Data analysis method based on brain map

Abstract

The present application proposes a data analysis method based on a mind map, which can generate a series of continuous and orderly interactive data chains by parsing the interactive operation behavior events between the user and the mind map. Furthermore, by performing entity feature and semantic feature analysis on the interactive data chain, the entity feature words that the user is concerned about can be extracted from it. Furthermore, based on the extracted entity feature words and semantic feature words, several data sections can be constructed. Each data section represents an aspect of the user's interaction with the mind map, such as the user's interest distribution, focus, etc. Finally, each data section is encoded to obtain a coding vector. These coding vectors represent the characteristics of the user on each data section in a numerical form, and convert the coding vector into an intuitive and easy-to-understand visual chart. These charts can clearly show the distribution and correlation of the user's characteristics on each data section, helping users to quickly understand the data analysis results.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular to a data analysis method based on brain mapping. Background Art

[0002] In the field of data analysis, with the explosive growth of information, how to effectively extract valuable information from massive data and provide users with personalized data display and interpretation has become an urgent problem to be solved. Traditional data analysis methods are often based on structured data tables, and use statistical analysis, data mining and other technical means to discover the rules and patterns in the data. However, this analysis method based on structured data tables is incapable of processing unstructured data, especially the interaction between users and mind maps (a commonly used visual thinking tool).

[0003] As an intuitive and easy-to-understand thinking tool, mind maps are widely used in many fields such as personal learning, team collaboration, and project management. When users interact with mind maps, a large number of operational behavior events will be generated, which contain rich information about user intentions and preferences. However, how to effectively capture, parse, and utilize these interactive operational behavior events to achieve a deep understanding of user needs is a problem that needs to be solved in the existing technology.

[0004] In the prior art, although there are some methods for analyzing user behavior data, these methods usually only focus on surface information such as the frequency and duration of user behavior, and lack the exploration of the deep intentions behind user behavior. In addition, when dealing with the interaction between users and mind maps, these methods often fail to fully utilize the structured information of mind maps, resulting in inaccurate and incomplete analysis results. Summary of the invention

[0005] The purpose of this application is to propose a data analysis method based on brain maps to solve or alleviate the technical problems existing in the prior art.

[0006] The technical solutions provided by the embodiments of this application are as follows:

[0007] A data analysis method based on brain mapping, comprising:

[0008] Get data retrieval request;

[0009] Parsing the data retrieval request to determine data retrieval content;

[0010] Converting the data retrieval content into a data retrieval vector;

[0011] Based on the data retrieval vector, a match is performed with a semantic annotation index table of unstructured multimodal data in a database to retrieve multimodal data matching the data retrieval request.

[0012] In this application, when using the mind map, users will generate various interactive operation behaviors, such as adding nodes, deleting nodes, editing content, dragging nodes, etc. These behavioral events can directly reflect the user's intentions and preferences. To this end, by parsing the interactive operation behavior events between the user and the mind map, a series of continuous and orderly interactive data chains can be generated. These data chains record every interaction between the user and the mind map in detail, providing a basis for subsequent in-depth analysis. Further, the interactive data chain is analyzed for entity features and semantic features, and entity feature words (such as names, place names, company names, etc.) and semantic feature words (such as themes, categories, emotions, etc.) that users are concerned about can be extracted from them. These feature words directly reflect the user's interests and concerns. Furthermore, according to the extracted entity feature words and semantic feature words, several data sections can be constructed. Each data section represents an aspect of the user's interaction with the mind map, such as the user's interest distribution, focus, etc. Finally, each data section is encoded to obtain a coding vector. These coding vectors represent the characteristics of the user on each data section in a numerical form, providing a data basis for subsequent visual analysis. Using visualization techniques (such as charts, heat maps, relationship network diagrams, etc.), the encoding vectors are converted into intuitive and easy-to-understand visualization charts. These charts can clearly show the characteristic distribution and correlation of users on various data slices, helping users to quickly understand the data analysis results, so that user portraits can be further formed based on the visualized data analysis results. User portraits are a comprehensive description of user characteristics, including user interests, preferences, behavior patterns, and other aspects. Through user portraits, we can have a deeper understanding of user needs and provide strong support for subsequent personalized recommendations and service optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Hereinafter, some specific embodiments of the present application will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0014] Figure 1 This is a flow chart of a data analysis method based on a mind map according to an embodiment of the present application. DETAILED DESCRIPTION

[0015] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the embodiments of the present application should fall within the scope of protection of the embodiments of the present application.

