Data processing method and electronic equipment

Abstract

The invention discloses a data processing method and electronic equipment, and the method comprises the steps: determining a knowledge graph of a target user and attribute information of first input data in response to the obtained first input data input to a target application by the target user; generating multiple groups of guide data for the first input data based on the knowledge graph and the attribute information, wherein the multiple groups of guide data have a target logic relationship; and generating a corresponding target response result based on feedback data of the target user to the multiple groups of guide data.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a data processing method and electronic equipment. Background Art

[0002] Driven by information technology, various intelligent systems are gradually being integrated into people's learning, skills development, and other knowledge acquisition scenarios, playing a positive role in knowledge dissemination and improving learning efficiency. However, shortcomings remain in the key link of interactive communication between intelligent systems and users. Summary of the Invention

[0003] The technical solutions provided in this application are as follows:

[0004] The first aspect of the present application provides a data processing method, comprising:

[0005] In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data;

[0006] generating a plurality of sets of guidance data for the first input data based on the knowledge graph and the attribute information, wherein the plurality of sets of guidance data have a target logical relationship between them; and

[0007] Generate corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data.

[0008] Determining the target user's knowledge graph and the attribute information of the first input data includes:

[0009] Identifying semantic attributes and / or category attributes of target content in the first input data, and determining attribute information of the first input data based on the semantic attributes and / or category attributes and the context of the first input data;

[0010] Obtain historical learning data of the target user, and determine a knowledge graph of the target user based on the historical learning data and attribute information of the first input data.

[0011] The generating of multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information includes:

[0012] Splitting the first input data into a plurality of knowledge point data based on the knowledge graph and the attribute information;

[0013] Based on the various pieces of knowledge point data and the logical relationships therebetween, a number of heuristic questions for the various pieces of knowledge point data are generated as multiple groups of guiding data for the first input data.

[0014] The generating of multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information includes:

[0015] generating and processing each content data in the first input data based on the attribute information to obtain a plurality of guide data for each content data;

[0016] Filtering out a plurality of target guidance data that matches the knowledge graph of the target user from the plurality of guidance data; and

[0017] The target guidance data are sorted based on the knowledge graph as multiple groups of guidance data for the first input data.

[0018] Generating corresponding target response results based on the target user's feedback data on the multiple sets of guidance data includes:

[0019] The guidance data of the target group in the guidance data is updated based on the evaluation result of the feedback data.

[0020] The updating of the guidance data of the target group in the guidance data based on the evaluation result of the feedback data includes at least one of the following:

[0021] If the evaluation result indicates that the target user's acceptance of the guidance data is lower than a first threshold, updating the logical relationship between the guidance data of the target group in the guidance data;

[0022] If the evaluation result indicates that the target user's acceptance of the guidance data is greater than a second threshold, updating the logical order of the guidance data of the target group in the guidance data;

[0023] When the evaluation result indicates that the target user's acceptance of the guidance data is greater than a third threshold, the guidance data is regenerated by introducing target knowledge points based on the knowledge graph.

[0024] Generating corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data further includes at least one of the following:

[0025] Generate corresponding target knowledge point explanation content and / or application examples based on the feedback data;

[0026] A learning suggestion matching the target user is generated based on the feedback data, and the learning suggestion is associated with the knowledge graph of the target user.

[0027] Also includes:

[0028] guiding the target user to learn knowledge data related to the first input data based on the guidance data;

[0029] A corresponding target feedback interface is generated and updated based on the feedback data, and the target feedback interface provides operation controls or recommended content for updating the knowledge graph of the target user.

[0030] Also includes:

[0031] Obtaining interaction data between a target user and the target application, updating the target user's knowledge graph based on the interaction data, and / or generating corresponding positive evaluation content based on the interaction data;

[0032] The target application is an application that can call the at least one processing model to execute a target processing function, and the interaction data includes data as input to the processing model and feedback data for the inference result output by the processing model.

[0033] Another aspect of the present application provides an electronic device, including at least one target application and at least one processing model capable of running on the target application, wherein the processing model can be called by the target application to perform at least one of the following:

[0034] In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data;

[0035] generating a plurality of sets of guidance data for the first input data based on the knowledge graph and the attribute information, wherein the plurality of sets of guidance data have a target logical relationship between them; and

[0036] Generate corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the originals and elements are not necessarily drawn to scale.

[0038] Figure 1 A flowchart of a data processing method provided in Example 1 of the present application;

[0039] Figure 2 A flowchart of a data processing method provided in Example 2 of the present application;

[0040] Figure 3A flowchart of a data processing method provided in Example 3 of the present application;

[0041] Figure 4 A flowchart of a data processing method provided in Example 4 of the present application;

[0042] Figure 5 A flowchart of a data processing method provided in Example 5 of the present application;

[0043] Figure 6 A flowchart of a data processing method provided in Example 8 of the present application;

[0044] Figure 7 A flowchart of a data processing method provided in Example 9 of the present application;

[0045] Figure 8 Another flowchart of a data processing method provided in Example 9 of the present application;

[0046] Figure 9 An interactive schematic diagram provided for this application;

[0047] Figure 10 A schematic diagram of the structure of a data processing device provided in this application. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present application in conjunction with the accompanying drawings. The terms used in the implementation methods of the present application are only used to explain the specific embodiments of the present application and are not intended to limit the present application.

[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0050] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, and this is merely a way of distinguishing the objects of the same attributes when describing them in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0052] Reference Figure 1 , is a flow chart of a data processing method provided in Example 1 of the present application, such as Figure 1 As shown, the method may include but is not limited to the following steps:

[0053] Step S101: In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data.

[0054] When users interact with the target application, they can choose from a variety of input methods to meet user needs in different scenarios (such as answering questions or submitting learning content). Specifically, they can include:

[0055] Text input: Users input content in natural language to ask questions. Whether it is a short reply or a long description, the target application can parse and understand different forms of text expression.

[0056] Image input: Users can upload image data such as handwritten questions and charts (such as geometric figures, flow charts, etc.). For example, the user triggers the "Upload Image" operation entry in the target application and selects the image file to be uploaded from the local storage device. In addition to selecting images from the local storage device, the target application can also support direct photo uploading. For example, near the "Upload Image" operation entry, a "Photo" option can be set. After the user clicks this option, the user can take a photo and upload the math problem he or she has handwritten in the homework book, or take a photo and upload pictures containing complex geometric figures. The target application can receive and process this image information.

[0057] Formula input: Supports users to input complex mathematical formulas using standard mathematical input methods (such as LaTeX).

[0058] Audio input: Users can submit information through voice input. For example, in language learning scenarios, users can read aloud an English article. The target application can receive the audio data and then perform speech recognition on the audio, converting it into text for grammatical analysis and semantic understanding. In meeting recording scenarios, users can also directly record the meeting audio, and the application will process the audio and extract key information.

[0059] Video input: Users can upload video files. For example, in a physics experiment course, students can record and upload a video of themselves conducting a physics experiment. The target application can then analyze the video frame by frame, identifying key steps, operating techniques, and experimental phenomena. In the education field, the application can also extract and analyze the teaching content of the teaching videos uploaded by teachers, assisting in the organization and optimization of teaching resources.

[0060] In the case where the target input data includes multiple data in different modalities, the target input data may be processed in a unified format. Specifically, this may include:

[0061] Text processing: Based on natural language processing technology (such as the multimodal large language model (LLM)), text input is subjected to grammatical analysis, keyword extraction, and semantic understanding.

[0062] Multimodal LLM has powerful natural language processing capabilities and can accurately analyze the structure and semantic information of text, thereby providing accurate and comprehensive information for subsequent analysis. Compared with traditional methods, it can better handle complex contexts and diverse text expressions.

[0063] Image processing: Multimodal LLM is used to recognize text in images (such as handwritten homework content), while image analysis technology is used to interpret mathematical symbols, tables, and charts in images.

[0064] Formula recognition: Using the mathematical formula recognition technology of multimodal LLM, complex mathematical formulas are identified and converted into standardized expressions to facilitate subsequent calculations and analysis.

[0065] Audio processing: First, use speech recognition technology to convert audio data into text, and then perform subsequent operations according to text processing methods, such as grammatical analysis and semantic understanding, to extract valuable information from the audio.

[0066] Video processing: Analyze the video frame by frame, extract key frame images, extract the audio part of the video for audio processing, and use image recognition technology to identify and process the key frame images. Combine the image and audio processing results to fully extract the effective information in the video.

[0067] From a content perspective, the first input data can be a subject matter question, such as in mathematics, involving complex geometric proofs and function calculation problems; in physics, including application questions on the principles of mechanics and understanding electromagnetic concepts. The first input data can also be a general consultation question, such as a user asking about the day's weather, surrounding traffic conditions, the opening hours of a certain tourist attraction, and other life-related questions. In addition, the first input data can also be a question statement in a professional field, such as in the medical field, a patient's consultation on the diagnosis of a disease or a doctor's research and discussion on treatment plans; in the legal field, it can be a party's need to interpret legal provisions or a lawyer's analysis and judgment of a case.

