Cross-application data and process intelligent circulation method and system based on user intention prediction in cloud desktop environment

Abstract

The invention discloses a cross-application data and process intelligent circulation method and system based on user intention prediction in a cloud desktop environment, and belongs to the technical field of cloud computing, artificial intelligence and human-computer interaction. The method comprises the following steps: monitoring a cross-application operation behavior sequence and context data of a user in a cloud desktop environment; predicting a user work intention and subsequent steps by using an intention learning model based on the data; extracting or generating associated data from the source application program according to the prediction result; and automatically pre-filling the associated data or providing intelligent operation recommendation when the user switches to the target application program or executes the operation conforming to the prediction. The system comprises a user behavior monitoring module, an intention prediction engine module, a data extraction and generation module and an intelligent interaction execution module. According to the method, cross-application data transmission and process connection are intelligently assisted by predicting the user intention, repeated operation of the user is reduced, and the working efficiency and the cloud desktop use experience are improved.

Description

Technical Field

[0001] The present invention relates to the fields of cloud computing technology, artificial intelligence technology and human-computer interaction technology, and in particular to a method and system for realizing intelligent flow of data and workflow across applications based on user intent prediction in a cloud desktop environment. Background Art

[0002] Cloud desktops provide users with a flexible and secure work environment, where they often need to use multiple applications simultaneously to complete complex tasks. For example, a user might need to extract information from an email, enter it into a customer relationship management (CRM) system, generate a report based on the CRM data, and finally send it to colleagues via instant messaging. Traditionally, users have to frequently switch between different applications, manually copying, pasting, searching, and entering data. This process is not only inefficient and error-prone, but also increases cognitive load and operational fatigue.

[0003] Existing technologies include some operating systems or productivity tools that facilitate data sharing between applications, such as universal clipboard functionality or simple script automation tools. However, these tools often lack a deep understanding of and ability to predict users' true work intentions. They are unable to proactively and intelligently assist users in efficiently transferring data and task flows between different applications, making them unable to meet users' demands for a highly intelligent and personalized office experience. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the purpose of the present invention is to provide a method and system for intelligent cross-application data and process flow based on user intention prediction in a cloud desktop environment, aiming to intelligently learn user behavior and predict user intentions, thereby realizing automatic pre-filling of data across applications, intelligent recommendation, and automation or semi-automation of part of the work process, thereby significantly improving the work efficiency and fluency of users in the cloud desktop environment.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for intelligently transferring cross-application data and processes based on user intention prediction in a cloud desktop environment comprises: monitoring a user's cross-application operation behavior sequence and the contextual data involved in the operation in the cloud desktop environment; analyzing and predicting the user's most likely current work intention and subsequent operation steps using a preset intention learning model based on the operation behavior sequence and contextual data; extracting or generating associated data from a source application based on the predicted user work intention and subsequent operation steps; and automatically pre-filling the associated data into corresponding fields of the target application when the user switches to a target application or performs an operation consistent with the prediction, or providing the user with intelligent operation recommendations based on the associated data.

[0007] Preferably, the cross-application operation behavior sequence includes: application startup and switching order, file opening and saving records, data copy and paste behavior, and interaction history of specific interface elements.

[0008] Preferably, the context data includes: the type of the currently active application, the window title, the document content fragment, the email subject, and the schedule information in the user's calendar.

[0009] Preferably, the intention learning model adopts a machine learning algorithm to establish a typical user workflow pattern library by learning a large number of user behavior samples, and performs probabilistic reasoning based on real-time behavior data to predict intentions.

[0010] Preferably, the extraction or generation of the associated data includes: identifying and extracting key information entities from interface elements, document content or structured data of the source application, or generating new data items based on the predicted intent combination.

[0011] Preferably, the intelligent operation recommendation includes: recommending possible target applications or common operations for the data after the user selects specific data; or recommending data sources that may need to be input when the user enters a specific application interface.

