Electronic whiteboard input method automatic switching method and system based on environment recognition

Abstract

The invention relates to the technical field of input method management, in particular to an electronic whiteboard input method automatic switching method and system based on environment recognition. The method comprises the following steps: constructing an environment information model by monitoring system language and context environment in real time and integrating voice recognition, text analysis and image processing technologies; reasoning and dynamically updating an input method mapping table by utilizing a Markov logic network algorithm; based on the mapping table, identifying a user input demand through a prediction algorithm and automatically switching to the most suitable input method; user feedback and use habits are recorded, a deep learning algorithm is used for analysis, an input method mapping rule and a switching algorithm are continuously optimized, and intelligent input method management is achieved. According to the method, the input method is automatically adjusted by sensing the change of the used language and the context environment in real time, so that the user experience and the intelligent level of the system are improved, the efficient and automatic switching of the electronic whiteboard input method is realized, and the input efficiency and the use experience of the user are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of input method management, and in particular to an automatic switching method and system of electronic whiteboard input methods based on environment recognition. Background Art

[0002] With the rapid development of information technology and the increasing demand for efficient office work, electronic whiteboards have been widely used as a new interactive tool. Most of the existing electronic whiteboard input methods are fixed input modes, which cannot automatically adjust the input method according to the changes in the language and context environment. For example, in a multilingual teaching scenario, the system may need to automatically switch input methods according to different languages; in meeting records of different topics, the system needs to switch different input methods (such as text, symbols, charts, etc.) according to the context content. In these scenarios, the traditional electronic whiteboard input method switching requires manual intervention, cumbersome operation, and low efficiency. At present, there are still the following problems: most of the existing technologies rely on manual operation or simple preset rules, and cannot monitor and respond to changes in the system language and context environment in real time, resulting in insufficient timeliness and accuracy of input method switching, affecting the user's experience; the existing technology lacks intelligent reasoning and dynamic update capabilities, and cannot be intelligently adjusted according to the user's historical behavior and current environment, resulting in the input method switching failing to meet the user's personalized needs. Summary of the invention

[0003] To solve the above problems, the present invention provides an automatic switching method and system for electronic whiteboard input methods based on environment recognition, which solves the problem of how to realize automatic and intelligent switching of input methods in a multi-language and multi-scenario electronic whiteboard usage environment, realizes efficient automatic switching of electronic whiteboard input methods, and improves the user's input efficiency and usage experience.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] On the one hand, a method for automatically switching an electronic whiteboard input method based on environment recognition comprises the following steps:

[0006] S1: Monitor the system language and context in real time by calling the API interface, extract environmental data by integrating speech recognition, text analysis and image processing technology, and build a real-time environmental information model;

[0007] S2: using a Markov logic network algorithm to infer the environment information model, and dynamically updating the input method mapping table according to the user's historical behavior and the current environment;

[0008] S3: Based on the environment information model and the updated input method mapping table, a prediction algorithm is used to identify the user's input needs and automatically switch to the appropriate input method;

[0009] S4: Record user satisfaction and usage habits with input method switching, use deep learning algorithms to analyze feedback data, and continuously optimize input method mapping rules and switching algorithms.

[0010] Furthermore, the environmental data includes data on system language settings, active applications, the type of current application, usage scenarios, and user input behavior.

[0011] Furthermore, it also includes setting a context change listener to trigger a context change event when a change in the active application is detected; using a timer or an operating system callback mechanism to periodically or in real time detect changes in the current active application, and once a change is detected, triggering corresponding event processing.

[0012] Furthermore, it also includes adding or modifying the mapping relationship between language, context environment and input method through the setting interface; providing a user interface to allow the user to configure the input method mapping relationship between language and context environment; after the user saves the settings, the system updates the mapping table in the memory and persists it to the configuration file or database.

[0013] Furthermore, the step S1 specifically includes the following steps:

[0014] By calling the system API interface, the environmental data is collected in real time;

[0015] Use speech recognition technology to analyze user audio input, identify language features and extract potential usage intent;

[0016] Use natural language processing technology to parse user text input, extract key contextual information, and identify language patterns and input requirements;

[0017] Analyze the image data captured by the camera through image processing algorithms to identify the current application scenario;

[0018] Integrate the above multimodal data and build a real-time environmental information model through the Bayesian network algorithm.