[0016] Figure 1 This is a flow chart of a method for analyzing data based on a brain map according to an embodiment of the present application. Figure 1 As shown, it includes:

[0017] Obtain the user's interactive operation behavior events on the mind map;

[0018] Parsing the mind map based on the interactive operation behavior events to generate an interactive data chain;

[0019] Performing entity feature and semantic feature analysis on the interactive data link to extract entity feature words and semantic feature words;

[0020] Constructing a plurality of data sections according to the entity feature words and the semantic feature words;

[0021] Encoding each of the data slices to obtain a coding vector;

[0022] Based on the coding vector, a visual data analysis result is generated to form a user portrait.

[0023] In this application, when using the mind map, users will generate various interactive operation behaviors, such as adding nodes, deleting nodes, editing content, dragging nodes, etc. These behavioral events can directly reflect the user's intentions and preferences. To this end, by parsing the interactive operation behavior events between the user and the mind map, a series of continuous and orderly interactive data chains can be generated. These data chains record every interaction between the user and the mind map in detail, providing a basis for subsequent in-depth analysis. Further, the interactive data chain is analyzed for entity features and semantic features, and entity feature words (such as names, place names, company names, etc.) and semantic feature words (such as themes, categories, emotions, etc.) that users are concerned about can be extracted from them. These feature words directly reflect the user's interests and concerns. Furthermore, according to the extracted entity feature words and semantic feature words, several data sections can be constructed. Each data section represents an aspect of the user's interaction with the mind map, such as the user's interest distribution, focus, etc. Finally, each data section is encoded to obtain a coding vector. These coding vectors represent the characteristics of the user on each data section in a numerical form, providing a data basis for subsequent visual analysis. Using visualization techniques (such as charts, heat maps, relationship network diagrams, etc.), the encoding vectors are converted into intuitive and easy-to-understand visualization charts. These charts can clearly show the characteristic distribution and correlation of users on various data slices, helping users to quickly understand the data analysis results, so that user portraits can be further formed based on the visualized data analysis results. User portraits are a comprehensive description of user characteristics, including user interests, preferences, behavior patterns, and other aspects. Through user portraits, we can have a deeper understanding of user needs and provide strong support for subsequent personalized recommendations and service optimization.

[0024] Optionally, obtaining the user's interactive operation behavior event on the mind map includes:

[0025] Based on the front-end monitoring component embedded in each node in the mind map, the user's interactive operation on the mind map is monitored to generate the interactive operation behavior event.

[0026] Optionally, the front-end monitoring component embedded in each node in the mind map monitors the user's interactive operation on the mind map to generate the interactive operation behavior event, including:

[0027] Configure an HTML element for each node in the mind map, and render the HTML element to the corresponding node to form an interactive node structure;

[0028] Embedding a JavaScript event listener in the interactive node structure and encapsulating it as a DOM element;

[0029] Based on the DOM element, the user's interactive operation on the mind map is monitored to generate the interactive operation behavior event.

[0030] To this end, configuring HTML elements for each node in the mind map and embedding JavaScript event listeners to monitor the processing of user interaction operations can achieve the following technical benefits:

[0031] (1) Improve the accuracy of capturing user interaction data: By embedding JavaScript event listeners directly into the nodes of the mind map, the user's detailed operation behavior on each node, such as clicking, dragging, editing, etc., can be accurately captured, thereby providing users with a more accurate data analysis basis.

[0032] (2) Enhance the real-time nature of data analysis: Since JavaScript event listeners are run on the front end, they can capture user interaction behaviors in real time without waiting for the response of the back-end server. This greatly speeds up data analysis and enables users to obtain timely feedback.

[0033] (3) Reduced data transmission costs: Since interactive operation behavior data is captured and processed directly at the front end, there is no need to send this data to the back-end server for processing. This reduces the cost of data transmission and also reduces the load on the back-end server.