[0068] In this embodiment, the knowledge graph can reflect the target user's knowledge level and record the knowledge mastered by the target user in detail in a structured manner. Taking the learning scenario as an example, the knowledge graph will record the students' mastery of knowledge points in various subjects. Assuming that students have a solid grasp of triangle-related knowledge, such as the interior angle theorem of a triangle, the Pythagorean theorem and other properties and theorems in their past mathematics learning, the target application can mark and record the corresponding position in the knowledge graph. In terms of quadrilateral knowledge, if students have many questions about the classification of quadrilaterals (such as parallelograms, trapezoids, etc.), judgment conditions (such as the judgment theorem of parallelograms), etc., the knowledge graph will also accurately reflect this situation. Based on the information recorded by the knowledge graph, the target application can accurately judge that the student is currently in the learning and exploration stage of quadrilateral knowledge, thereby providing a strong basis for subsequent data processing and guidance.

[0069] The attribute information of the first input data may be used to describe at least one attribute of the first input data, such as the field, discipline, or industry to which the first input data belongs, or the difficulty attribute or background (context situation) possessed by the first input data.

[0070] Step S102: Generate multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information, and the multiple sets of guidance data have a target logical relationship between them.

[0071] Target logical relationships can include, but are not limited to, progressive relationships from easy to difficult and from easy to complex. Specifically, multiple sets of guiding data guide the target user's thinking in a sequential order of increasing difficulty and difficulty. For example, a target user uploads an image containing the following question: To create a "beautiful countryside" suitable for human habitation, Liushu Village has decided to build a 200cm-wide circular path around a small circular lake with a diameter of 30m. What is the area of ​​this path? The target application determines the knowledge graph and attribute information based on the image uploaded by the target user. Based on this knowledge graph and attribute information, the target application first asks a basic question, "1. Look carefully, how many circles are there in this question?" to guide the target user to initially observe the geometric structure of the question. It then asks, "2. Have you learned which formula to use to calculate the area of ​​a circle?" to remind the target user of this formula. Finally, it asks, "3. Given the diameter and path width, can you calculate the radius of each circle?" to gradually guide the target user to the key steps of solving the problem. This progressive relationship allows target users to gradually master problem-solving ideas and avoid learning obstacles caused by a sudden increase in difficulty.

[0072] Target logical relationships can also include a hierarchical structure consisting of root nodes and leaf nodes. The root node's guiding data is typically a core question directly related to the first input data, and subsequent guiding data expands around this core question and becomes increasingly detailed. For example, when exploring questions related to biological evolution, the root node question, "Please explain what are the main driving forces of biological evolution?" serves as the starting point for the entire guiding system, providing direction for subsequent analysis. Subsequent guiding data, such as "How does natural selection play a role in biological evolution?" and "What are the effects of gene mutations on biological evolution?", provide in-depth analysis of different driving forces. Each sub-question can be further refined. For example, for "How does natural selection play a role in biological evolution," further questions can be asked, such as "Which characteristics of organisms does natural selection affect?" and "How does the mechanism of natural selection vary under different environments?" This hierarchical structure is like a large tree, from the trunk to the branches and leaves, helping target users to fully and systematically understand the problem and form a complete knowledge system.

[0073] Target logical relationships can also include: logical relationships from large to small. Taking the analysis of a historical event as an example, the first set of guiding data, "What was the historical background of the historical event and what overall impact did it have on society at the time?", guides students to grasp the overall picture of the event from a macro perspective; the second set of guiding data, "Who were the main participants in the event, and what were their positions and motivations?", focuses the perspective on the key figures and major contradictions in the event; the subsequent guiding data, "In what context was a key decision made, and what role did it play in the development of the event?", delves into the specific details of the event. This logical relationship from large to small is like observing things through a telescope, first seeing the overall outline and then gradually focusing on the details. It helps target users to deeply understand the problem from both macro and micro levels and deepen their knowledge.

[0074] In this embodiment, it is possible but not limited to calling at least one first processing model based on the knowledge graph and attribute information to perform generation processing on the first input data to generate multiple groups of guidance data for the first input data.

[0075] The output method of multiple sets of guidance data is highly flexible and can be selected according to actual needs.

[0076] For example, the target application can display all generated sets of guidance data to the target user at once. This output method allows the target user to independently choose the order in which to answer the guidance data according to their own learning pace and thinking process, gradually deepening their thinking and cultivating their independent thinking and problem-solving skills.

[0077] Of course, the target application can also choose to output one or more sets of guidance data at a time, wait for feedback from the target user, and then output the next set of guidance data based on the target user's feedback. This batch output method is more interactive and targeted. Based on the target user's feedback, the target application can accurately understand the target user's understanding of the knowledge points and thinking direction, and promptly adjust the difficulty and content of subsequent guidance data to ensure that the guidance process is more closely aligned with the target user's actual needs and improve guidance effectiveness.

[0078] Step S103: Generate corresponding target response results based on the target user's feedback data on the multiple sets of guidance data.

[0079] The target user's feedback data on multiple sets of guidance data can reflect the target user's understanding of the guidance data, depth of thinking, and knowledge needs during the interaction process.

[0080] In this embodiment, feedback data may include, but is not limited to, choices or responses to guidance data. For example, in the process of guiding a math word problem, the guidance data may be "Given that two sides of a triangle are 3 and 5, what is the possible length of the third side? (Give a range of values)." The user might respond with "The third side is greater than 2 and less than 8." This direct response to the guidance question can clearly demonstrate the user's mastery of the relevant knowledge and problem-solving approach.

[0081] Feedback data can also include, but is not limited to, new questions or reflections. As target users reflect on and respond to guidance data, they may develop new questions or ideas, leading to new questions. These new questions may be closely related to the knowledge points covered in the guidance data, or they may extend to other related or unrelated knowledge points.

[0082] New questions or reflections may be related to the guiding data. For example, in the history discipline, the guiding data revolves around the question "The impact of the Industrial Revolution on the social structure of Britain." During their responses, the target user may question the role of a specific technological invention during the Industrial Revolution (such as the steam engine) in driving changes in social structure. This may lead them to raise new questions closely related to the guiding data, such as "How did the invention of the steam engine specifically influence the formation and development of the British working class?"

[0083] New questions or thoughts may also be unrelated to the guiding data. For example, when discussing mathematical functions, a user suddenly asked, "Will knowledge about functions be used in quantum computing?" This cross-disciplinary question reflects the user's divergent and exploratory thinking.

[0084] In this embodiment, in response to obtaining first input data inputted by a target user into a target application, the target user's knowledge graph and attribute information of the first input data are determined, and multiple sets of guidance data specific to the first input data are generated based on the knowledge graph and attribute information. This ensures that the multiple sets of guidance data do not contain a large amount of content that exceeds the target user's knowledge level, thereby avoiding frustration caused by a large knowledge gap and preventing interaction obstacles caused by difficulty in understanding. Furthermore, the guidance data bureau will not introduce too many knowledge points unrelated to the key content in the first input data, reducing the user's time wasted on irrelevant information and thus improving learning or problem-solving efficiency.

[0085] Furthermore, the multiple sets of guidance data have a target logical relationship, which can guide the target user to explore the deeper logic and background knowledge of the problem, improving the target user's thinking expansion and comprehension ability. Furthermore, based on the target user's feedback data on the multiple sets of guidance data, corresponding target response results are generated, enhancing the stickiness and interactivity between the user and the target application.

[0086] In addition, it can understand various user input methods and provide a more diversified and flexible interactive experience.

[0087] As another optional embodiment of the present application, refer to Figure 2 , is a flow chart of a data processing method provided in Example 2 of the present application. This embodiment is mainly an implementation method of the above step S101. Figure 2 As shown, the following steps may be included but not limited to:

[0088] Step S1011: In response to obtaining first input data input by a target user into a target application, identify semantic attributes and / or category attributes of the target content in the first input data, and determine attribute information of the first input data based on the semantic attributes and / or type attributes and the context of the first input data.

[0089] Semantic attributes can be used to reveal the user intent represented by the first input data. For example, the first input data could be asking about a specific knowledge point, engaging in daily conversation, or issuing specific task instructions. By analyzing semantic attributes, we can preliminarily determine the target user's interaction purpose for this input.

[0090] In this embodiment, the semantic attributes of the target content may be identified by, but are not limited to, at least one of the following methods:

[0091] Step S11: Extract semantic information from the target content using techniques such as part-of-speech tagging, named entity recognition, and semantic role tagging to determine the semantic attributes of words or concepts. For example, part-of-speech tagging can determine whether a word is a noun, verb, or adjective, thereby understanding its semantic role in a sentence.

[0092] Step S12: The information in the semantic knowledge graph can be used to determine the semantic attributes of the entities and concepts of the target content in the target input data. For example, searching for the entity "apple" in the knowledge graph can obtain its related semantic attributes, such as "fruit", "red", "round", etc.