[0019] Furthermore, the formula of the environment information model is as follows:

[0020]

[0021] Where EIM represents the result of the environmental information model, that is, the optimal environmental state; E represents the set of environmental states; n represents the number of environmental variables; α i and β i represents the weight parameter; Sim(E i , H i ) indicates the current environment state E i With the historical state Hi Similarity; P(E i |x i ) means that in a given environment variable x i Under the condition, the environmental state E i The probability of x i Indicates the value of the i-th environment variable; H i Represents the historical state of the i-th environment variable.

[0022] Furthermore, the step S2 specifically includes the following steps:

[0023] Extracting current environment characteristic variables from the environment information model;

[0024] Call the user history behavior database to retrieve the history input method usage records that match the current environment characteristics;

[0025] The Markov logic network algorithm is used to calculate the association probability between the current environment characteristics and the historical input method usage records, and a preliminary input method mapping table is generated;

[0026] Apply an adaptive weight adjustment mechanism to optimize the preliminary mapping table according to real-time environmental changes;

[0027] Based on the optimized mapping table, the system input method configuration is updated in real time.

[0028] Furthermore, the formula of the Markov logic network algorithm is as follows:

[0029]

[0030] Where P(F|G) represents the probability of selecting input method G under a specific environment state E; Z represents the normalization constant; N represents the number of input methods; m represents the number of environmental characteristic variables; w ij Indicates input method F i and environmental characteristics j The weight parameter between ij (F i , G j ) indicates input method F i and environmental characteristics j The correlation measure between .

[0031] Furthermore, the step S3 specifically includes the following steps:

[0032] Extracting the input method candidate set that best matches the current environment features from the updated input method mapping table;

[0033] Use dynamic programming algorithm to predict user input needs, calculate fitness to select the input method suitable for the current environment, and automatically switch to the input method;

[0034] After switching the input method, monitor the user's input behavior and feedback, and evaluate and adjust the adaptability of the input method.

[0035] Furthermore, the construction process of the dynamic programming algorithm includes the following steps:

[0036] Define the state space, including the current environment characteristics and the historical data of user input behavior;

[0037] Construct a state transition equation to describe the relationship between changes in environmental characteristics and input method switching;

[0038] Initialize the weights and transition probabilities of the state space based on historical data;

[0039] Apply dynamic programming algorithm to recursively calculate each candidate input method to solve the optimal input method switching path;

[0040] Based on the recursive calculation results, the fitness of each candidate input method is evaluated and the input method with the highest fitness is selected.

[0041] Furthermore, the step S4 specifically includes the following steps:

[0042] Collect user operation data during input method switching, including switching frequency, usage duration, and error rate;

[0043] Collect user satisfaction with input method switching through user interface feedback and record users' subjective evaluation of input method selection in different scenarios;

[0044] A deep learning algorithm is used to jointly analyze the collected user operation data and feedback data to generate an optimization model; the deep learning algorithm constructs and trains a multi-layer neural network model, combines supervised learning and unsupervised learning methods, and performs pattern recognition and trend analysis on the user operation data and feedback data;

[0045] Adjust the input method mapping table and switching algorithm based on the optimization model, and dynamically update system parameters.

[0046] On the other hand, an automatic switching system for electronic whiteboard input methods based on environment recognition includes an environment monitoring module, a reasoning mapping module, an input method switching module and a feedback optimization module connected in sequence;

[0047] The environment monitoring module is used to call the API interface to monitor the system language and context environment in real time, extract environmental data by integrating speech recognition, text analysis and image processing technology, and build a real-time environmental information model;

[0048] The inference mapping module is used to use the Markov logic network algorithm to infer the environment information model and dynamically update the input method mapping table according to the user's historical behavior and the current environment;

[0049] The input method switching module is used to identify the user's input requirements and automatically switch to a suitable input method based on the environment information model and the updated input method mapping table using a prediction algorithm;

[0050] The feedback optimization module is used to record user satisfaction and usage habits with input method switching, analyze feedback data using a deep learning algorithm, and continuously optimize input method mapping rules and switching algorithms.

