorithm, continuously reducing the error and gradually bringing the model output closer to the labeled result. Through repeated training on large amounts of data, the model gradually learns the mapping relationships and patterns between request information and applications, intents, and scenarios.

[0107] When a user enters a first request, it is fed into the trained second intent segmentation model. Based on the patterns and features learned during training, the model conducts a comprehensive analysis of the input request. It understands the meaning and function of each word at the lexical level, grasps the sentence structure at the syntactic level, and then, at the semantic level, combines the knowledge base and historical user data to deeply understand the true intent of the request.

[0108] The model ultimately outputs the corresponding first application information, first intent information, and first scenario information. For example, for a user input, "Find a nearby shop that can repair a computer," the model, based on the knowledge acquired during training, identifies the first intent information as "computer repair" and the first scenario information as "life services." Based on the internal association database and the learned mapping relationship, it outputs the first application information, such as "XX computer repair service platform."

[0109] The results are directly output through the second intent splitting model. Since the model has been trained with a large amount of targeted data, it can more accurately capture the key elements in the request information and directly link them to the most appropriate application, greatly improving the efficiency from request to application startup.

[0110] Because the second training data contains a rich variety of second sample request information, the model learns intent, scenario, and application associations across various expressions during training. This enables the model to accurately analyze relevant information even when faced with new, even vaguely expressed or personalized requests, leveraging its strong generalization capabilities.

[0111] By outputting information quickly and accurately, users can find the applications that meet their needs in the shortest possible time, reducing wait times and operational complexity. Whether searching for daily life services or specific applications for work or study, users can appreciate the efficiency and convenience of the intelligent system, greatly improving their satisfaction and user stickiness.

[0112] like Figure 3 As shown, the first request information input by the user is obtained. Then, the first request information is input into the intent splitting model. The model processes the intent splitting description, the scenario, and the relationship between the scenario and the app, and outputs the splitting result, which includes the query, scenario, and app information.

[0113] Next, determine whether there is an APP name:

[0114] If there is an APP name: further determine whether the APP exists in the scene library. If it exists, return a single APP package name; if not, return empty.

[0115] If there is no APP name: determine whether the scene exists in the scene library. If it does, return the APP list package name; if not, return empty.

[0116] Then, determine the APP installation status:

[0117] When returning a single APP package name: Determine whether the single APP is installed. If it is installed, directly launch the APP; if not, recommend installation and return failure.

[0118] When the app list package name is returned: the app list is matched sequentially to determine whether it is installed. If an app is already installed, it is launched; if none of the apps in the list are installed, it is recommended for installation. The entire process involves splitting the user request information and determining the app's related information to determine and launch the app. If the app is not installed, it is recommended for installation.

[0119] In an embodiment of the present application, intent analysis is performed on a first request message input by a user to obtain first intent information and first scenario information. Based on the first scenario information, a first application is launched. By directly analyzing the scenario information, compatible applications are quickly selected, significantly reducing the time required to launch the application. Processing the first intent information in the first application accurately executes the first intent information in the first request message, improving the efficiency of responding to the first request message.

[0120] The information processing method provided in the embodiment of the present application can be executed by an information processing device. In the embodiment of the present application, the information processing device provided in the embodiment of the present application is described by taking the information processing device executing the information processing method as an example.

[0121] Figure 4 4 is a block diagram of an information processing device provided in an embodiment of the present application. The device 400 includes:

[0122] Receiving module 410, configured to receive first request information input by a user;

[0123] An analysis module 420 is configured to perform intent analysis on the first request information to obtain first intent information and first scenario information;

[0124] A starting module 430, configured to start a first application according to the first scenario information;

[0125] The processing module 440 is configured to process the first intent information in the first application.

[0126] In a possible embodiment, the analysis module 420 is specifically configured to:

[0127] Inputting the first request information into a first intent splitting model, and outputting the first intent information and the first scenario information;

[0128] The first intent splitting model is trained based on multiple groups of first training data, and each group of the first training data includes: first sample request information, first label intent information and first label scene information associated with the first sample request information.

[0129] In a possible embodiment, the starting module 430 includes:

[0130] a determination module, configured to determine, based on a scenario database, at least one application information matching the first scenario information, the scenario database comprising a plurality of intent information and at least one application information associated with each of the intent information;

[0131] The determining module is further configured to determine the first application information from the at least one application information;

[0132] The starting module 430 is further configured to start the first application indicated by the first application information.

[0133] In a possible embodiment, the determination module is specifically configured to:

[0134] displaying the at least one piece of application information, wherein the at least one piece of application information is arranged according to the recommended order information;

[0135] In response to a selection input of the first application information among the at least one application information, the first application information is determined.