[0093] Step S13: For some complex semantic attributes, which are difficult to accurately obtain through automated technology, manual annotation may be required. Manual annotation can combine professional knowledge and contextual understanding to determine semantic attributes more accurately.

[0094] Category attributes can be used to reveal information such as the subject category (such as Chinese, history, geography, physics, chemistry, etc.) or application field (such as architecture, civil engineering, economics, politics, biomedicine, urban development and construction, automobile industry, PC industry, renewable energy, ESG, law, etc.) to which the first input data belongs.

[0095] In this embodiment, the category attribute may be identified by, but is not limited to, at least one of the following methods:

[0096] Step S21: Based on professional knowledge and experience, the target content in the target input data is analyzed and judged to determine its category attributes. For example, based on a subject knowledge base covering various disciplines, the input content is classified into the subject category, knowledge domain, etc., to determine whether it belongs to a subject category such as mathematics, physics, or chemistry, or a more specific sub-field.

[0097] Step S22: Determine the category attributes of the target content by extracting the category features of the target content according to the machine learning model.

[0098] To further improve the accuracy of identifying user intent, subject category, or application domain, attribute information of the first input data may be determined by comprehensively considering the context of the first input data. The attribute information of the first input data may be used to describe, but is not limited to, semantic attributes, category attributes, and difficulty attributes. Difficulty attributes may include, but are not limited to, the context of the problem (e.g., whether it is an algebraic problem, a geometry problem, or another type of problem) and the difficulty level of the problem.

[0099] The context of the first input data may include, but is not limited to, historical interaction data or historical task execution data, such as previous user questions and operation records in the target application. The context of the first input data may also include, but is not limited to, the source or origin of the question, such as when the first input data comes from a specific learning material or discussion topic.

[0100] By combining the context, the meaning and background of the first input data can be understood more comprehensively and deeply, thereby more accurately determining the attribute information of the first input data.

[0101] Step S1012: Obtain historical learning data of the target user, and determine the knowledge graph of the target user based on the historical learning data and attribute information of the first input data.

[0102] The target user's historical learning data comes from a wide range of sources. For example, it can be the target user's learning interaction data in the target application, or it can be records of learning obtained through other applications or devices and using other channels, such as the user's course learning records on other online learning platforms, learning records of reading e-books, etc.

[0103] In this embodiment, the learning records related to the knowledge point involved in the first input data in the historical learning data can be determined. Specifically, this may include: the target user's learning frequency of the knowledge point, such as how often the target user comes into contact with materials or courses related to the knowledge point. A high learning frequency may mean that the target user has a certain degree of attention to this knowledge point or has a learning need; the target user's learning depth, such as whether the target user has deeply studied the advanced content of the knowledge point and completed related extension exercises or projects; and the target user's learning outcomes, such as the target user's scores in knowledge point-related tests and the quality of homework completion.

[0104] Next, we can build the target user's knowledge graph through the above learning records that can reflect the target user's actual mastery of the knowledge points.

[0105] In this embodiment, the knowledge graph can also be updated based on the difficulty and importance of the knowledge points reflected in the first input data. For example, for difficult knowledge points, if the target user has little exposure to or poor mastery of these knowledge points in historical learning data, but has some exposure to these knowledge points in the first input data, the target application can mark the target user's potential weaknesses in these knowledge points in the knowledge graph. For more important knowledge points, the target application can focus on the user's learning status to ensure that the knowledge graph accurately reflects the target user's cognitive status on these key knowledge points.

[0106] The target user's knowledge graph can comprehensively and meticulously record the target user's mastery of various subjects or application fields, various knowledge points (such as different levels such as proficiency, basic mastery, and less understanding), learning progress (such as which stage of learning has been reached, whether the relevant learning tasks have been completed, etc.) and the relationship between knowledge (such as which knowledge points have logical connections, which knowledge points are the basis of other knowledge points, etc.).

[0107] In this embodiment, the introduction of context helps to identify the accuracy of user intention or subject category or application field, especially when faced with some semantically ambiguous or input data with multiple interpretations. The context can provide key clues to more accurately determine the attribute information of the first input data.

[0108] By collecting the target user's historical learning data and combining it with the attribute information of the first input data to determine the target user's knowledge graph, we can comprehensively and meticulously record the user's mastery of various disciplines or application fields, knowledge points, learning progress, and the relationships between knowledge. This enables the knowledge graph to truly and accurately reflect the user's cognitive state, providing the target application with a comprehensive perspective on the user.

[0109] Furthermore, updating the knowledge graph based on the difficulty and importance of knowledge points reflected in the first input data can promptly reflect changes in the user's learning process. For example, for knowledge points that are difficult and poorly mastered in previous learning but are involved in the current input, the user's weak points are marked, which helps the target application provide targeted learning suggestions and guidance to the user, helping the user to fill in knowledge gaps.

[0110] Focus on knowledge points with higher importance, ensure that the knowledge graph can accurately reflect the user's cognitive status on these key knowledge points, so that the target application can adjust the learning path and guidance strategy according to the user's learning focus and needs, and improve the user's learning effect and efficiency.

[0111] As another optional embodiment of the present application, refer to Figure 3 , is a flow chart of a data processing method provided in Example 3 of the present application. This embodiment is mainly an implementation method of the above step S102. Figure 3 As shown, the following steps may be included but not limited to:

[0112] Step S1021: split the first input data into multiple knowledge point data based on the knowledge graph and the attribute information.

[0113] There are certain logical relationships between the split knowledge point data. These logical relationships can be the order of knowledge, cause-effect relationship, inclusion relationship, etc.

[0114] For example, suppose a student inputs a mathematical geometry problem (i.e., one embodiment of the first input data) into a heuristic problem-solving agent application (i.e., one embodiment of the target application): "In an isosceles trapezoid, the upper base is 5, the lower base is 9, and both sides are 6. Find the height of the isosceles trapezoid and the length of one of its diagonals, and explain the solution."

[0115] Through the semantic attributes of the mathematical geometry problem, it can be determined that the user's intention is to seek a solution to this geometry problem and hopes to obtain specific solution steps and guidance.

[0116] Through the category attributes of this mathematical geometry problem, it can be determined that it belongs to the geometry discipline problem, specifically the trapezoid-related knowledge field in plane geometry.

[0117] This mathematical geometry problem involves the properties of isosceles trapezoids, the application of the Pythagorean theorem, and some possible geometric calculation techniques. The difficulty of this mathematical geometry problem is medium to high.

[0118] Based on this attribute information, the heuristic problem-solving agent can make a preliminary judgment on the problem, which may involve the properties of isosceles trapezoids (such as the two sides of an isosceles trapezoid are equal, and the two angles on the same base are equal), the Pythagorean theorem, and the area calculation of triangles and trapezoids.

[0119] Based on the student's knowledge map, it can be determined that the student has a good grasp of the basic properties of isosceles trapezoids (such as definition and the equality of the two sides), but has weak links in the application of the Pythagorean theorem in complex geometric figures and how to transform trapezoid problems into triangle problems to solve the length of the diagonal.

[0120] From the knowledge graph, we can see that the student has learned the basic knowledge of isosceles trapezoids, but has not yet systematically learned the advanced content of using the Pythagorean theorem to solve complex trapezoid problems.

[0121] The knowledge graph clearly records the relationship between isosceles trapezoids and triangles. For example, by drawing auxiliary lines, the isosceles trapezoid problem can be converted into a triangle problem for solution.

[0122] Combining attribute information and knowledge graphs, the heuristic problem-solving agent application can break this problem down into three sequential knowledge points:

[0123] Knowledge Point 1: Review of the properties of isosceles trapezoids, including the definition of an isosceles trapezoid, the equality of its two sides, the equality of two angles on the same base, and the method of drawing auxiliary lines to transform an isosceles trapezoid into a triangle and a parallelogram.

[0124] Knowledge Point 2: Application of the Pythagorean Theorem in Right Triangles. The focus is on how to identify right triangles in complex geometric figures and use the Pythagorean Theorem for calculations.

[0125] Knowledge Point 3: Use the properties of an isosceles trapezoid and the Pythagorean theorem to solve the height and diagonal length of a trapezoid, and apply the previous two knowledge points to practical problems.

[0126] Step S1022: Generate a number of heuristic questions for each knowledge point data based on the respective knowledge point data and the logical relationships therebetween as multiple groups of guiding data for the first input data.

[0127] In this embodiment, a set of heuristic questions may be generated for each piece of knowledge point data. A set of heuristic questions may include one or more heuristic questions arranged from easy to difficult.

[0128] For example, a set of heuristic questions can be divided into multiple levels, such as basic understanding, application and expansion, and comprehensive innovation. Basic understanding questions are designed to test students' understanding of the basic concepts, definitions, and principles of a knowledge point; application and expansion questions require students to apply knowledge points to specific situations and solve practical problems; and comprehensive innovation questions encourage students to integrate multiple knowledge points and engage in innovative thinking and solutions.