[0051] The beneficial effects of the present invention are as follows: by calling the API interface and integrating speech recognition, text analysis, and image processing technology, the present invention can monitor and analyze the system language and context environment in real time, build an accurate real-time environmental information model, and ensure the timeliness and accuracy of input method switching. The environmental information model is inferred by using the Markov logic network algorithm, and the input method mapping table is dynamically updated in combination with the user's historical behavior and the current environment to achieve intelligent input method switching to meet the personalized needs of users. Based on the environmental information model and the updated input method mapping table, a prediction algorithm is used to identify the user's input needs, and automatically switch to the appropriate input method to improve the user's input efficiency and user experience. The user's satisfaction and usage habits with input method switching are recorded, and the feedback data is analyzed using a deep learning algorithm, and the input method mapping rules and switching algorithms are continuously optimized, so that the system can continuously adapt and improve and provide better services. By intelligently identifying and automatically switching input methods, the operation of users manually switching input methods is reduced, the convenience and satisfaction of users are improved, and the user experience is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The present invention is a flowchart of an automatic switching method of an electronic whiteboard input method based on environment recognition.

[0053] Figure 2 It is a flowchart of step S2 provided in an embodiment of the present invention.

[0054] Figure 3 The present invention is a module schematic diagram of an automatic switching system of electronic whiteboard input method based on environment recognition. DETAILED DESCRIPTION

[0055] See also Figure 1-3 As shown, the present invention relates to an automatic switching method and system for electronic whiteboard input methods based on environment recognition.

[0056] Example 1

[0057] An automatic switching method of an electronic whiteboard input method based on environment recognition comprises the following steps:

[0058] S1: Monitor the system language and context environment in real time by calling the API interface, extract environmental data by integrating speech recognition, text analysis and image processing technology, and build a real-time environmental information model; the environmental data includes system language settings, active applications, the type of current application, usage scenarios and user input behavior data.

[0059] Wherein, the step S1 specifically includes the following steps:

[0060] By calling the system API interface, the environmental data is collected in real time;

[0061] Specifically, call the operating system's API interface to obtain the language setting of the current system. For example, in Windows system, you can obtain the language setting of the current user interface by calling the `GetUserDefaultUILanguage` function. Get the currently active application and its window title information through the API interface provided by the operating system. For example, in Windows system, you can use the `GetForegroundWindow` function to obtain the handle of the currently active window, and then use the `GetWindowText` function to obtain the title of the window. Current application type: Determine the type of application (such as browser, document editor, communication tool, etc.) based on the process name or window title of the active application. Usage scenario data: Call the operating system API interface to obtain data on the current usage scenario, such as whether screen sharing is in progress, whether it is connected to a specific network, etc. User input behavior data: Record user behavior data such as keyboard input and mouse clicks, which can be obtained through system hooks or monitoring tools.

[0062] Use speech recognition technology to analyze user audio input, identify language features and extract potential usage intent;

[0063] Specifically, the user's audio input is captured through a microphone. The audio signal captured by the microphone needs to be preprocessed, such as removing noise and echo, to improve the accuracy of speech recognition. The audio signal is converted into text using a speech recognition engine (such as Google Speech-to-Text, Microsoft Azure Speech Services, etc.). This process involves multiple steps such as acoustic models, language models, and decoding algorithms. The language features in the audio text are analyzed, such as voice intonation, speaking speed, and pauses. These features can help identify the user's emotional state and speaking intention. Based on the identified text and language features, the user's potential usage intention is extracted.