[0012] Preferably, the method further includes: for cross-application operation processes that are frequently executed by users and have a fixed pattern, after obtaining user confirmation, recording them and converting them into automated micro-process scripts that can be triggered with one click or under conditions.

[0013] Preferably, the method further includes: providing an enhanced clipboard function, which can intelligently manage and recommend multiple copies of historically copied data content based on the context of the currently active application and the characteristics of the target input field, and support intelligent conversion of data formats.

[0014] A cross-application data and process intelligent flow system based on user intention prediction in a cloud desktop environment, comprising: a cloud server, a service terminal and a cloud desktop deployed on the service terminal; an intelligent flow assistance module is integrated in the cloud desktop; the intelligent flow assistance module comprises: a user behavior monitoring module, an intention prediction engine module, a data extraction and generation module and an intelligent interaction execution module; the user behavior monitoring module is used to capture the user's cross-application operation behavior sequence and related context data in the cloud desktop environment; the intention prediction engine module has a built-in intention learning model, which is used to analyze the data collected by the user behavior monitoring module and predict the user's current work intention and subsequent possible operations; the data extraction and generation module is used to extract or combine and generate related data required to complete subsequent operations from the source application according to the prediction results of the intention prediction engine module; the intelligent interaction execution module is used to automatically fill the related data provided by the data extraction and generation module into the corresponding position of the target application when the user switches to the target application or performs a specific operation, or present relevant intelligent operation recommendations to the user interface.

[0015] Preferably, the intent learning model is deployed on a cloud server, and the model is trained and updated by continuously collecting and analyzing desensitized user behavior data from one or more service terminals.

[0016] Preferably, the intelligent flow assistance module also includes a micro-process management module for recording, storing and executing user-defined or system-recommended automated cross-application micro-process scripts.

[0017] This invention discloses the following technical effects: By intelligently learning users' cross-application operating habits and workflows in a cloud desktop environment, the invention can relatively accurately predict users' operational intentions, proactively transferring required data between different applications, recommending subsequent operations, or executing automated process segments. This effectively reduces repetitive tasks such as manual switching between applications, data search, and copying and pasting, shortens task completion time, reduces operational errors, significantly improves the work efficiency and intelligent experience of cloud desktop users, and makes workflows more coherent and smooth. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1A flowchart of a method for intelligently transferring cross-application data and processes based on user intent prediction provided by an embodiment of the present invention;

[0020] Figure 2 Module diagram of the cross-application data and process intelligent flow system based on user intent prediction provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] The purpose of the present invention is to provide a method and system for intelligent cross-application data and process flow based on user intention prediction in a cloud desktop environment. By predicting user intention, cross-application data flow and part of the workflow are automatically or semi-automatically processed to improve user work efficiency.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] Figure 1 A flow chart of a method for intelligently transferring cross-application data and processes based on user intent prediction provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the method provided by the present invention includes:

[0025] Step S101: Monitor user cross-application operation behavior and context. The system's built-in user behavior monitoring module continuously records a series of user operations in the cloud desktop environment. For example, a user opens the email application, reads an email containing an order number and customer name, copies the order number, and then switches to the CRM application. The monitored content includes application startup sequence, window focus switching, keyboard input (after desensitization), mouse clicks, data copy and paste events, as well as contextual information such as the current application window title and partially visible text content.

[0026] Step S102: Analyze the behavior sequence and predict user intent. The intent prediction engine module receives the monitored behavior sequence and contextual data. For example, based on the behavior sequence of "reading the order email -> copying the order number -> opening the CRM application," combined with the characteristics of the CRM application, the intent learning model might predict that the user's next intent is to "query or enter relevant information about this order in the CRM." By learning from a large number of similar behavior patterns, the model can identify common user workflow segments.