[0129] Each heuristic question can be closely centered around the core content of the knowledge point data, ensuring that the question accurately reflects the key points and examination focus of the knowledge point data, and avoiding questions that are irrelevant to the knowledge point data or deviate from the topic.

[0130] Of course, the heuristic question groups corresponding to different knowledge point data can also follow a logical relationship of organization from easy to difficult. For example, in the subject of mathematics, if the basic concept knowledge points (such as the classification of numbers) are involved first, the corresponding heuristic question groups are relatively simple, mainly testing students' understanding and memory of basic concepts; then the applied operation knowledge points (such as simple four arithmetic operation application problems) involved, the difficulty of the heuristic question groups will be appropriately increased, testing students' ability to use concepts to solve practical problems; and then the comprehensive and complex knowledge points involved (such as comprehensive problems combining equations and geometry) will be further increased in difficulty, requiring students to use multiple knowledge points in a comprehensive manner to solve problems. Students can gradually deepen their understanding and mastery of knowledge points at different levels by answering each group of heuristic questions in this order, and ultimately build a complete and systematic knowledge system.

[0131] In this embodiment, multiple sets of guidance data can guide the target user to recall various knowledge points and answer questions according to the knowledge points. During the answering process, the target user can gradually clarify the solution ideas, master the solution methods, and improve the efficiency and accuracy of the problem-solving.

[0132] In this embodiment, by combining the knowledge graph and attribute information to split the first input data (such as a subject question) into multiple knowledge point data, it is possible to accurately determine the key knowledge points involved in the target user's problem solving. On this basis, a number of heuristic questions are generated for each knowledge point data based on the logical relationships between each knowledge point data. This follows the knowledge cognition law of progressing from easy to difficult and from basic to comprehensive, gradually improving the target user's knowledge application and thinking ability, and enhancing the stickiness and interactivity between the user and the target application.

[0133] As another optional embodiment of the present application, refer to Figure 4 , is a flow chart of a data processing method provided in Example 4 of the present application. This embodiment is mainly an implementation method of the above step S102. Figure 4 As shown, the following steps may be included but not limited to:

[0134] Step S1023: Perform generation processing on each content data in the first input data based on the attribute information to obtain a number of guiding data for each content data.

[0135] The attribute information can be used to determine the direction of the generation processing, so that the number of guiding data conforms to the attributes described by the attribute information.

[0136] Each content data in the first input data can be a statement, a paragraph, a keyword or a phrase, etc. For example, in the text generation scenario, for a news report, it can be split into description paragraphs of each news event, quoted statements of key figures, important data, etc.; in the mathematical problem scenario, a complex problem can be split into known conditions, solution objectives and other different content data.

[0137] The forms of the guiding data are diverse. It can be label options for further perfecting the content data, or other alternative solutions.

[0138] For example, in the text generation scenario, if a content data involves an idiom or an abbreviation, the guiding data can be further confirmation label options for the idiom or the abbreviation. For example, for the idiom "waiting for gains without pains", the guiding data can be set as "Is this idiom used appropriately? (A. Appropriate; B. Inappropriate, should be changed to [other idiom])", so that the target user can think about whether the idiom is used appropriately in the text and give alternative solutions.

[0139] In addition to label options, the guiding data can also be other possible expression forms or content supplements. For example, when describing a historical event, the guiding data can be "Another common expression for this historical event is [specific expression]", or "Regarding the impact of this event, [relevant impact content] can also be supplemented", to help users enrich and optimize the text content.

[0140] Step S1024: Screen out a number of target guiding data that match the knowledge graph of the target user from the number of guiding data.

[0141] In this embodiment, the knowledge graph of the target user can record information such as the knowledge points already mastered by the target user and the cognitive level. According to the knowledge graph of the target user, it can be judged whether the knowledge points involved in each guiding data are within the user's knowledge range, and whether the user's mastery of the knowledge points is sufficient to handle the guiding data.

[0142] If the guidance data involves knowledge points that exceed the user's current knowledge level, or requires the user to have higher cognitive abilities and expertise to understand and answer, then this guidance data can be excluded. For example, for a user who is new to programming, guidance data involving advanced algorithm optimization and the application of complex data structures is not suitable as target guidance data.

[0143] At the same time, it is also possible to exclude guidance data that is too simple and has no practical learning value for users. For example, for users who have already mastered basic grammar knowledge, some guidance data on simple grammar rules is not necessary as target guidance data.

[0144] Step S1025: sort the target guidance data based on the knowledge graph to obtain multiple groups of guidance data for the first input data.

[0145] In this embodiment, the target user's mastery of various knowledge points or their associated relationships can be determined based on the knowledge graph, thereby determining the user's mastery or proficiency in different knowledge points. For example, if the target user has recently learned a certain knowledge point, then the user is likely to be more familiar with it; if a certain knowledge point is closely related to other knowledge points that the target user has already mastered, then the target user is likely to have a certain understanding of it.

[0146] According to the target users' mastery or proficiency in relevant knowledge points, several target guidance data are sorted. For example, the guidance data corresponding to the knowledge points that users are more familiar with and have a better mastery of are arranged first, and then the guidance data corresponding to the knowledge points that users are relatively less familiar with and have a lower mastery of are arranged, so that the several target guidance data present a logical relationship from easy to difficult.

[0147] By sorting, multiple sets of guided data can be aligned with the user's learning patterns and cognitive processes, allowing them to gradually deepen their learning and thinking. For example, in mathematics learning, students can first be asked to answer some guided data related to basic concepts and simple calculations, and then gradually transition to guided data involving comprehensive applications and complex reasoning. This helps students gradually build a knowledge system and improve their problem-solving skills.

[0148] In this embodiment, each content data in the first input data is generated and processed separately based on the attribute information to obtain a number of guidance data for each content data, which can reduce the guidance data that does not match the attributes of the first input data, making the guidance data more accurate and effective. On this basis, a number of target guidance data that matches the knowledge graph of the target user is screened out from the several guidance data, so that the guidance data exposed to the target user can be as far as possible within the scope of its understanding and processing, providing personalized guidance for the target user. In addition, after sorting the several target guidance data based on the knowledge graph as multiple groups of guidance data for the first input data, the user can process the guidance data step by step, which is in line with the human cognitive logic from shallow to deep and from easy to difficult. For example, in a creative design scenario, first guiding the user to optimize the familiar design elements, and then guiding them to try new design styles and techniques, will help the user better understand and accept.

[0149] As another optional embodiment of the present application, refer to Figure 5 , is a flow chart of a data processing method provided in Example 5 of the present application. This embodiment is mainly an implementation method of the above step S103. Figure 5 As shown, the following steps may be included but not limited to:

[0150] Step S1031: Update the guidance data of the target group in the guidance data based on the evaluation result of the feedback data.

[0151] In this embodiment, the feedback data may be evaluated from at least one of the following dimensions to obtain an evaluation result:

[0152] Correct Answer Rate: This system records the target user's responses to questions posed by the system in real time and calculates the correct answer rate. For example, in a knowledge quiz scenario, if the target user correctly answers multiple questions in a row for a certain knowledge module, it indicates a good grasp of this knowledge; if the user frequently answers incorrectly, it indicates difficulty understanding this knowledge module.

[0153] Answering speed: In addition to accuracy, time spent answering is also a key evaluation factor. Quick and accurate answers indicate a high level of familiarity with the subject matter. However, taking too long may indicate a lack of understanding and a need for further refinement.

[0154] Task Progress: The target user's (e.g., student's) task progress (e.g., learning progress) is determined by their performance in answering questions. If the target user consistently correctly answers multiple questions of a certain difficulty level, it is considered that they have a good grasp of this knowledge and are progressing quickly. Conversely, if the target user frequently answers incorrectly, it is considered that they have difficulty with this knowledge and are progressing slowly.

[0155] User acceptance of guidance data: Analyze target users' operational behavior and feedback regarding guidance data to understand their acceptance of guidance data. For example, if users frequently follow guidance data prompts and do not raise significant objections later in the process, this indicates a high degree of user acceptance of the guidance data. Conversely, if users repeatedly ignore guidance data or suggest changes, this indicates that the guidance data may not be aligned with user needs.

[0156] Frequency of new questions: Record the number of times target users submit new questions while processing the guidance data. Frequent submissions may indicate that the current guidance data doesn't fully meet their needs, or that there are issues with the guidance process that lead to new questions. A low frequency of new questions suggests that the guidance data is able to guide users through their tasks successfully to a certain extent.

[0157] In this embodiment, by updating the guidance data of the target group in the guidance data, the task advancement path (e.g., learning path) can be dynamically adjusted. For example, if the target user's task advancement progress is fast, the target application can appropriately increase the difficulty of the questions and introduce more challenging knowledge point related guidance data; if the target user's task advancement progress is slow, the target application can reduce the difficulty of the questions, or return to the target user's previously mastered knowledge point related guidance data for consolidation, and at the same time adjust the order of the questions, putting more basic questions that are more helpful in understanding the current difficulty in advance, so as to help the target user better complete the task.