[0064] Use natural language processing technology to parse user text input, extract key contextual information, and identify language patterns and input requirements;

[0065] Specifically, the text input by the user is preprocessed, including word segmentation, removal of stop words and punctuation marks, etc. The text is tagged with parts of speech and named entity recognition using natural language processing models (such as spaCy, NLTK, etc.) to extract key information in the text, such as time, place, and person. Through dependency syntactic analysis, the dependency relationship between words in the text is determined to understand the semantic structure of the sentence. For example, for the sentence "I need to book a flight ticket for tomorrow", dependency syntactic analysis can identify "I" as the subject, "need" as the verb, "book" as the object, and "tomorrow's flight ticket" as the object of the reservation. Based on the results of part-of-speech tagging and dependency syntactic analysis, key information in the context is extracted. For example, the time information "tomorrow", the action "book", and the object "flight ticket" are identified. By analyzing the language pattern and structure of the text, the user's input needs are identified. For example, it is determined whether the user is asking a question, issuing a command, or expressing emotion.

[0066] Analyze the image data captured by the camera through image processing algorithms to identify the current application scenario;

[0067] Specifically, the environmental image data is captured in real time through the camera. The camera can be a built-in camera of a computer or mobile phone, or it can be an external camera. The captured image is preprocessed, including grayscale, denoising and image enhancement, to improve the accuracy of subsequent processing. The image is analyzed using image processing algorithms (such as convolutional neural networks, deep learning models, etc.) to identify the current application scenario. For example, identify whether the user is in an office environment, a home environment or an outdoor environment. On the basis of scene recognition, further object detection and recognition are performed to determine the specific objects and activities in the image. For example, identify whether the user is using a computer, reading a document or talking to others. Based on the object detection and recognition results, the user's behavior pattern is analyzed. For example, a user using a computer in an office environment may be processing work; reading a document in a home environment may be studying or looking up information.

[0068] Integrate the above multimodal data and build a real-time environmental information model through the Bayesian network algorithm.

[0069] Specifically, the multimodal data obtained from the system API interface, speech recognition, natural language processing and image processing technology are integrated. These data include system language settings, active application information, current application type, usage scenario data, user input behavior data, audio text, language features, usage intention, text context key information and image recognition results. Design the structure of the Bayesian network and determine the nodes and their dependencies. The nodes of the Bayesian network can represent different data types or features, such as system language, user intention, usage scenario, etc., and the edges represent the dependencies between these nodes. Use the collected data to learn the parameters of the Bayesian network and determine the conditional probability distribution between the nodes. Parameter learning can use methods such as maximum likelihood estimation or Bayesian estimation. Through the Bayesian network, reasoning is performed to calculate the posterior probability of different nodes, thereby inferring the user's current status and potential needs. For example, based on the system language settings, active applications and user audio input, the user's usage intention can be inferred; based on image recognition results and text input, the user's usage scenario and behavior pattern can be inferred. According to actual usage and feedback, the Bayesian network model is continuously optimized to improve the accuracy and robustness of the model.

[0070] Furthermore, the formula of the environment information model is as follows:

[0071]

[0072] Where EIM represents the result of the environmental information model, that is, the optimal environmental state; E represents the set of environmental states; n represents the number of environmental variables; α i and β i represents the weight parameter; Sim(E i , H i ) indicates the current environment state E i With the historical state H i Similarity; P(E i |x i ) means that in a given environment variable x i Under the condition, the environmental state E i The probability of x i Indicates the value of the i-th environment variable; H i Represents the historical state of the i-th environment variable.

[0073] S2: using a Markov logic network algorithm to infer the environment information model, and dynamically updating the input method mapping table according to the user's historical behavior and the current environment;

[0074] Wherein, the step S2 specifically includes the following steps:

[0075] Extracting current environment characteristic variables from the environment information model;

[0076] Call the user history behavior database to retrieve the history input method usage records that match the current environment characteristics;

[0077] It should be noted that the user history behavior database records the user's use of input methods in different environments. The records in the database include timestamps, environmental features (such as language settings, application types, usage scenarios), and the type of input method used. Define retrieval rules to search for matching historical records in the database based on the currently extracted environmental feature variables. Matching rules can be defined based on the similarity or degree of complete matching of features. Use SQL queries or other database query methods to retrieve historical input method usage records that match the current environmental features. These records will serve as the data basis for subsequent calculations of association probabilities.