[0027] Step S103: Extract or generate related data based on the predicted intent. In the above example, the data extraction and generation module identifies the copied "order number" as key related data. If the predicted intent is to create a new entry, it may also attempt to extract other relevant information such as "customer name" from the email context. In some cases, if it predicts that the user may need to perform calculations or conversions based on the source data (such as date format conversion or currency conversion), the module can also perform these operations and generate new data.

[0028] Step S104: Execute intelligent interaction (pre-population or recommendation) in the target application. When the user actually switches to the CRM application and clicks the "Order Query" input box, the intelligent interaction execution module can automatically fill the "Order Number" extracted in step S103 into the input box. Alternatively, when the user first opens the CRM application, the system pops up a small prompt: "Do you want to query the order number XXX you just copied?" If the user selects a piece of text (such as a to-do item in a meeting minutes), the system may recommend "Create a calendar reminder" or "Send to a task management application."

[0029] Step S105 (optional): Record and execute automated micro-processes. If a user frequently performs a series of fixed cross-application operations (for example, selecting the latest sales report from the download folder -> opening the BI application -> importing the report -> generating a monthly chart), after the user completes it manually several times, the system can prompt the user whether to save this process as a "one-click generation of monthly reports" micro-process. After the user agrees, the micro-process management module will record the operation sequence (including application path, file name rules, interface operations, etc.), and the user can subsequently execute it automatically or semi-automatically by clicking a button or a specific trigger condition (such as the completion of the download of a new report).

[0030] Step S106 (optional): Manage and recommend enhanced clipboard content. When a user copies data, it not only goes into the system clipboard but is also recorded by the intelligent transfer assistance module (with a configurable maximum number and expiration date). When the user is ready to paste, if the system detects that the input field of the current target application has specific format requirements or contextual cues, the enhanced clipboard can prioritize the most matching historical copy items or provide simple format conversion options.

[0031] refer to Figure 2 , a cross-application data and process intelligent flow system based on user intent prediction in a cloud desktop environment. It is integrated into the user's cloud desktop environment. Its core intelligent flow auxiliary modules include:

[0032] User Behavior Monitoring Module: Responsible for comprehensively capturing various user operation events and related contextual information within the cloud desktop. This module acts as the system's sensory system, providing raw data for subsequent intelligent analysis. It can be a lightweight agent that records application launches, window switches, focus changes, clipboard operations, keyboard and mouse input (key content can be desensitized), and partial application content (such as obtained through the Accessibility API).

[0033] Intent Prediction Engine: This is the "brain" of the system. It receives data from the User Behavior Monitoring Module and uses built-in intent learning models (such as sequence analysis-based recurrent neural networks (RNNs), Transformer models, or simpler rule-based reasoning and pattern matching engines) to analyze user behavior patterns and predict the user's most likely current task intent (such as "find customer information," "create a meeting," "process an order," etc.), as well as the likely next action or target application. This model can be trained and updated in the cloud to adapt to a wider range of user behaviors.

[0034] Data Extraction and Generation Module: Based on the output of the intent prediction engine, this module intelligently extracts key data needed to complete the next step from the source application of the user's current or previous action. For example, it can extract the meeting time, location, and participants from an email body; or extract the product name and price from a web form. It may also cleanse, format, or combine the extracted data to generate new data as needed.

[0035] Intelligent Interaction Execution Module: This module bridges system intelligence and user actions. When the predicted user intent matches the user's actual action (such as opening a new app or activating an input box), the module takes action:

[0036] Automatic pre-population: Automatically fill in the corresponding fields of the target application with the extracted or generated contextual data.

[0037] Smart recommendations: Recommend possible operations or data input items to users on the interface through non-intrusive methods (such as small bubble prompts, right-click menu enhancements).

[0038] Call microflow: If the current intent matches a defined microflow and the trigger conditions are met, the user can be prompted or the microflow can be automatically executed.

[0039] (Optional) Micro-Process Management Module: This module is responsible for creating, storing, editing, and executing user-defined or system-recommended automated micro-process scripts. Users can define their own frequently used cross-application processes by simply recording or dragging and dropping.