[0158] As another optional embodiment of the present application, a data processing method is provided in Example 6 of the present application. This embodiment is mainly an implementation of the above-mentioned step S1031, and may specifically include but is not limited to at least one of the following:

[0159] Step S31 : when the evaluation result indicates that the target user's acceptance of the guidance data is lower than a first threshold, updating the logical relationship between the guidance data of the target group in the guidance data.

[0160] If the evaluation result indicates that the target user's acceptance of the guidance data is below a first threshold, this may indicate that there are issues with the logical relationships between the guidance data, making it difficult for the user to understand and follow the guidance. For example, in a knowledge quiz or task execution scenario, if the steps between guidance data jump too much or do not follow the user's familiar cognitive logic, the user may become confused, thereby reducing their acceptance of the guidance data.

[0161] To improve this situation, the target app can update the logical relationships between the target group's guidance data within the guidance data. This can be done in a variety of ways, such as adjusting the original logic from difficult to easy to easy to easy, or changing the logic from deep to shallow to shallow to deep. Through such adjustments, the guidance data can better align with user cognitive habits, making it easier for users to understand and accept.

[0162] Consider a software onboarding scenario where the original logic of the onboarding data is to first guide the user through an advanced function before introducing basic functions. However, users frequently ignore the onboarding data or suggest changes, leading the system to determine that the user's acceptance of the onboarding data is low. In this case, the system can update the onboarding data to first introduce basic functions and then gradually guide the user through advanced functions, making the onboarding process more aligned with the user's cognitive order.

[0163] The first threshold can be set as needed and is not limited in this application.

[0164] In this embodiment, the second prompt information generated based on the feedback data may be input into a second processing model, and the second processing model may process the second prompt information to generate an evaluation result of the feedback data.

[0165] In this embodiment, the third prompt information generated based on the evaluation result of the feedback data and the guidance data can be input into the first processing model, and the first processing model can process the information to generate the updated guidance data of the target group.

[0166] Step S32: if the evaluation result indicates that the target user's acceptance of the guidance data is greater than a second threshold, update the logical order of the guidance data of the target group in the guidance data.

[0167] When the evaluation result indicates that the target user's acceptance of the guidance data is greater than the second threshold, the current guidance data can be considered to be effective to some extent, but there may be room for further optimization. At this time, the target application can consider updating the logical order of the guidance data of the target group in the guidance data. The logical order update can be to adjust the order of presentation of each element under the premise that the logical relationship of the current guidance data is basically reasonable. The logical order update may not change the internal logical connection between the elements in the guidance data, but only adjust the order in which they appear.

[0168] In this embodiment, the order of the guidance data can be adjusted based on the target user's answering performance, task progress, and other information. For example, if the target user shows a higher interest in certain knowledge points or masters them faster during the learning process, the guidance data for these knowledge points can be presented earlier, allowing the user to learn more deeply. Alternatively, if the user frequently answers questions of a certain difficulty level incorrectly, the guidance data for related basic knowledge points can be presented earlier, helping the user consolidate their basic knowledge before moving on to more in-depth learning.

[0169] The second threshold value can be set as needed and is not limited in this application. The second threshold value can be greater than the first threshold value.

[0170] Step S33: When the evaluation result indicates that the target user's acceptance of the guidance data is greater than a third threshold, the guidance data is regenerated by introducing target knowledge points based on the knowledge graph.

[0171] The third threshold value can be set as needed and is not limited in this application. The third threshold value can be greater than the second threshold value.

[0172] If the evaluation result indicates that the target user's acceptance of the guidance data is greater than the third threshold, it indicates that the target user has adapted well to the current guidance method and may wish to further expand their knowledge or delve deeper into related content. In this case, the target application can introduce target knowledge points based on the knowledge graph and regenerate the guidance data, providing the user with a richer and more in-depth learning experience.

[0173] The regenerated guidance data may include more extensive content, deeper questions, or more complex tasks to guide users to deeper learning and exploration. At the same time, the target application will ensure that the new guidance data matches the user's current knowledge level and learning needs to avoid placing excessive learning pressure on users.

[0174] For example, in an academic research guidance scenario, users have a good grasp of the basic knowledge in a certain research field through previous guidance data and are highly receptive to the guidance data. The target application, based on the knowledge graph, discovers that there are cutting-edge research directions and hot topics in this research field that are relevant to the user's current learning content. The target application can then incorporate these target knowledge points and regenerate guidance data, providing users with an introduction to cutting-edge research directions, guidance on relevant research methods, and some open research questions, guiding them into deeper academic research exploration.

[0175] As another optional embodiment of the present application, a data processing method is provided in Example 7 of the present application. This embodiment is mainly an implementation of the above-mentioned step S103, and may specifically include but is not limited to at least one of the following:

[0176] Step S1032: Generate corresponding target knowledge point explanation content and / or application examples based on the feedback data.

[0177] In this embodiment, whether the target user has difficulties and errors can be detected based on the feedback data, such as stuck in answering questions, no operation for a long time, confusing operation steps, wrong answers, etc.

[0178] Problem identification: Problem identification is performed based on specific knowledge or skill points. For example, in a programming learning scenario, if a user encounters an error while writing code related to a loop structure, the target application can analyze the code and provide error messages to identify potential misunderstandings or operational errors in the loop structure's conditional judgments, loop execution logic, or loop termination conditions. Alternatively, if a student incorrectly answers a question about functions, the target application can identify potential gaps in the student's understanding of functions.

[0179] The target application can record relevant learning information where the target user encounters difficulties or answers incorrectly, including the incorrect questions, tasks, operation steps, incorrect answers, and error time.

[0180] The target application can determine the target user's current knowledge level based on the recorded learning information. Based on the target user's knowledge level, the application can provide concise and easy-to-understand explanations of relevant knowledge points or skill points where problems are detected. For example, when explaining programming loop structures, the application might begin with the basic concepts of loop structures, such as "a loop structure allows a program to repeatedly execute a section of code according to certain rules, like an automated repetitive operation machine." Alternatively, for function questions, the application might explain the definition and basic properties of functions, while also integrating the student's previous learning data to highlight key points that students tend to overlook or confuse.

[0181] Explanation can be in various forms, including text description, chart display, animation demonstration, etc. For some complex algorithms or operation processes, animation demonstration can more intuitively show their execution process and help users understand.

[0182] In this embodiment, the target application can generate an application instance by combining information such as the target user's age, occupation, life background, and knowledge mastery.

[0183] Application examples should be representative and actionable, authentic to the target user's real-life or work situations. For example, for new employees learning project management, a small team activity could be used as an example to illustrate the specific operations and interrelationships within project management, such as project planning, resource allocation, and progress monitoring. Alternatively, for junior high students learning functions, the system could use the relationship between car speed and time as an example to illustrate the changing relationship between the independent and dependent variables in a function, helping students more intuitively grasp the concept of functions.

[0184] In this embodiment, the fourth prompt information generated based on the feedback data can be input into the first processing model, and the first processing model can process it to generate target knowledge point explanation content and / or application examples.

[0185] Step S1033: Generate a learning suggestion that matches the target user based on the feedback data, and associate the learning suggestion with the knowledge graph of the target user.

[0186] In this embodiment, error analysis can be performed on the target user's answers, operation results, etc. in the feedback data. Not only can the correctness of the answer and the success of the operation be determined, but the reasons for the error or failure can also be deeply analyzed, such as misunderstanding of concepts, calculation errors, lack of proficiency in skill application, deviation in problem-solving ideas, etc. For example, in a software operation skill improvement scenario, when a user completes a software function operation task, an error occurs. The system analysis finds that this is because the user is not familiar with the use of a tool in the software, resulting in an incorrect operation step. Or, when a student calculates the area of ​​a geometric figure, the system analysis finds that the student's answer is incorrect because of the incorrect application of the area formula.

[0187] Based on the above error analysis results and the target user's knowledge graph, tailored learning suggestions are provided to the target user. For example, for the problem of unfamiliarity with the use of tools in the above software operations, the system may suggest that the user re-learn the tool's usage tutorial, provide some simple practice tasks to familiarize the user with the tool's operating interface and basic functions, and provide detailed step-by-step examples of how to correctly use the tool to complete the task. Alternatively, for the case of incorrect use of the above area formula, the system may suggest that students review the derivation process of the area formula, provide some similar but simpler area calculation problems for students to consolidate their practice, and provide detailed step-by-step examples of how to correctly use the area formula to solve problems.

[0188] In this embodiment, the fifth prompt information generated based on the feedback data can be input into the first processing model, and the first processing model can generate learning suggestions after processing.

[0189] Learning suggestions can include different levels of error correction. For example, for beginners, suggestions may focus more on consolidating basic knowledge and basic operations; for users with some experience, suggestions may focus more on improving advanced skills and solving complex problems. For example, when learning painting skills, for beginners, the system may recommend more line practice and basic color matching; for users with a certain foundation, the system may recommend trying different painting styles and practicing creation.