[0078] The Markov logic network algorithm is used to calculate the association probability between the current environment characteristics and the historical input method usage records, and a preliminary input method mapping table is generated;

[0079] It should be noted that the Markov logic network (MLN) combines the probability representation of the Markov network and the expressive power of the first-order logic. The model contains a set of logical formulas and weights that represent the relationship between different environmental features and input method usage records. The reasoning process is as follows: Input the currently extracted environmental feature variables into the MLN model. : MLN uses an inference algorithm to calculate the probability of use of each input method in the current environment. The inference algorithm calculates the association probability of each input method by traversing all logical formulas and weights. Based on the calculation results, a preliminary input method mapping table is generated. The mapping table lists the probability values ​​of each input method, which represent the applicability of the input method in the current environment.

[0080] Apply an adaptive weight adjustment mechanism to optimize the preliminary mapping table according to real-time environmental changes;

[0081] Specifically, the system monitors changes in environmental characteristics in real time, such as when a user switches applications or changes usage scenarios. Define an adaptive weight adjustment strategy to dynamically adjust the weights of logical formulas in MLN according to environmental changes. For example, when it is detected that a user frequently uses a certain input method, the system can increase the weight of the input method accordingly. Optimize the initially generated input method mapping table so that it can more accurately reflect the current environment and user habits. The optimization process can be based on a comprehensive analysis of real-time feedback and historical data.

[0082] Based on the optimized mapping table, the system input method configuration is updated in real time.

[0083] Furthermore, the formula of the Markov logic network algorithm is as follows:

[0084]

[0085] Where P(F|G) represents the probability of selecting input method G under a specific environment state E; Z represents the normalization constant; N represents the number of input methods; m represents the number of environmental characteristic variables; w ij Indicates input method F i and environmental characteristics j The weight parameter between ij (F i , G j ) indicates input method F i and environmental characteristics j The correlation measure between .

[0086] S3: Based on the environment information model and the updated input method mapping table, a prediction algorithm is used to identify the user's input needs and automatically switch to the appropriate input method;

[0087] Wherein, the step S3 specifically includes the following steps:

[0088] Extracting the input method candidate set that best matches the current environment features from the updated input method mapping table;

[0089] Use dynamic programming algorithm to predict user input needs, calculate fitness to select the input method suitable for the current environment, and automatically switch to the input method;

[0090] After switching the input method, monitor the user's input behavior and feedback, and evaluate and adjust the adaptability of the input method.

[0091] The construction process of the dynamic programming algorithm includes the following steps:

[0092] Define the state space, including the current environment characteristics and the historical data of user input behavior;

[0093] Construct a state transition equation to describe the relationship between changes in environmental characteristics and input method switching;

[0094] Initialize the weights and transition probabilities of the state space based on historical data;

[0095] Apply dynamic programming algorithm to recursively calculate each candidate input method to solve the optimal input method switching path;

[0096] Based on the recursive calculation results, the fitness of each candidate input method is evaluated and the input method with the highest fitness is selected.

[0097] It should be noted that the state space is defined as follows: The state space includes the current environment characteristics and the historical data of user input behavior. Each state in the state space represents a specific combination of environment characteristics and input behavior. For example, a state may include the environment characteristics of "entering English in a text editor" and the input behavior of "last time using the English input method".

[0098] Constructing state transition equations: The state transition equations describe the relationship between changes in environment characteristics and input method switching. For example, if a user switches from a Chinese input environment to an English input environment, the state transition equation will describe the process of switching from a Chinese input method to an English input method.

[0099] Initialize the weight and transition probability of the state space: The system initializes the weight and transition probability of the state space based on historical data. This data may include the frequency of input method usage and switching records of users in different environments. For example, if a user frequently uses a certain input method in a certain environment, the weight of the input method will be higher, and the probability of switching to the input method will also increase.

[0100] Recursively calculate the optimal input method switching path: The system applies a dynamic programming algorithm to recursively calculate each candidate input method to solve the optimal input method switching path. This path takes into account the current environment characteristics, historical data, and the fitness of the input method. For example, the system calculates the steps and fitness required to switch to a certain input method in the current environment, and selects the optimal path for switching.