[0034] To this end, the above process of configuring HTML elements for each node in the mind map and embedding JavaScript event listeners provides the following exemplary code for product implementation:

[0035]

[0036]

[0037] Optionally, configuring an HTML element for each node in the mind map, and rendering the HTML element to the corresponding node to form an interactive node structure, includes:

[0038] Traversing the mind map data structure and adding interactive event logic to recursively iterate each node in the mind map to dynamically create a corresponding HTML element for each node and add it to a specified container;

[0039] Performing interactive logic binding on the designated container to render the HTML element to a corresponding node to obtain a virtual dynamic tree;

[0040] Interactive event delegates are allocated to the virtual dynamic tree to form an interactive node structure.

[0041] To this end, the above process of dynamically creating HTML elements for each node in the mind map and assigning interactive event delegation can achieve the following technical benefits:

[0042] (1) Dynamicity and flexibility: By dynamically creating HTML elements through recursive iteration of the mind map data structure, it can adapt to mind maps of different sizes and structures without the need to predefine a fixed HTML structure, which improves the dynamicity and flexibility of the system.

[0043] (2) Efficient event handling: By assigning interactive event delegates to the virtual dynamic tree, efficient event handling can be achieved. Event delegation allows setting event listeners on parent elements to manage events of their child elements, avoiding the performance overhead of setting listeners for each node separately.

[0044] (3) Easy to maintain and expand: The use of recursion and event delegation makes the code more concise and easy to maintain, and the existing system can be easily expanded when new interactive functions need to be added.

[0045] To this end, the above process of dynamically creating HTML elements for each node in the mind map and assigning interactive event delegation provides the following exemplary code for product implementation:

[0046]

[0047]

[0048] In this example, the createNodeElement function is responsible for dynamically creating HTML elements based on node data, and the renderMindMap function recursively traverses the mind map data structure and creates elements and binds interactive events for each node. In this way, an interactive HTML representation is dynamically generated for a mind map with arbitrary structure and size, and responds to user interactions through event listeners.

[0049] Optionally, embedding a JavaScript event listener in the interactive node structure and encapsulating it as a DOM element includes:

[0050] Create a virtual DOM node, and mount the virtual DOM node on a specified container to form a virtual DOM node tree;

[0051] The virtual DOM node tree is mapped to the interactive node structure, so as to embed a JavaScript event listener in the interactive node structure and encapsulate it as a DOM element.

[0052] To this end, the above process of embedding JavaScript event listeners in the interactive node structure and encapsulating them as DOM elements can achieve the following technical benefits by creating a virtual DOM node tree and mapping it to the interactive node structure:

[0053] (1) Improve performance: Virtual DOM technology first changes the DOM tree in memory and then applies the changed part to the real DOM, reducing the performance loss caused by frequent operations on the real DOM and improving the response speed and fluency of the application.

[0054] (2) Simplify code logic: Using virtual DOM can simplify code logic because developers only need to operate the virtual DOM tree in memory without directly operating the complex real DOM structure. At the same time, virtual DOM also provides a wealth of APIs and tools to facilitate developers to perform state management and component-based development.

[0055] (3) Cross-platform compatibility: Virtual DOM technology does not depend on a specific browser or platform, so it has better cross-platform compatibility. Developers can use the same code base to build applications in different environments, reducing development and maintenance costs.

[0056] To this end, the above process provides the following exemplary codes implemented by the product:

[0057]

[0058] Optionally, mapping the virtual DOM node tree to the interactive node structure to embed a JavaScript event listener in the interactive node structure and encapsulate it as a DOM element includes:

[0059] Instantiating an event handler to map the virtual DOM node tree to the interactive node structure to obtain an enhanced interactive node structure;

[0060] Based on the enhanced interactive node structure, the event attribute of each node is bound and initialized, so as to embed the JavaScript event listener in the interactive node structure and encapsulate it as a DOM element;

[0061] Among them, the enhanced interactive node structure is the interactive node structure after being instantiated by the event processor. Each node contains the original data and structural information, and is also attached with event listening logic for processing user interactions. The event attribute binding and initialization processing of each node refers to the predefined event type on each node, associating it with the corresponding event processing function, and ensuring that when these events are triggered, they can respond to the user's interactive behavior, and finally convert these nodes with event listeners into actual DOM elements through the framework's rendering mechanism.