[0136] In a possible embodiment, the analysis module 420 is specifically configured to:

[0137] Inputting the first request information into a second intent splitting model, and outputting the first application information, the first intent information, and the first scenario information;

[0138] The second intent splitting model is trained based on multiple sets of second training data, each set of the second training data including: second sample request information, label application information associated with the second sample request information, second label intent information, and second label scenario information;

[0139] The starting module is further configured to start the first application indicated by the first application information.

[0140] In an embodiment of the present application, intent analysis is performed on a first request message input by a user to obtain first intent information and first scenario information. Based on the first scenario information, a first application is launched. By directly analyzing the scenario information, compatible applications are quickly selected, significantly reducing the time required to launch the application. Processing the first intent information in the first application accurately executes the first intent information in the first request message, improving the efficiency of responding to the first request message.

[0141] The information processing device in the embodiment of the present application can be an electronic device or a component in an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or a device other than a terminal. For example, the electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a mobile Internet device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc. It can also be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., and the embodiment of the present application does not specifically limit it.

[0142] The information processing device of the embodiment of the present application may be a device having an action system. The action system may be an Android action system, an iOS action system, or other possible action systems, which are not specifically limited in the embodiment of the present application.

[0143] The information processing device provided in the embodiment of the present application can implement each process implemented in the above method embodiment. To avoid repetition, it will not be described here.

[0144] Alternatively, as Figure 5 As shown, an embodiment of the present application also provides an electronic device 510, including a processor 511, a memory 512, and a program or instruction stored in the memory 512 and executable on the processor 511. When the program or instruction is executed by the processor 511, the various steps of any of the above-mentioned information processing method embodiments are implemented, and the same technical effect can be achieved. To avoid repetition, they will not be described here.

[0145] It should be noted that the electronic devices in the embodiments of the present application include the above-mentioned mobile electronic devices and non-mobile electronic devices.

[0146] Figure 6 A schematic diagram of the hardware structure of an electronic device implementing an embodiment of the present application.

[0147] The electronic device 600 includes but is not limited to components such as a radio frequency unit 601 , a network module 602 , an audio output unit 603 , an input unit 604 , a sensor 605 , a display unit 606 , a user input unit 607 , an interface unit 608 , a memory 609 , and a processor 610 .

[0148] Those skilled in the art will understand that the electronic device 600 may also include a power source (such as a battery) to power each component, and the power source may be logically connected to the processor 610 through a power management system, thereby implementing functions such as charging, discharging, and power consumption management through the power management system. Figure 6 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be repeated here.

[0149] The user input unit 607 is configured to receive first request information input by a user;

[0150] Processor 610, configured to perform intent analysis on the first request information to obtain first intent information and first scenario information;

[0151] The processor 610 is further configured to start a first application according to the first scenario information;

[0152] The processor 610 is further configured to process the first intent information in the first application.

[0153] Optionally, the processor 610 is further configured to input the first request information into a first intent splitting model, and output the first intent information and the first scenario information;

[0154] The first intent splitting model is trained based on multiple groups of first training data, and each group of the first training data includes: first sample request information, first label intent information and first label scene information associated with the first sample request information.

[0155] Optionally, the processor 610 is further configured to determine, based on a scenario database, at least one application information matching the first scenario information, the scenario database including multiple intent information and at least one application information associated with each intent information;

[0156] The processor 610 is further configured to determine first application information from the at least one application information;

[0157] The processor 610 is further configured to start the first application indicated by the first application information.

[0158] Optionally, the processor 610 is further configured to display the at least one application information, wherein the at least one application information is arranged according to the recommended order information;

[0159] The processor 610 is further configured to determine the first application information in response to a selection input of the first application information in the at least one application information.

[0160] Optionally, the processor 610 is further configured to input the first request information into a second intent splitting model, and output the first application information, the first intent information, and the first scenario information;

[0161] The second intent splitting model is trained based on multiple sets of second training data, each set of the second training data including: second sample request information, label application information associated with the second sample request information, second label intent information, and second label scenario information;

[0162] The processor 610 is further configured to start the first application indicated by the first application information.

[0163] In an embodiment of the present application, intent analysis is performed on a first request message input by a user to obtain first intent information and first scenario information. Based on the first scenario information, a first application is launched. By directly analyzing the scenario information, compatible applications are quickly selected, significantly reducing the time required to launch the application. Processing the first intent information in the first application accurately executes the first intent information in the first request message, improving the efficiency of responding to the first request message.