[0101] Evaluate the fitness of candidate input methods: Based on the calculation results of the dynamic programming algorithm, the system evaluates the fitness of each candidate input method. Input methods with higher fitness are more in line with the current environment characteristics and user input needs. For example, if a certain input method has the highest fitness in the current environment, the system selects it as the optimal input method.

[0102] Automatically switch to the most adaptable input method: Once the optimal input method is determined, the system will automatically switch to it to ensure that users get the best input experience in the current environment.

[0103] S4: Record user satisfaction and usage habits with input method switching, use deep learning algorithms to analyze feedback data, and continuously optimize input method mapping rules and switching algorithms.

[0104] Wherein, the step S4 specifically includes the following steps:

[0105] Collect user operation data during input method switching, including switching frequency, usage duration, and error rate;

[0106] Collect user satisfaction with input method switching through user interface feedback and record users' subjective evaluation of input method selection in different scenarios;

[0107] A deep learning algorithm is used to jointly analyze the collected user operation data and feedback data to generate an optimization model; the deep learning algorithm constructs and trains a multi-layer neural network model, combines supervised learning and unsupervised learning methods, and performs pattern recognition and trend analysis on the user operation data and feedback data;

[0108] Adjust the input method mapping table and switching algorithm based on the optimization model, and dynamically update system parameters.

[0109] It also includes setting a context change listener to trigger a context change event when a change in the active application is detected; using a timer or the operating system's callback mechanism to periodically or in real time detect changes in the current active application, and once a change is detected, triggering corresponding event processing.

[0110] Specifically, initialize the context change listener in the system service to ensure that the listener starts automatically when the system starts. Determine the scope of the context change that needs to be monitored, including active applications, application types, usage scenarios, etc. Use the API provided by the operating system (such as WMI of Windows, NSWorkspace of macOS, dbus of Linux, etc.) to obtain the currently active application and its related information. Set a timer and call the API interface regularly to obtain the latest system status to ensure the real-time nature of the environmental information. Monitor the system environment in real time through the listener, and immediately trigger a context change event when a change in active application or usage scenario is detected. After the event is triggered, call the corresponding event processing mechanism, update the environmental information model, and switch the input method according to the new context environment.

[0111] Set a reasonable timing cycle (such as once per second) and use the timer to regularly check the current system environment to ensure a quick response to environmental changes. Consider system performance in the implementation of the timer, minimize the occupation of system resources, and avoid performance issues caused by frequent API calls. Use the callback mechanism provided by the operating system, such as the Windows message loop, macOS notification center, etc., to automatically trigger the callback function when the system environment changes. Implement the logic for obtaining and updating environmental information in the callback function to ensure that you respond to environmental changes in the first place. Detect changes in information such as active applications, application types, and usage scenarios in real time. Once changes are detected, update the environmental information model immediately. Trigger the corresponding event processing logic and recalculate the input method switching strategy based on the new environmental information to ensure that the input method always adapts to the current usage environment.

[0112] It also includes adding or modifying the mapping relationship between language, context environment and input method through the setting interface; providing a user interface to allow users to configure the input method mapping relationship between language and context environment; after the user saves the settings, the system updates the mapping table in the memory and persists it to the configuration file or database.

[0113] Specifically, design a user-friendly configuration interface that allows users to easily add and modify the mapping relationship between language and context environment and input method. : The interface should include functions such as language selection, application type selection, input method mapping settings, and provide preview and save options. After the user saves the settings, the system will immediately update the mapping table in memory to ensure that the new configuration can take effect immediately. At the same time, the updated configuration is persistently stored in the configuration file or database so that the user's customized settings can be restored after the system is restarted. During the input method switching process, priority is given to the customized rules set by the user through the configuration interface to ensure that the user's personalized needs are met. Combine user configuration with the rules automatically generated by the system to dynamically adjust the input method switching strategy to achieve the best user experience.