[0062] To this end, the above process of mapping the virtual DOM node tree to the interactive node structure, embedding JavaScript event listeners therein and encapsulating them as DOM elements can achieve the following technical benefits:

[0063] (1) Improve code reusability and maintainability: By separating the event processing logic from the node data structure, the code is highly modularized, so that changes to node data and modifications to event processing logic can be made independently, improving code reusability and maintainability.

[0064] (2) Improve user interaction experience: By binding and initializing event attributes for each node, timely response and correctness of user interaction are ensured, providing users with a smoother and more consistent interaction experience.

[0065] (3) Reduce the number of DOM operations: Through virtual DOM technology, the number of direct operations on the real DOM is reduced, avoiding frequent DOM updates and redrawing, and improving the performance and response speed of the application.

[0066] To this end, the above process provides the following exemplary codes implemented by the product:

[0067]

[0068]

[0069] In the sample code above, a MindMapNode component is defined to represent the nodes of the mind map, and a click event handler function handleClick is added to it. The selected state of the node is managed through useState, and the handler function is bound to the node element through the onClick attribute. Then, in the MindMap component, the nodes and their child nodes are recursively rendered to form the entire mind map. Finally, the ReactDOM.render method of React is used to render the mind map into the specified container. As a result, an interactive mind map DOM structure with event listeners is obtained.

[0070] Optionally, monitoring the user's interactive operation on the mind map based on the DOM element to generate the interactive operation behavior event includes:

[0071] Parsing the event object of the DOM element to monitor the user's interactive operation on the mind map;

[0072] Generate the corresponding InteractionEvent object based on the monitored interactive operations;

[0073] Performing validity verification on the InteractionEvent object to obtain a valid verification measurement value;

[0074] According to the effective verification metric value, context analysis is performed on the InteractionEvent object to generate the interactive operation behavior event.

[0075] To this end, the above-mentioned process of monitoring the user's interactive operations on the mind map based on DOM elements and generating interactive operation behavior events can achieve the following technical benefits:

[0076] (1) Accurately capture user behavior: By parsing the event objects of DOM elements, every interaction between the user and the mind map can be accurately captured, such as clicking, dragging, zooming, etc.

[0077] (2) Real-time feedback and response: After generating the corresponding InteractionEvent object, the system can provide real-time feedback and response to user behavior, providing an instant user experience.

[0078] (3) Behavior analysis and optimization: Validation and context analysis of InteractionEvent objects can help developers gain a deeper understanding of user usage habits and needs, thereby optimizing the functions and interaction design of mind maps.

[0079] (4) Enhance system security: Through validity verification, invalid or malicious events can be filtered out, enhancing the security and stability of the system.

[0080] To this end, the above process provides the following exemplary codes implemented by the product:

[0081]

[0082]

[0083] Optionally, parsing the event object of the DOM element to monitor the user's interactive operation on the mind map includes:

[0084] Performing callback processing on the event object of the DOM element to determine the event type and event object;

[0085] Determine a trajectory movement diagram of the user interaction in the mind map according to the event type and event object;

[0086] The user's interactive operation on the mind map is monitored according to the trajectory movement diagram of the user interaction in the mind map.

[0087] To this end, the above-mentioned analysis of the event object of the DOM element to monitor the processing process of the user's interactive operation on the mind map can achieve the following technical benefits:

[0088] (1) Clear event type: The event type and event object are determined through callback processing, so that we can clearly know what type of operation the user performed, such as click, drag, scroll, etc.

[0089] (2) Trajectory tracking: Based on the event type and event object, the trajectory of user interaction in the mind map is determined, which helps to track the path and pattern of users browsing the mind map.

[0090] (3) Accurate monitoring: By monitoring the user’s interactive operations on the mind map, real-time data on the user’s interaction with the mind map can be obtained, providing strong support for subsequent user behavior analysis and optimization.

