Search

KR-20260063902-A - AI ASSISTANT POP-UP PROVISION SYSTEM AND METHOD

KR20260063902AKR 20260063902 AKR20260063902 AKR 20260063902AKR-20260063902-A

Abstract

The present invention relates to an AI assistant popup providing system and method for recommending functions optimized for an ongoing task by learning a user's work patterns through time-series analysis of user input data. The AI assistant popup providing system comprises: an AI assistant popup providing server that collects data on a series of user inputs entered during a task using a program running on at least one user terminal, learns the user's work patterns and preferences through time-series analysis of the collected user input data, and, when a trigger execution for an AI assistant popup is confirmed, generates and provides an AI assistant popup including recommendation functions related to the currently ongoing task based on the learned user's work patterns and preferences; and a data storage server that classifies and stores the collected user input data and the learned user's work patterns and preferences by user.

Inventors

  • 이준석
  • 박철현

Assignees

  • (주)인바이즈

Dates

Publication Date
20260507
Application Date
20241031

Claims (13)

  1. An AI assistant popup providing server that collects data on a series of user inputs entered during a task using a program running on at least one user terminal, learns the user's task patterns and preferences through time-series analysis of the collected user input data, and, when a trigger execution for an AI assistant popup is confirmed, generates and provides an AI assistant popup including a recommendation function related to the currently ongoing task based on the learned user's task patterns and preferences; and An AI assistant popup providing system comprising: a data storage server that classifies and stores the collected user input data and learned user work patterns and preferences by user.
  2. In paragraph 1, An AI assistant popup providing system characterized by the above-mentioned AI assistant popup providing server recognizing a focused program and a text drag for a currently running task from at least one user terminal, checking whether a trigger for an AI assistant popup is executed, and if the trigger execution of the AI assistant is confirmed, generating an AI assistant popup that includes a recommendation function for the recognized text drag among the functions of the focused program and providing it to the at least one user terminal.
  3. In paragraph 2, The above-mentioned collected user input data includes browser data including URL conversion history, dwell time, search keywords, and bookmarks, and desktop data including shortcuts, mouse movements, executed programs, function usage history, and focus conversion history. The AI assistant popup providing server described above is characterized by applying a Gated Recurrent Unit (GRU) algorithm and a Contents Based Filtering (CBF) technique to collected user input data to analyze and learn each user's work patterns and preferences in a time series, and predicting a recommendation function based on the learned user's work patterns and preferences to generate an AI assistant popup with a user-customized UI/UX.
  4. In paragraph 3, The above AI assistant popup providing server collects user input data including at least URL conversion history, focus conversion history, and function usage history from multiple user terminals, processes the collected user input data into work environment bundle data, and classifies running programs according to work groups based on the processed work environment bundle data. An AI assistant popup providing system characterized by activating programs classified as having the same work environment as the currently focused program and deactivating programs classified as having a different work environment to optimize the current work environment, and adding metadata to and saving the work environment information of the deactivated programs.
  5. In paragraph 4, The AI assistant popup is characterized by comprising at least a first area for displaying a search target selected by dragging text, a second area for recommending a search engine, a third area for providing AI functions including the dictionary meaning and translation of the search target, a fourth area for including a text-based AI prompt recommendation type and a generative AI question input window for direct querying, and a fifth area for providing a widget including additional functions based on a focused program.
  6. In paragraph 5, An AI assistant popup providing system characterized by the above trigger being performed by a user input device and executed through a preset gesture, movement to a preset specific area within a browser, or pressing an input button configured on the user input device.
  7. In a method for providing an AI assistant popup using an AI assistant popup providing system, Step 1: Continuously monitoring a working environment running on a user terminal providing a computing environment to collect user input data for a series of user inputs entered during the work; Step 2, inferring the current working environment based on active program information; Step 3 for predicting recommendation features based on the inferred current working environment and collected user input data; and A method for providing an AI assistant popup, comprising the step of: 4, generating an AI assistant popup including a predicted recommendation function and providing it to the user terminal when trigger execution is confirmed.
  8. In paragraph 7, the above 2 steps are, A step of recognizing a program running on the above user terminal; A step of analyzing the collected user input data and performing calculations on it as work environment bundled data; A step of classifying the above-mentioned running program based on computationally processed work environment bundle data; A step of optimizing the current working environment by activating programs classified as being in the same working environment as the currently focused program and deactivating programs classified as being in a different working environment; and A method for providing an AI assistant popup, characterized by further including the step of adding metadata to and saving the working environment information of a disabled program.
  9. In paragraph 8, the above three steps are, A step of extracting user work patterns and preferences by performing time series analysis on pre-stored user input data using the GRU (Gated Recurrent Unit) algorithm and Contents Based Filtering (CBF) technique; A step of building a learning model that predicts recommendation functions based on extracted user activity patterns and preferences; and A method for providing an AI assistant popup, characterized by further including the step of predicting a recommendation function by learning personalized user work patterns through a learning model built based on currently collected user input data and an inferred current work environment.
  10. In paragraph 9, the above-mentioned four steps are, A step of recognizing an action by dragging text during the execution of a focused program among currently active programs; A step of checking whether the operation by dragging the text above is a trigger execution; and A method for providing an AI assistant popup, characterized by further including the step of, if the result of verification is that the trigger is executed, generating and providing a user-customized AI assistant popup based on a predicted recommendation function.
  11. In Clause 10, prior to the above Step 1, A step of linking with the above user terminal and cloud; A step of loading files from linked user terminals and the cloud, performing indexing, and storing them on a data storage server; and It further includes the step of continuously monitoring the user terminal and the cloud regarding changes to the stored file and reflecting them in the stored file—changes being additions, modifications, or deletions to the file—; The above 4th step is a step of optimizing query splitting and search results through a RAG (Retrieval-Augmented Generation) technique upon a search query request via the AI assistant popup, and generating a response based on an LLM (Large Language Model) for the search results to perform file search; and A method for providing an AI assistant popup, characterized by further including the step of providing file exploration results through the AI assistant popup.
  12. In Paragraph 7, A method for providing an AI assistant popup, wherein the AI assistant popup described above comprises at least a first area for displaying a search target selected by dragging text, a second area for recommending a search engine, a third area for providing AI functions including the dictionary meaning and translation of the search target, a fourth area including a text-based AI prompt recommendation type and a generative AI question input window for direct querying, and a fifth area for providing a widget including additional functions based on a focused program.
  13. In Paragraph 7, A method for providing an AI assistant popup, characterized in that the above trigger is performed by a user input device and is executed through a preset gesture, movement to a preset specific area within a browser, or pressing an input button configured on the user input device.