[0190] Learning suggestions can be presented in various forms, including text descriptions, video demonstrations, and online course recommendations. For some complex operating skills, video demonstrations can more intuitively show the correct operating methods, helping users better understand and master them.

[0191] As another optional embodiment of the present application, refer to Figure 6 , is a flow chart of a data processing method provided in Example 8 of the present application, such as Figure 6 As shown, the method may include but is not limited to the following steps:

[0192] Step S201: In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data.

[0193] Step S202: Generate multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information, and the multiple sets of guidance data have a target logical relationship between them.

[0194] Step S203: Generate corresponding target response results based on the target user's feedback data on the multiple sets of guidance data.

[0195] The detailed process of steps S201-S203 can be found in the relevant introduction of steps S101-S103 in Example 1, and will not be repeated here.

[0196] Step S204: guiding the target user to learn knowledge data related to the first input data based on the guidance data.

[0197] The target application can present guidance data to the target user in a variety of intuitive and easy-to-understand ways. For example, when guiding users to learn about geometric figures in mathematics, such as triangles, text descriptions can be used to explain the definition of triangles and their basic classifications (equilateral triangles, isosceles triangles, and general triangles).

[0198] Graphical presentations can also be used to enhance user understanding of key concepts. For the aforementioned triangle knowledge, graphical examples of different types of triangles can be displayed, allowing users to intuitively see the shape characteristics of triangles and the differences between them. Furthermore, animations can be used to illustrate the dynamic changes in key concepts. For example, demonstrating the proof that the sum of the interior angles of a triangle is 180°, using animations to gradually show the changes and calculations of angles, helps users better grasp this key concept.

[0199] Step S205: Generate and update a corresponding target feedback interface based on the feedback data, wherein the target feedback interface provides operation controls or recommended content for updating the knowledge graph of the target user.

[0200] In this embodiment, if the user triggers an operation control for updating the knowledge graph of the target user, the target application can update the knowledge graph of the target user based on the feedback data.

[0201] Based on the updated knowledge graph, recommended content can be generated. Still using the target user as a student as an example, a comprehensive and in-depth learning report can be automatically generated based on the updated knowledge graph. This report can cover multiple important dimensions, including learning progress, error analysis, and knowledge mastery.

[0202] In terms of learning progress, the report will list in detail the tasks that the students have completed, such as the chapters studied, the homework completed, etc., and clearly point out the weak links, that is, the knowledge points with low mastery or loosely related knowledge systems in the knowledge map.

[0203] In the error analysis section, the report will provide a detailed statistical analysis of error types, such as conceptual misunderstandings, calculation errors, logical reasoning errors, etc., and analyze the error frequency. By combining the knowledge graph, it will identify frequently occurring knowledge points and their related knowledge points, and conduct an in-depth analysis of the root causes of the errors.

[0204] Knowledge point mastery is displayed through intuitive charts and text descriptions, demonstrating the student's mastery of each knowledge point, as well as the relationships between knowledge points and the learning path. Learning reports are presented using a variety of elements, including text, formulas, images, and tables, so that students can clearly and comprehensively understand their learning status.

[0205] Based on the learning report, personalized learning feedback content can be further generated, mainly including course recommendations, exercise recommendations and video recommendations.

[0206] Course recommendations: We conduct an in-depth analysis of the knowledge points and weak links in students' knowledge graphs, and recommend relevant course content to students based on the correlation and learning order between the knowledge points in the knowledge graph. These courses will not only cover students' weak points, but also help them build a complete knowledge system and deepen their understanding of relevant knowledge. For example, if a student shows a poor grasp of the function knowledge graph in the mathematics subject, we recommend a series of courses from basic function concepts to complex function applications, including function graph drawing, function property analysis, and the application of functions in practical problems.

[0207] Exercise recommendations: For the knowledge points marked in the knowledge graph where students frequently make mistakes, suitable exercises can be carefully selected from the massive exercise library. These exercises will be classified and sorted according to multiple dimensions such as difficulty, question type, and relevance of knowledge points to form a set of targeted exercise sets. By practicing these exercises, students can consolidate weak knowledge points and improve their problem-solving and knowledge application abilities. For example, for the knowledge points in mechanics in the physics subject where students frequently make mistakes, the system will recommend a series of exercises ranging from the application of simple mechanics principles to complex mechanics comprehensive problems, covering a variety of question types such as multiple-choice questions, fill-in-the-blank questions, and calculation questions.

[0208] Video Recommendations: Based on students' learning progress and interests, and taking into account the characteristics and importance of each knowledge point in the knowledge graph, relevant video courses can be recommended. These video courses not only explain the knowledge points in detail, but also combine real-life examples and problem-solving techniques to present the learning content in a vivid and visual way, enhancing students' learning experience and comprehension. For example, for a knowledge point in a specific historical period in the history subject, the system will recommend a video course that includes vivid historical stories, expert interpretations, and cultural relics displays to stimulate students' learning interest.

[0209] Users can interact with the generated recommendations in the target feedback interface. For example, students can select recommended exercises and practice them on the target feedback page. After students select exercises, the target application can record the students' practice process and results, including answering time, answer accuracy, reasons for errors, etc. At the same time, the target application can provide timely feedback on the answer results and detailed analysis based on the students' practice, helping students understand their learning effects and existing problems. In addition, students' practice data will be updated in real time to the knowledge graph, so that the target application can further adjust the learning recommendation strategy and provide students with more accurate and personalized learning support.

[0210] In this embodiment, steps S204 - S205 may also be performed after at least one of the above steps S31 - S33 .

[0211] As another optional embodiment of the present application, refer to Figure 7, is a flow chart of a data processing method provided in Example 9 of the present application, such as Figure 7 As shown, the method may include but is not limited to the following steps:

[0212] Step S301: In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data.

[0213] Step S302: Generate multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information, and the multiple sets of guidance data have a target logical relationship between them.

[0214] Step S303: Generate corresponding target response results based on the target user's feedback data on the multiple sets of guidance data.

[0215] The detailed process of steps S301-S303 can be found in the relevant introduction of steps S101-S103 in Example 1, and will not be repeated here.

[0216] Step S304: Obtain interaction data between the target user and the target application, update the target user's knowledge graph based on the interaction data, and / or generate corresponding positive evaluation content based on the interaction data.

[0217] The target application is an application that can call the at least one processing model to execute a target processing function, and the interaction data includes data as input to the processing model and feedback data for the inference result output by the processing model.

[0218] The following uses students as an example to explain knowledge graph updates in detail. For example, the raw information that students actively input into the target application (such as submitted homework answers, problem-solving steps, and stated learning questions) can serve as learning data input to the processing model. This data can reflect students' learning behavior and current knowledge mastery.

[0219] After the processing model generates inference results (such as homework grading results and problem answers) based on the input learning data, students will provide feedback on these results, generating feedback data. Feedback can take various forms, including judgments on the correctness of answers, understanding of problem-solving ideas, and further questions about knowledge points.

[0220] Based on learning data and feedback data, the students' learning progress, error types, test scores, etc. in the knowledge graph can be adjusted to ensure that the knowledge graph is consistent with the students' actual learning situation.

[0221] In this embodiment, the frequency and difficulty of students' errors can be analyzed based on learning data and feedback data, and the students' weak links in corresponding knowledge points can be marked in the knowledge graph.

[0222] Based on the updated knowledge graph, we can analyze the relationships between students' knowledge points and learning resources such as course content, exercises, and video courses. For example, if the knowledge graph shows that a student has a good understanding of a mathematical theorem, the target application will recommend related extension exercises, practical application cases, and relevant mathematical competition questions to help students further deepen their understanding and application of the theorem.

[0223] Through the real-time updated knowledge graph, students' learning paths can be accurately tracked, understanding their key learning points, difficulties, and progress at different stages. For example, the target application can plot a student's learning time curve for each knowledge point over the course of a semester, analyzing their learning status and efficiency at different time periods.

[0224] Furthermore, the knowledge graph can help target applications identify bottlenecks in student learning—critical knowledge points that hinder further learning or performance improvement. Based on the weaknesses identified in the knowledge graph, the target application can analyze the relationships between these weak links to identify the root causes of students' learning difficulties, thereby providing more targeted solutions.

[0225] Furthermore, based on the updated knowledge graph, the target application can analyze students' knowledge needs and interests and recommend personalized learning resources, functions, or services. For example, if the knowledge graph shows that a student has a strong interest in a certain subject area and has mastered a certain foundation, the target application will recommend more in-depth courses, research reports, or related academic exchange activities in that field to meet the student's needs for further in-depth learning and exploration.

[0226] After the target response generated for the target input data is presented to the target user, the user can provide feedback based on their understanding and needs. This feedback data may be presented in various forms, such as text comments entered directly by the user on the target application's interactive interface (such as "This answer is not very clear" or "I understand part of the content, but I still have some questions"), user thumbs-up or thumbs-down of the response, and further questions raised by the user regarding the response.

[0227] The positive evaluation content may include but is not limited to: first positive evaluation content and second positive evaluation content.