[0091] To this end, the above process provides the following exemplary code (based on JavaScript) for product implementation:

[0092]

[0093]

[0094] Optionally, generating a corresponding InteractionEvent object according to the monitored interactive operation includes:

[0095] Generate a continuous event sequence based on the monitored interactive operations;

[0096] Determining an interactive operation mode according to the continuous event sequence;

[0097] According to the interactive operation mode, a corresponding InteractionEvent object is generated.

[0098] To this end, the above process of generating corresponding InteractionEvent objects based on monitored interactive operations can achieve the following technical benefits:

[0099] (1) Event serialization: By generating a continuous event sequence, the complete process of the user’s interaction with the mind map can be clearly recorded, providing an accurate data basis for subsequent analysis and processing.

[0100] (2) Operation mode recognition: Based on the continuous event sequence, the user's interactive operation mode can be determined, such as dragging, zooming, clicking, etc., which helps the system execute corresponding logic according to different operation modes.

[0101] (3) Standardized event object: The generated InteractionEvent object provides a standardized way to represent user interaction operations, making it easier for different modules or components to communicate and collaborate.

[0102] (4) Improve system responsiveness: Since the system can quickly respond to user operations based on standardized InteractionEvent objects, the system responsiveness and user experience are improved.

[0103] (5) Easy to expand and maintain: By defining clearly defined InteractionEvent objects and event processing logic, it is easy to expand new interactive functions and improve existing functions while reducing maintenance costs.

[0104] Example code:

[0105] To this end, the above process provides the following exemplary code (based on JavaScript) for product implementation:

[0106]

[0107]

[0108] Optionally, the performing validity verification on the InteractionEvent object to obtain a valid verification measurement value includes:

[0109] According to the area and permission configuration data of the interactive operation, the validity of the InteractionEvent object is verified to obtain a valid verification measurement value.

[0110] To this end, the above process of validating the InteractionEvent object to obtain a valid validation metric value can achieve the following technical benefits:

[0111] (1) Security enhancement: By verifying the area and permission configuration data of interactive operations, it can ensure that users can only operate within the permitted scope, preventing illegal or unauthorized operations, thereby improving the security of the system.

[0112] (2) Error reduction: Validation can promptly detect and prevent invalid or erroneous interactive operations, thereby avoiding system errors or data corruption caused by erroneous operations and improving system stability and reliability.

[0113] (3) Improved user experience: When a user attempts to perform an invalid or unauthorized operation, the system can provide immediate feedback to prevent the user from making invalid attempts, thereby saving the user's time and improving the user experience.

[0114] (4) Enhanced system maintainability: Through validity verification, the system can clearly identify which operations are legal and which are invalid, which helps system administrators or developers maintain and optimize the system.

[0115] (5) Strong adaptability: By verifying the region and permission configuration data of interactive operations, the system can flexibly adapt to different permission settings and scenario requirements, improving the adaptability and scalability of the system.

[0116] To this end, the above process provides the following exemplary code (based on Java) implemented by the product:

[0117]

[0118]

[0119] Optionally, performing context analysis on the InteractionEvent object according to the valid verification metric value to generate the interactive operation behavior event includes:

[0120] Removing redundant operations from the InteractionEvent object, merging consecutive similar events, and enhancing event processing to obtain a context analysis event stream;

[0121] According to the effective verification metric value, a behavior path analysis is performed on the context analysis event stream to generate the interactive operation behavior event.

[0122] To this end, the above process of contextual analysis of the InteractionEvent object, including removing redundant operations, merging consecutive similar events, enhancing event processing, and performing behavior path analysis based on effective verification metrics, can achieve the following technical benefits:

[0123] (1) Optimize data processing flow: By removing redundant operations and merging consecutive similar events, repeated calculations in the data processing process can be greatly reduced, the data processing flow can be optimized, and data processing efficiency can be improved.

[0124] (2) Improve data quality: Enhancing the event processing steps can ensure the accuracy and completeness of the data and improve data quality by adding additional contextual information to make the data more meaningful.