[0164] It should be understood that in an embodiment of the present application, the input unit 604 may include a graphics processing unit (GPU) 6041 and a microphone 6042, and the graphics processor 6041 processes image data of a static picture or video image obtained by an image capture device (such as a camera) in a video image capture mode or an image capture mode. The display unit 606 may include a display panel 6061, and the display panel 6061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 607 includes a touch panel 6071 and at least one of other input devices 6072. The touch panel 6071 is also called a touch screen. The touch panel 6071 may include two parts: a touch detection device and a touch controller. Other input devices 6072 may include but are not limited to a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and an action stick, which will not be repeated here. The memory 609 can be used to store software programs and various data, including but not limited to applications and action systems. The processor 610 may integrate an application processor and a modem processor, wherein the application processor mainly processes the action system, user pages and applications, etc., and the modem processor mainly processes wireless communications. It is understandable that the modem processor may not be integrated into the processor 610.

[0165] The memory 609 can be used to store software programs and various data. The memory 609 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, applications or instructions required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 609 may include a volatile memory or a non-volatile memory, or the memory x09 may include both volatile and non-volatile memory. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM), and a direct memory bus random access memory (DRRAM). The memory 609 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0166] Processor 610 may include one or more processing units. Optionally, processor 610 integrates an application processor and a modem processor. The application processor primarily handles operations related to the operating system, user interface, and application programs, while the modem processor primarily processes wireless communication signals, such as a baseband processor. It is understood that the modem processor may not be integrated into processor 610.

[0167] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned information processing method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0168] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0169] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned information processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0170] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0171] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned information processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0172] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising 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, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be noted that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0173] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), including a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0174] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. An information processing method, characterized in that: The method comprises: receiving first request information input by a user; Performing intent analysis on the first request information to obtain first intent information and first scenario information; Starting a first application according to the first scenario information; Process the first intent information in the first application.

2. The method according to claim 1, characterized in that The analyzing the first request information to obtain the first intent information and the first scenario information includes: Inputting the first request information into a first intent splitting model, and outputting the first intent information and the first scenario information; The first intent splitting model is trained based on multiple groups of first training data, and each group of the first training data includes: first sample request information, first label intent information and first label scene information associated with the first sample request information.

3. The method according to claim 1, characterized in that The starting the first application according to the first scenario information includes: Determining, based on a scenario database, at least one application information matching the first scenario information, the scenario database comprising a plurality of intent information and at least one application information associated with each of the intent information; determining first application information from the at least one application information; Start the first application indicated by the first application information.

4. The method according to claim 3, characterized in that The determining the first application information from the at least one application information includes: displaying the at least one piece of application information, wherein the at least one piece of application information is arranged according to the recommended order information; In response to a selection input of the first application information among the at least one application information, the first application information is determined.

5. The method according to claim 1, wherein The analyzing the first request information to obtain the first intent information and the first scenario information includes: Inputting the first request information into a second intent splitting model, and outputting the first application information, the first intent information, and the first scenario information; The second intent splitting model is trained based on multiple sets of second training data, each set of the second training data including: second sample request information, label application information associated with the second sample request information, second label intent information, and second label scenario information; The method further comprises: Start the first application indicated by the first application information.

6. An information processing device, characterized in that The device comprises: A receiving module, configured to receive first request information input by a user; an analysis module, configured to perform intent analysis on the first request information to obtain first intent information and first scenario information; A starting module, configured to start a first application according to the first scenario information; A processing module is used to process the first intent information in the first application.

7. The device according to claim 6, characterized in that The analysis module is specifically used to: Inputting the first request information into a first intent splitting model, and outputting the first intent information and the first scenario information; The first intent splitting model is trained based on multiple groups of first training data, and each group of the first training data includes: first sample request information, first label intent information and first label scene information associated with the first sample request information.

8. The device according to claim 6, characterized in that The startup module includes: a determination module, configured to determine, based on a scenario database, at least one application information matching the first scenario information, the scenario database comprising a plurality of intent information and at least one application information associated with each of the intent information; The determining module is further configured to determine the first application information from the at least one application information; The starting module is specifically configured to start the first application indicated by the first application information.

9. The device according to claim 8, characterized in that The determining module is specifically configured to: displaying the at least one piece of application information, wherein the at least one piece of application information is arranged according to the recommended order information; In response to a selection input of the first application information among the at least one application information, the first application information is determined.

10. The device according to claim 6, characterized in that The analysis module is specifically used to: Inputting the first request information into a second intent splitting model, and outputting the first application information, the first intent information, and the first scenario information; The second intent splitting model is trained based on multiple sets of second training data, each set of the second training data including: second sample request information, label application information associated with the second sample request information, second label intent information, and second label scenario information; The starting module is specifically configured to start the first application indicated by the first application information.

Images