[0040] (Optional) Enhanced Clipboard Manager: Provides more powerful features than the standard clipboard, such as history, content preview, format conversion suggestions, and context-based paste recommendations.

[0041] These modules work together to make cross-application operations in the cloud desktop smoother and more efficient. For example, when a user copies text from a document containing an address and switches to the map app, the system can predict that the user intends to find the address and automatically fill it into the map app's search box.

[0042] The beneficial effects of the present invention are as follows: By deeply learning and understanding the user's work habits and intentions in a cloud desktop environment, the present invention can proactively predict and assist users in completing cross-application data transfer and process integration, greatly reducing the user's manual switching between different software, data search, copy and paste, and other tedious operations. This not only significantly improves work efficiency and reduces the probability of error, but also makes the entire workflow more natural and coherent, optimizing the user's cloud desktop experience. For enterprises, this means that employees can devote more energy to the core business itself, rather than wasting time on repetitive auxiliary operations.

[0043] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0044] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for intelligent cross-application data and process flow based on user intent prediction in a cloud desktop environment, characterized by: The following steps are involved: a monitoring user in the cloud desktop environment cross-application operation behavior sequence and the context data associated with the operation behavior sequence; b. Based on the operation behavior sequence and the context data, using a preset intention learning model to analyze and predict the user's most likely current work intention and subsequent operation steps; c. extracting or combining related data from source applications related to the sequence of operational behaviors based on the predicted user work intention and the subsequent operational steps; d. When the user switches to a target application related to the predicted subsequent operation step or performs an operation consistent with the prediction, the associated data is automatically pre-filled into the corresponding user interface field of the target application, or an intelligent operation recommendation based on the associated data is provided to the user.

2. The method according to claim 1, characterized in that The cross-application operation behavior sequence in step a includes: the user's launch and switching order of applications, the user's data copy and paste behavior, or at least one of the user's interaction history with specific user interface elements; the intention learning model in step b uses a machine learning algorithm to analyze the operation behavior sequence and context data to predict the work intention.

3. The method according to claim 1, characterized in that The method also includes: for cross-application operation processes that are frequently executed by users and have fixed patterns, after obtaining user confirmation, recording them and converting them into automated micro-process scripts that can be triggered by the user with one click or automatically triggered by the system according to specific conditions.

4. The method according to claim 1, wherein The method further includes providing an enhanced clipboard function, wherein the enhanced clipboard can intelligently manage and recommend to the user multiple copies of historically copied data contents according to the context of the currently active application and the characteristics of the target input field.

5. A cross-application data and process intelligent flow system based on user intention prediction in a cloud desktop environment, characterized by: It is deployed in a cloud desktop environment and includes: a. a user behavior monitoring module, suitable for capturing the user's cross-application operation behavior sequence and related contextual data in the cloud desktop environment; b. an intention prediction engine module with a built-in intention learning model, suitable for analyzing the data collected by the user behavior monitoring module and predicting the user's current work intention and subsequent possible operations; c. a data extraction and generation module, suitable for extracting or combining the associated data required to complete subsequent operations from the source application related to the user's operation behavior based on the prediction results of the intention prediction engine module; d. an intelligent interaction execution module, suitable for automatically filling the associated data provided by the data extraction and generation module into the corresponding user interface position of the target application when the user switches to the target application related to the predicted subsequent operation or performs a specific operation, or presenting relevant intelligent operation recommendations to the user interface.

6. The system according to claim 5, characterized in that The system further comprises: a micro-process management module adapted to record, store, manage and execute automated cross-application micro-process scripts that are user-defined or generated by the system based on user behavior recommendations.

7. The system according to claim 5, characterized in that The intent learning model is deployed on the cloud server side, and the model is trained and updated by continuously collecting and analyzing desensitized user behavior data from one or more service terminals; alternatively, the intent learning model is deployed locally on the service terminal, and personalized learning and prediction are performed using local data.

Images