[0125] (3) Revealing user behavior patterns: Behavioral path analysis can reveal users’ interactive operation behavior patterns, which is of great significance for understanding user behavior, optimizing user experience, and achieving personalized recommendations.

[0126] (4) Support decision making: Based on the generated interactive operation behavior events, enterprises can understand user needs and market trends more accurately, thereby supporting more effective decision making.

[0127] (5) Improve system performance: Reducing redundant operations and merging similar events can reduce the consumption of system resources such as CPU, memory, and network bandwidth, thereby improving the overall performance of the system.

[0128] The above processing process provides the following exemplary codes implemented by the product:

[0129]

[0130]

[0131] Optionally, parsing the mind map based on the interactive operation to generate an interactive data chain includes:

[0132] Based on the interactive operation, extracting the node objects selected on the mind map to generate a user behavior sequence;

[0133] Performing a breadth-first search on the user behavior sequence to determine a context graph between selected node objects;

[0134] The interactive data link is generated according to the context graph between the node objects.

[0135] To this end, the above-mentioned process of parsing the mind map based on interactive operations to generate an interactive data chain can achieve the following technical benefits:

[0136] (1) Enhanced user experience: By extracting the node objects selected by the user and generating a user behavior sequence, the system can more accurately understand the user's intentions and needs, thereby providing more personalized services or feedback and enhancing the user experience.

[0137] (2) Constructing contextual associations: By determining the context graph between the selected node objects through breadth-first search, the system can construct the association relationships between nodes and form a meaningful contextual network, which provides an important basis for subsequent data analysis and decision support.

[0138] (3) Generate interactive data chains: Based on the context graph between node objects, the system can generate interactive data chains, which are data structures consisting of a series of interrelated nodes and edges. This data structure is not only easy to store and query, but also can clearly show the path and logic of user behavior, which helps to visualize and analyze user behavior.

[0139] (4) Support for complex scenario analysis: When dealing with complex mind map structures and user behaviors, the above processing can effectively capture subtle changes in user behavior and complex relationships between nodes, providing support for data analysis in complex scenarios.

[0140] (5) Improving the intelligence level of the system: Through an in-depth understanding of user behavior and the construction of relationships between nodes, the system can gradually improve its own intelligence level and provide users with more intelligent and accurate services.

[0141] To this end, the above processing process provides the following exemplary codes implemented by the product:

[0142]

[0143]

[0144] Please note that the MindMap class in the sample code is a mind map model, which contains methods for adding nodes and edges, as well as methods for extracting user behavior sequences, breadth-first search, and generating interactive data links. In addition, selected_ids represents the list of node IDs selected by the user.

[0145] Optionally, the performing entity feature and semantic feature analysis on the interactive data link to extract entity feature words and semantic feature words therefrom includes:

[0146] Performing entity feature analysis on the interactive data link based on a set topic model to extract entity feature words therefrom;

[0147] Performing dependency syntactic analysis on the interactive data link based on a set topic model to extract semantic feature words therefrom;

[0148] Performing sentiment analysis on the interactive data link based on a set sentiment analysis model to extract sentiment feature words therefrom;

[0149] Based on the sentiment feature words, the feature distribution correlation between the entity feature words and the semantic feature words is determined.

[0150] To this end, the above-mentioned process of analyzing entity features and semantic features of the interactive data link can achieve the following technical benefits:

[0151] (1) Deep understanding of user intent: Through entity feature analysis, the system can identify key entities in user operations and thus understand the user’s focus and intent. This helps the system provide more accurate and personalized services.

[0152] (2) Improving semantic understanding capabilities: Dependency parsing helps the system understand the grammatical relationship between node objects, thereby grasping the deeper meaning of user operations. This enhances the system’s semantic understanding capabilities, enabling it to handle more complex user requests.

[0153] (3) Emotional recognition and feedback: Sentiment analysis can identify the emotional tendencies in user operations, such as positive, negative, or neutral. This helps the system understand the user's emotional state and adjust the feedback strategy accordingly to enhance the user experience.

[0154] (4) Enhanced feature correlation analysis: Determining the feature distribution correlation between entity feature words and semantic feature words based on sentiment feature words helps the system discover the underlying rules and patterns behind user behavior, providing strong support for further intelligent services such as recommendation and prediction.