[0228] The first positive evaluation content is different from second positive evaluation content generated for the target input data.

[0229] The second positive evaluation content generally does not involve a specific response to the target response result feedback. For example, in response to a user's question about a textbook content they don't understand (i.e., one implementation of the target input data), the user might generate a comment like "You're really great! Taking the initiative to learn about knowledge points you don't understand is the beginning of progress" (i.e., one implementation of the second positive evaluation content).

[0230] First, positive evaluation content can not only contain encouraging elements, but also goal-guiding information.

[0231] Target guidance information can be mainly used to guide users to learn other content or guide users to understand through other means when users do not accept or understand the target response results.

[0232] Of course, goal-directed information can also be used to demonstrate learning achievements.

[0233] Guiding learning of other content can be understood as:

[0234] Based on user feedback, we can analyze potential knowledge gaps or interests and guide users to learn more about related content. For example, if a user doesn't understand a particular application scenario of a mathematical function, we can guide them to learn about other, simpler or more complex applications of that function, helping them understand functions from different perspectives.

[0235] For example, in a heuristic problem-solving agent application, a user did not understand the explanation of the application of trigonometric functions to complex geometric shapes in the target response. The heuristic problem-solving agent application generated the first positive evaluation content based on the user feedback data, which included the target guidance information: "Although the application of trigonometric functions to complex geometric shapes is a bit difficult to understand, we can start by learning how to apply them to simple triangles. Once we have mastered the basics, it will be much easier to work on complex shapes. Below are some recommended exercises and learning materials for applying trigonometric functions to simple triangles."

[0236] Guidance is understood in other ways, which can be understood as:

[0237] Different users have different learning styles and preferences. Some users may be more suited to understanding knowledge through intuitive methods such as videos and animations, while others prefer to learn through reading text and doing exercises. If a user does not understand the current text-based response, you can use user profile data (if previously collected) and feedback data to guide the user to other more suitable methods of understanding.

[0238] For example, in a heuristic problem-solving agent application, if a user doesn't understand the textual explanation of a grammatical point in the target response, the target guidance message in the first positive review generated by the heuristic problem-solving agent application could be: "It seems the textual explanation might be difficult to understand. We've prepared some interesting and engaging grammar explanation videos for you to watch to better grasp this grammar point. Click here to watch."

[0239] Demonstrating learning achievements can be understood as:

[0240] Real-time tracking and recording of students' learning achievements, such as how many questions they completed, which knowledge points they learned, and how much study time they completed, can be displayed to users as goal guidance, allowing them to intuitively see their learning progress and results. By displaying learning achievements, users gain a sense of accomplishment and satisfaction, thereby enhancing their learning motivation.

[0241] In this embodiment, the generation method and focus of positive evaluation content can be flexibly and accurately adjusted according to the learning performance and progress shown by students at different learning stages, so as to always adapt to the students' learning status and effectively maintain and improve students' learning enthusiasm.

[0242] For example, when students repeatedly make mistakes on a particular knowledge point during their learning process, the target application can keenly detect this and adjust its strategy for generating positive feedback. At this point, the positive feedback will focus on providing students with more encouragement and guidance. For example, positive feedback might include: "This knowledge point is indeed quite difficult, but you have diligently tried again and again. This perseverance is truly commendable. Come, let's delve deeper into the relevant examples and explanations. I believe that with your perseverance, you will definitely master it thoroughly." Through such targeted feedback, students can feel that the target application recognizes their efforts, while providing clear guidance and strengthening their confidence in overcoming difficulties.

[0243] If a student demonstrates rapid learning progress at a certain stage of learning, the target application will also promptly adjust the way positive evaluation content is generated. At this time, the positive evaluation content will give students a higher degree of affirmation and encouragement, and appropriately provide more challenging suggestions. For example, the positive evaluation content may be: "Wow, your learning speed is amazing. You have mastered so many knowledge points so quickly! Next, why not challenge some more difficult content to further improve your abilities? I believe you will achieve greater breakthroughs in your studies." Such feedback not only affirms the students' excellent performance, but also inspires the students' desire to move towards higher goals, prompting students to continue to maintain a positive learning attitude.

[0244] By dynamically adjusting the way positive evaluation content is generated based on different learning stages, the target application can always closely fit the students' learning status, provide students with just the right incentives and guidance, and allow students to feel the support and attention of the target application at all stages of learning, thereby always maintaining a high level of learning enthusiasm.

[0245] In this embodiment, when a user sees the first positive evaluation content generated by the target application based on their feedback on the target response, they will feel the target application's attention and attention to them, and thus be more willing to interact with the target application. The target application can also continue to provide targeted feedback and guidance based on these user interactions, forming a virtuous cycle of interaction. For example, when interacting with students, the first positive evaluation content can enhance students' learning motivation and promote their enthusiasm through an incentive mechanism.

[0246] Next, the data processing method provided by this application is explained in conjunction with the learning scenario. Figure 8 As shown, during the learning process, students have multiple ways to input learning-related information, namely, they can input in text form through a text input box, use a voice microphone for voice input, and use an image upload interface to upload images, etc. With these multimodal input methods, students can input the first input data into the heuristic solution agent application (i.e., an implementation method of the target application).

[0247] After receiving the first input data, the heuristic solution agent application can perform multimodal processing on it. This processing process is similar to the method of uniformly formatting the target input data mentioned above, and will not be repeated here.

[0248] After processing, the heuristic answering agent application can generate several heuristic questions. The specific process of generating several heuristic questions can be found in the relevant introduction of steps S1021-S1022 above.

[0249] Afterwards, the students can start answering these heuristic questions. After the students answer the questions, the heuristic answering agent application can enter the learning path adjustment phase. The specific operation of this phase can refer to the relevant content of step S1031 above.

[0250] Students can continue answering questions in the new learning path.

[0251] For each question answered by the student, the heuristic answering agent is applied to determine whether the student has difficulty or makes errors in answering the question.

[0252] If there are difficulties or errors, corresponding explanations will be generated for the knowledge points where the problems occurred, and application examples will be generated based on actual application scenarios. In this way, students can better understand the knowledge and master problem-solving methods.

[0253] If there are no difficulties or errors: The heuristic problem-solving agent application can provide real-time feedback and error correction to ensure the accuracy of the student's answer, and generate learning suggestions based on the student's answer to guide students to further optimize their learning methods and improve learning outcomes.

[0254] After the student has answered the question and completed the above processing, the heuristic answering agent application can further determine whether the student actively asked the question.

[0255] If students ask questions proactively: This indicates that the students have strong learning initiative and curiosity. The application of heuristic answering agents can generate positive evaluation content, affirm and encourage students' proactive questioning behavior, and enhance students' learning confidence.

[0256] If the student does not actively ask questions: the heuristic answering agent application can continue to advance according to the established learning process, guide the student to the next learning link, and continue the learning activities.

[0257] The following will take the heuristic problem-solving agent application as the target application, and elaborate on the complete application process of the data processing method provided by this application in actual learning scenarios.

[0258] For example, Figure 9 As shown in the figure, students input a picture of a page in the workbook and a text description "I know it is to find the area of ​​a ring, but I just don't know how to calculate it, my mind is a mess" into the heuristic solution agent application. The picture contains the question "To create a suitable living environment for "beautiful countryside", Liushu Village decided to build a 200cm wide circular path around a small circular lake with a diameter of 30m in the village. What is the area of ​​this path?"

[0259] After receiving the image and text description input by the student, the Heuristic Answering Agent application explains that it will use heuristic questions to help the student understand the problem and generates three heuristic questions:

[0260] "1. Look carefully, how many circles are there in the question?" Guide students to pay attention to the number of geometric figures involved in the question, laying the foundation for subsequent calculations.

[0261] "2. Have you learned which formula to use to calculate the area of ​​a circle?" Help students review the basic knowledge of calculating the area of ​​a circle and strengthen their memory.

[0262] "3. Given the diameter and the width of the path, can you calculate the radius of each of the two circles?" Guide students to think about how to derive the key data needed to solve the problem from the known conditions.

[0263] As students answer these three heuristic questions, the system will evaluate their responses. Based on the evaluation results, the system will enter the learning path adjustment phase, update the heuristic questions, reduce the difficulty of the questions, and provide students with new heuristic questions:

[0264] "1. Do you remember the relationship between diameter and radius?" Ask questions based on basic concepts that students may have forgotten to help them consolidate their knowledge.

[0265] "2. What are the radii of the two circles in the question?" Guide students to calculate step by step and break down complex problems into simple steps.

[0266] Based on students' responses to new heuristic questions, we can generate explanations of key concepts and practical examples. For example, "In a circle,... the diameter of the small lake is 30cm, and the radius is... Keep at it and you'll fully grasp it!" This explanation helps students understand the key concepts. Then, we generate new heuristic questions, such as "Try to solve this problem: the area of ​​the ring," to guide students in applying their knowledge to real-world problems.