Description

AI Assistant Pop-Up Provision System and Method The present invention relates to artificial intelligence software technology, and more specifically, to an AI assistant popup providing system and method for recommending functions optimized for a currently executing task by learning a user's work patterns through time-series analysis of user input data generated in a computing environment. In a computer environment, an average of 3 to 5 SaaS programs are used to perform a single task, and as an average of over 200 shortcuts are used for each software, it is difficult to recognize these shortcuts during work. Since the development purpose of each software is different, it is difficult to standardize usability, such as UI, UX, and shortcuts, which leads to a decrease in work convenience. In addition, as illustrated in FIGS. 9(a) to (c), resources are distributed among software by job to achieve a single objective in a computer environment. For example, as shown in FIG. 9(a), for business-related jobs, resources may be distributed among software classified as data research, file exploration, and document creation; as shown in FIG. 9(b), for creator jobs, resources may be distributed among software classified as file exploration, 2D source creation, and video editing; and as shown in FIG. 9(c), for game streamer jobs, resources may be distributed among software classified as games, streaming, and broadcast management. In other words, as the amount of software used increases, resources become more decentralized, which leads to a problem where work efficiency is reduced as a significant amount of time is spent searching for and collecting resources during work. This inefficient process results in losses in time and cost. FIG. 1 is a schematic diagram illustrating an AI assistant popup providing system according to one embodiment of the present invention. FIG. 2 is a flowchart illustrating a method for providing an AI assistant popup according to an embodiment of the present invention. FIG. 3 is a flowchart for specifically explaining a method to optimize the current working environment of S120 shown in FIG. 2. Figure 4 is a drawing illustrating an example of a work environment optimized by the method shown in Figure 3. FIG. 5 is a flowchart for specifically explaining how to predict the operation pattern and recommendation function of S130 shown in FIG. 2. FIG. 6 is a flowchart for specifically explaining how to execute the AI assistant popup of S140 shown in FIG. 2. FIG. 7 is a drawing illustrating the UI of an AI assistant popup according to one embodiment of the present invention. FIG. 8 is a flowchart illustrating a file search method using an AI assistant popup according to one embodiment of the present invention. Figure 9 is a diagram illustrating resource distribution by job in a conventional computing environment. Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement the invention. However, since the description of the present invention is merely an example for structural or functional explanation, the scope of the present invention should not be interpreted as being limited by the embodiments described in the text. That is, since the embodiments are subject to various modifications and may take various forms, the scope of the present invention should be understood to include equivalents capable of realizing the technical concept. Furthermore, the objectives or effects presented in the present invention do not imply that a specific embodiment must include all of them or only such effects; therefore, the scope of the present invention should not be understood as being limited by them. Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. FIG. 1 is a schematic diagram illustrating an AI assistant popup providing system according to one embodiment of the present invention. Referring to FIG. 1, the AI assistant popup providing system according to the present invention may include an AI assistant popup providing server (100) and a data storage server (200). Here, the AI assistant popup providing server (100) is connected to at least one user terminal (10) via a network. The user terminal (10) is a computing device equipped with a user input device (11) for user input, and may be a desktop (PC) or a laptop, and may include at least one processor, memory, output device, display device, communication module, etc., although not shown. Such a user terminal (10) may be equipped with a plurality of programs for performing a user's task (TASK). In one embodiment, the program (used equally as application program, software, or application) may be pre-installed or stored at the time of manufacturing the user terminal (10), or may be installed based on data received from an external source when used later. The user input device (11) may be con