[0155] To this end, the above processing process provides the following exemplary codes implemented by the product:

[0156]

[0157]

[0158]

[0159] Optionally, constructing a plurality of data slices according to the entity feature words and the semantic feature words includes:

[0160] Based on the feature distribution relevance, multi-scale fusion is performed on the entity feature words and the semantic feature words to generate a multi-scale semantic link description;

[0161] According to the multi-scale semantic link description, several data slices are constructed.

[0162] To this end, the above processing process can achieve the following technical benefits:

[0163] (1) Improve the hierarchical nature of data understanding: By performing multi-scale fusion of entity feature words and semantic feature words to generate multi-scale semantic link descriptions, semantic information of different granularities can be captured, thereby more comprehensively understanding the meaning behind user behavior and data. This hierarchical understanding helps to achieve more accurate personalized services and recommendations.

[0164] (2) Enhance the pertinence and practicality of data slices: Data slices constructed based on multi-scale semantic link descriptions can provide targeted data analysis perspectives for specific scenarios or needs. Such data slices not only help business personnel quickly understand data, but also provide strong support for decision-making.

[0165] (3) Improve the flexibility of data analysis: Since data slices are constructed based on multi-scale semantic link descriptions, the construction method and parameter settings of slices can be flexibly adjusted according to actual needs. This flexibility makes data analysis more adaptable to the changing needs of different business scenarios.

[0166] To this end, the above processing process provides the following exemplary codes implemented by the product:

[0167]

[0168]

[0169] Optionally, constructing a plurality of data slices according to the multi-scale semantic link description includes:

[0170] Performing relevance analysis on the multi-scale semantic link description to obtain a relevance analysis community;

[0171] The relevance analysis community is perspectived to construct a plurality of data slices.

[0172] The above technical processing process has the following technical benefits:

[0173] (1) Improving the relevance of data facets: By performing relevance analysis on multi-scale semantic link descriptions, a set of highly relevant feature word communities can be obtained. Data facets constructed based on these communities will be more targeted and relevant, thus being able to more accurately reflect specific aspects or trends of the data.

[0174] (2) Enhanced interpretability of data slices: The formation of relevance analysis communities is based on the semantic connections and correlations between feature words, which makes the constructed data slices more interpretable. Business personnel can more easily understand the meaning and logic behind each data slice.

[0175] (3) Improve the flexibility of data analysis: Data slices built based on the relevance analysis community can be flexibly adjusted according to needs. When business needs change, relevance analysis can be re-performed and new data slices can be built to meet new analysis needs.

[0176] To this end, the exemplary code of the above process is:

[0177]

[0178] The above code uses the networkx library to build the graph and uses the best_partition function in the community library to perform community division.

[0179] Optionally, performing relevance analysis on the multi-scale semantic link description to obtain a relevance analysis community includes:

[0180] Based on applying a community detection algorithm, a correlation analysis is performed on the multi-scale semantic link description to obtain a cohesion estimate and a coupling estimate;

[0181] The multi-scale semantic link descriptions are clustered according to the cohesion measurement and the coupling measurement to generate a relevance analysis community.

[0182] To this end, the above processing process can achieve the following technical benefits:

[0183] (1) Improve the accuracy of community division: By applying the community detection algorithm and taking into account the cohesion and coupling evaluation, it is possible to more accurately divide communities of highly relevant feature words. This accuracy helps to build more targeted and effective data slices in the future.

[0184] (2) Enhance the comprehensibility of data structures: Cohesion and coupling measurements can quantify the degree of association between feature words, thereby helping to understand the relationship between different parts of the data structure. This helps business personnel better understand the data and make more informed decisions.

[0185] (3) Optimizing data processing efficiency: By clustering multi-scale semantic link descriptions, a small number of highly correlated communities can be generated, which reduces the complexity of subsequent data processing and improves data processing efficiency.

[0186] To this end, the above processing process provides the following exemplary codes implemented by the product:

[0187]

[0188]

[0189] Note: The affinity = 'precomputed' used in SpectralClustering in the above code means that the input is a pre-computed similarity matrix (such as an adjacency matrix), and the larger the element value in the matrix, the more similar it is. In this example, the weighted adjacency matrix is ​​directly used as the similarity matrix.