[0267] When students answer subsequent heuristic questions, the heuristic-solving agent determines whether they are struggling or making mistakes and generates learning suggestions. Furthermore, it can generate further heuristic questions, such as "1. Calculate the area of ​​a ring," "2. Calculate the radius," and "3. Calculate the area of ​​a sector." This guides students to expand their learning and deepen their understanding and mastery of relevant knowledge.

[0268] During the learning process, if students actively ask questions, positive evaluation content can be generated (such as, Wow, your question is so insightful!...), which can motivate students to continue learning and improve themselves.

[0269] Throughout the learning process, heuristic questions are generated continuously. The heuristic-solving agent continuously adjusts the content and difficulty of the heuristic questions based on students' learning progress and feedback, ensuring that the questions always align with students' learning progress and needs, guiding them to gradually deepen their learning and improve their learning outcomes.

[0270] Next, a data processing device provided by the present application is introduced. The data processing device introduced below and the data processing method introduced above can be referenced to each other.

[0271] Reference Figure 10 The data processing device includes: a first determining module 100, a first generating module 200 and a second generating module 300.

[0272] A first determining module 100 is configured to determine, in response to obtaining first input data input by a target user into a target application, a knowledge graph of the target user and attribute information of the first input data;

[0273] A first generating module 200 is configured to generate multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information, wherein the multiple sets of guidance data have a target logical relationship between them; and

[0274] The second generating module 300 is configured to generate corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data.

[0275] The first determination module 100 can be used to

[0276] Identifying semantic attributes and / or category attributes of target content in the first input data, and determining attribute information of the first input data based on the semantic attributes and / or category attributes and the context of the first input data;

[0277] Obtain historical learning data of the target user, and determine a knowledge graph of the target user based on the historical learning data and attribute information of the first input data.

[0278] The first generating module 200 may be specifically configured to:

[0279] Splitting the first input data into a plurality of knowledge point data based on the knowledge graph and the attribute information;

[0280] Based on the various pieces of knowledge point data and the logical relationships therebetween, a number of heuristic questions for the various pieces of knowledge point data are generated as multiple groups of guiding data for the first input data.

[0281] The first generating module 200 may be specifically configured to:

[0282] generating and processing each content data in the first input data based on the attribute information to obtain a plurality of guide data for each content data;

[0283] Filtering out a plurality of target guidance data that matches the knowledge graph of the target user from the plurality of guidance data; and

[0284] The target guidance data are sorted based on the knowledge graph as multiple groups of guidance data for the first input data.

[0285] The second generation module 300 may be specifically configured to:

[0286] The guidance data of the target group in the guidance data is updated based on the evaluation result of the feedback data.

[0287] The second generating module 300 updates the guidance data of the target group in the guidance data based on the evaluation result of the feedback data, which may specifically include at least one of the following:

[0288] If the evaluation result indicates that the target user's acceptance of the guidance data is lower than a first threshold, updating the logical relationship between the guidance data of the target group in the guidance data;

[0289] If the evaluation result indicates that the target user's acceptance of the guidance data is greater than a second threshold, updating the logical order of the guidance data of the target group in the guidance data;

[0290] When the evaluation result indicates that the target user's acceptance of the guidance data is greater than a third threshold, the guidance data is regenerated by introducing target knowledge points based on the knowledge graph.

[0291] The second generation module 300 may further include at least one of the following:

[0292] Generate corresponding target knowledge point explanation content and / or application examples based on the feedback data;

[0293] A learning suggestion matching the target user is generated based on the feedback data, and the learning suggestion is associated with the knowledge graph of the target user.

[0294] The data processing device may further include:

[0295] a guiding module, configured to guide the target user to learn knowledge data related to the first input data based on the guiding data;

[0296] The third generation and update module is used to generate and update the corresponding target feedback interface based on the feedback data, and the target feedback interface provides operation controls or recommended content for updating the knowledge graph of the target user.

[0297] The data processing device may further include:

[0298] a fourth generation and update module, configured to obtain interaction data between a target user and the target application, update the target user's knowledge graph based on the interaction data, and / or generate corresponding positive evaluation content based on the interaction data;

[0299] The target application is an application that can call the at least one processing model to execute a target processing function, and the interaction data includes data as input to the processing model and feedback data for the inference result output by the processing model.

[0300] In another embodiment of the present application, an electronic device is provided, including at least one target application and at least one processing model capable of running on the target application, wherein the processing model can be called by the target application to perform at least one of the following:

[0301] In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data;

[0302] generating a plurality of sets of guidance data for the first input data based on the knowledge graph and the attribute information, wherein the plurality of sets of guidance data have a target logical relationship between them; and

[0303] Generate corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data.

[0304] It should also be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided in this application, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines.

[0305] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course can also be implemented by special hardware including application-specific integrated circuits, special CPUs, special memories, special components, etc. In general, all functions performed by computer programs can be easily implemented with corresponding hardware, and the specific hardware structures used to implement the same function can also be diverse, such as analog circuits, digital circuits or special circuits, etc. However, for the present application, software program implementation is a better implementation method in most cases. 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 software product, which is stored in a readable storage medium, such as a computer's floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk or optical disk, etc., and includes a number of instructions to enable a computer device (which can be a personal computer, training equipment, or network equipment, etc.) to execute the methods described in each embodiment of the present application.

[0306] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0307] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, a computer, a training device or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website, a computer, a training device or a data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

Claims

1. A data processing method, comprising: In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data; generating a plurality of sets of guidance data for the first input data based on the knowledge graph and the attribute information, wherein the plurality of sets of guidance data have a target logical relationship between them; as well as, Generate corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data.

2. The data processing method according to claim 1, wherein: Determining the knowledge graph of the target user and the attribute information of the first input data includes: Identifying semantic attributes and / or category attributes of target content in the first input data, and determining attribute information of the first input data based on the semantic attributes and / or category attributes and the context of the first input data; Obtain historical learning data of the target user, and determine a knowledge graph of the target user based on the historical learning data and attribute information of the first input data.

3. The data processing method according to claim 1, wherein generating multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information comprises: Splitting the first input data into a plurality of knowledge point data based on the knowledge graph and the attribute information; Based on the various pieces of knowledge point data and the logical relationships therebetween, a number of heuristic questions for the various pieces of knowledge point data are generated as multiple groups of guiding data for the first input data.

4. The data processing method according to claim 1, wherein generating multiple sets of guidance data for the first input data based on the knowledge graph and the attribute information comprises: generating and processing each content data in the first input data based on the attribute information to obtain a plurality of guide data for each content data; Filtering out a plurality of target guidance data that matches the knowledge graph of the target user from the plurality of guidance data; as well as, The target guidance data are sorted based on the knowledge graph as multiple groups of guidance data for the first input data.

5. The data processing method according to claim 1, wherein generating corresponding target response results based on the target user's feedback data on the multiple sets of guidance data comprises: The guidance data of the target group in the guidance data is updated based on the evaluation result of the feedback data.

6. The data processing method according to claim 5, wherein updating the guidance data of the target group in the guidance data based on the evaluation result of the feedback data comprises at least one of the following: If the evaluation result indicates that the target user's acceptance of the guidance data is lower than a first threshold, updating the logical relationship between the guidance data of the target group in the guidance data; If the evaluation result indicates that the target user's acceptance of the guidance data is greater than a second threshold, updating the logical order of the guidance data of the target group in the guidance data; When the evaluation result indicates that the target user's acceptance of the guidance data is greater than a third threshold, the guidance data is regenerated by introducing target knowledge points based on the knowledge graph.

7. The data processing method according to claim 1, further comprising generating corresponding target response results based on the target user's feedback data on the multiple sets of guidance data, further comprising at least one of the following: Generate corresponding target knowledge point explanation content and / or application examples based on the feedback data; A learning suggestion matching the target user is generated based on the feedback data, and the learning suggestion is associated with the knowledge graph of the target user.

8. The data processing method according to claim 1 or 6, further comprising: guiding the target user to learn knowledge data related to the first input data based on the guidance data; A corresponding target feedback interface is generated and updated based on the feedback data, and the target feedback interface provides operation controls or recommended content for updating the knowledge graph of the target user.

9. The data processing method according to claim 1, further comprising: Obtaining interaction data between a target user and the target application, updating the target user's knowledge graph based on the interaction data, and / or generating corresponding positive evaluation content based on the interaction data; The target application is an application that can call the at least one processing model to execute a target processing function, and the interaction data includes data as input to the processing model and feedback data for the inference result output by the processing model.

10. An electronic device comprising at least one target application and at least one processing model capable of running on the target application, wherein the processing model can be called by the target application to perform at least one of the following: In response to obtaining first input data input by a target user into a target application, determining a knowledge graph of the target user and attribute information of the first input data; generating a plurality of sets of guidance data for the first input data based on the knowledge graph and the attribute information, wherein the plurality of sets of guidance data have a target logical relationship between them; as well as, Generate corresponding target response results based on the feedback data of the target user to the multiple sets of guidance data.

Images