[0114] Example 2

[0115] An automatic switching system for electronic whiteboard input methods based on environment recognition, comprising an environment monitoring module, a reasoning mapping module, an input method switching module and a feedback optimization module connected in sequence;

[0116] The environment monitoring module is used to call the API interface to monitor the system language and context environment in real time, extract environmental data by integrating speech recognition, text analysis and image processing technology, and build a real-time environmental information model;

[0117] The inference mapping module is used to use the Markov logic network algorithm to infer the environment information model and dynamically update the input method mapping table according to the user's historical behavior and the current environment;

[0118] The input method switching module is used to identify the user's input requirements and automatically switch to a suitable input method based on the environment information model and the updated input method mapping table using a prediction algorithm;

[0119] The feedback optimization module is used to record user satisfaction and usage habits with input method switching, analyze feedback data using a deep learning algorithm, and continuously optimize input method mapping rules and switching algorithms.

[0120] In this embodiment, an automatic switching system of an electronic whiteboard input method based on environment recognition is applied to an automatic switching method of an electronic whiteboard input method based on environment recognition described in Example 1, which will not be described in detail here.

[0121] In summary, the present invention can intelligently identify the user's input needs by real-time monitoring of system language, application type, user behavior and environmental changes, and automatically switch to the most suitable input method based on Markov logic network and dynamic programming algorithm. This improves input efficiency and reduces the frequency of users manually switching input methods. By comprehensively utilizing speech recognition, natural language processing and image processing technology, a real-time environmental information model is constructed, and reasoning is performed through a Bayesian network. This fusion of multimodal data greatly enhances the system's ability to understand user behavior and environment, making the switching of input methods more targeted and accurate.

[0122] During use, the present invention dynamically adjusts the input method mapping table based on the user's historical behavior data and real-time feedback. Combined with the deep learning algorithm for continuous optimization, the adaptive adjustment of the input method is realized, which not only meets the current needs of users, but also continuously improves the future use experience. Users can customize the mapping relationship between language and environment and input method through the configuration interface, and the system can also automatically adjust the input method according to environmental changes. This combination ensures that the personalized needs of users are met and further enhances the flexibility and ease of use of the system.

[0123] The present invention is provided with a context environment change listener and a timer, which can monitor and respond to changes in the environment in real time, ensuring that the input method is always adapted to the user's current usage environment. This design makes the switching of input methods more efficient and the user experience smoother. By recording the user's usage habits and feedback, the system uses a deep learning algorithm to analyze these data and continuously optimizes the input method mapping rules and switching algorithms. This feedback-based optimization mechanism ensures that the system can better adapt to the user's needs over time and improve user satisfaction.

[0124] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for automatically switching electronic whiteboard input methods based on environment recognition, characterized in that: The following steps are involved: S1: Monitor the system language and context in real time by calling the API interface, extract environmental data by integrating speech recognition, text analysis and image processing technology, and build a real-time environmental information model; S2: using a Markov logic network algorithm to infer the environment information model, and dynamically updating the input method mapping table according to the user's historical behavior and the current environment; S3: Based on the environment information model and the updated input method mapping table, a prediction algorithm is used to identify the user's input needs and automatically switch to the appropriate input method; S4: Record user satisfaction and usage habits with input method switching, use deep learning algorithms to analyze feedback data, and continuously optimize input method mapping rules and switching algorithms.

2. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: The environment data includes data on system language settings, active applications, the type of current application, usage scenarios, and user input behavior.

3. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: It also includes setting up a context change listener to trigger a context change event when a change in the active application is detected; Use a timer or operating system callback mechanism to periodically or in real time detect changes in the currently active application. Once a change is detected, trigger the corresponding event processing.

4. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: It also includes adding or modifying the mapping relationship between language, context environment and input method through the setting interface; providing a user interface to allow users to configure the input method mapping relationship between language and context environment; after the user saves the settings, the system updates the mapping table in the memory and persists it to the configuration file or database.

5. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: The step S1 specifically includes the following steps: By calling the system API interface, the environmental data is collected in real time; Use speech recognition technology to analyze user audio input, identify language features and extract potential usage intent; Use natural language processing technology to parse user text input, extract key contextual information, and identify language patterns and input requirements; Analyze the image data captured by the camera through image processing algorithms to identify the current application scenario; Integrate the above multimodal data and build a real-time environmental information model through the Bayesian network algorithm.

6. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 5, characterized in that: The formula of the environmental information model is as follows: Where EIM represents the result of the environmental information model, that is, the optimal environmental state; E represents the set of environmental states; n represents the number of environmental variables; α i and β i represents the weight parameter; Sim(E i , H i ) indicates the current environment state E i With the historical state H i Similarity; P(E i |x i ) means that in a given environment variable x i Under the condition, the environmental state E i The probability of x i Indicates the value of the i-th environment variable; H i Represents the historical state of the i-th environment variable.

7. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: The step S2 specifically includes the following steps: Extracting current environment characteristic variables from the environment information model; Call the user history behavior database to retrieve the history input method usage records that match the current environment characteristics; The Markov logic network algorithm is used to calculate the association probability between the current environment characteristics and the historical input method usage records, and a preliminary input method mapping table is generated; Apply an adaptive weight adjustment mechanism to optimize the preliminary mapping table according to real-time environmental changes; Based on the optimized mapping table, the system input method configuration is updated in real time.

8. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 7, characterized in that: The formula of the Markov logic network algorithm is as follows: Where P(F|G) represents the probability of selecting input method G under a specific environment state E; Z represents the normalization constant; N represents the number of input methods; m represents the number of environmental characteristic variables; w ij Indicates input method F i Environmental characteristics j The weight parameter between ij (F i , G j ) indicates input method F i Environmental characteristics j The correlation measure between .

9. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: The step S3 specifically comprises the following steps: Extracting the input method candidate set that best matches the current environment features from the updated input method mapping table; Use dynamic programming algorithm to predict user input needs, calculate fitness to select the input method suitable for the current environment, and automatically switch to the input method; After switching the input method, monitor the user's input behavior and feedback, and evaluate and adjust the adaptability of the input method.

10. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 9, characterized in that: The construction process of the dynamic programming algorithm includes the following steps: Define the state space, including the current environment characteristics and the historical data of user input behavior; Construct a state transition equation to describe the relationship between changes in environmental characteristics and input method switching; Initialize the weights and transition probabilities of the state space based on historical data; Apply dynamic programming algorithm to recursively calculate each candidate input method to solve the optimal input method switching path; Based on the recursive calculation results, the fitness of each candidate input method is evaluated and the input method with the highest fitness is selected.

11. The method for automatically switching electronic whiteboard input methods based on environment recognition according to claim 1, characterized in that: The step S4 specifically comprises the following steps: Collect user operation data during input method switching, including switching frequency, usage duration, and error rate; Collect user satisfaction with input method switching through user interface feedback and record users' subjective evaluation of input method selection in different scenarios; A deep learning algorithm is used to jointly analyze the collected user operation data and feedback data to generate an optimization model; the deep learning algorithm constructs and trains a multi-layer neural network model, combines supervised learning and unsupervised learning methods, and performs pattern recognition and trend analysis on the user operation data and feedback data; Adjust the input method mapping table and switching algorithm based on the optimization model, and dynamically update system parameters.

12. An automatic switching system for electronic whiteboard input method based on environment recognition, characterized in that: The system is applied to an automatic switching method of an electronic whiteboard input method based on environment recognition as described in any one of claims 1 to 11, comprising an environment monitoring module, a reasoning mapping module, an input method switching module and a feedback optimization module connected in sequence; The environment monitoring module is used to call the API interface to monitor the system language and context environment in real time, extract environmental data by integrating speech recognition, text analysis and image processing technology, and build a real-time environmental information model; The inference mapping module is used to use the Markov logic network algorithm to infer the environment information model and dynamically update the input method mapping table according to the user's historical behavior and the current environment; The input method switching module is used to identify the user's input requirements and automatically switch to a suitable input method based on the environment information model and the updated input method mapping table using a prediction algorithm; The feedback optimization module is used to record user satisfaction and usage habits with input method switching, analyze feedback data using a deep learning algorithm, and continuously optimize input method mapping rules and switching algorithms.

Images