[0190] Optionally, the correlation analysis community is perspectively analyzed to construct several data sections, including:

[0191] Determining the centrality of the relevance analysis community;

[0192] Based on the size of the centrality, keywords of the first several relevance analysis communities are extracted in sequence and used as corresponding community labels;

[0193] According to the community tags, a pivot table is constructed to perspective each of the relevance analysis communities to construct a plurality of data sections.

[0194] To this end, the above processing process can achieve the following technical benefits:

[0195] (1) Clarifying community importance: By calculating the correlation and analyzing the centrality of the community, we can clarify the importance or influence of each community in the overall data, which helps to prioritize and analyze important communities in the subsequent process.

[0196] (2) Providing an intuitive data view: Community labels extracted based on centrality can be used as identifiers of data slices, allowing users to intuitively understand the content and characteristics of each data slice, thereby performing data analysis and mining more efficiently.

[0197] (3) Optimize the data analysis process: By constructing a pivot table, you can conduct a detailed perspective on each correlation analysis community, thereby discovering the correlation patterns, trends, or anomalies within the community, and providing support for data-driven decision-making.

[0198] To this end, the above processing process provides the following exemplary codes implemented by the product:

[0199]

[0200]

[0201] Optionally, encoding each of the data slices to obtain a coding vector includes:

[0202] Performing feature cross-analysis between the data slices to determine dependency preferences between the slices;

[0203] Based on the dependency preference between the slices, the data slices are grouped to obtain several groups of data slices;

[0204] Perform principal component analysis on each set of data slices to obtain the principal components of the coded objects;

[0205] The encoding object principal component is mapped to a set encoding space to obtain an encoding vector.

[0206] To this end, the above processing process can achieve the following technical benefits:

[0207] (1) Revealing the dependency between facets: By performing feature cross-analysis on data facets, we can reveal the dependency and association between different data facets. This dependency preference helps us understand the intrinsic connection between data facets and provides an important basis for subsequent data processing and analysis.

[0208] (2) Optimizing data processing efficiency: By grouping data based on inter-aspect dependency preferences, data aspects with similar characteristics or dependencies can be grouped together. This can reduce data redundancy and improve the efficiency of data processing and analysis.

[0209] (3) Extract key features: By performing a principal component analysis on each set of data slices, the main components of the coded object, i.e., the core features of the data, can be extracted. These core features can represent the main information of the entire data slice and provide strong support for subsequent data analysis and mining.

[0210] (4) Convenient for data visualization: By mapping the principal components of the coded object to the set coding space, the coded vectors can be obtained. These coded vectors can be visualized in two-dimensional or three-dimensional space, making the results of data analysis and mining more intuitive and easy to understand.

[0211] To this end, the above processing process provides the following exemplary code for product implementation (using Python as an example, using pandas and scikit-learn libraries):

[0212]

[0213]

[0214]

[0215] Optionally, generating a visualized data analysis result according to the encoding vector includes:

[0216] Determine the dimension of the encoding vector, load the corresponding visualization model according to the size of the dimension, and draw a visualized data analysis result, wherein the visualized data analysis result includes at least one of a scatter plot, a parallel coordinate plot, and a heat map.

[0217] To this end, the above processing process can achieve the following technical benefits:

[0218] (1) Visually display data analysis results: By loading a visualization model that matches the encoding vector dimension, the results of data analysis can be intuitively displayed. Whether it is a scatter plot, parallel coordinates plot or heat map, it can help users quickly understand the relationship and pattern between data.

[0219] (2) Adapting to data of different dimensions: Loading different visualization models according to the dimension of the encoding vector can ensure that the visualization of the data analysis results matches the dimension of the data. In this way, whether the dimension of the data is high or low, a suitable visualization method can be found.

[0220] (3) Improve data exploration efficiency: Through the visualization of data analysis results, users can more intuitively explore the patterns and trends in the data. This intuitive data display method can help users find problems, propose hypotheses and verify them more quickly.

[0221] (4) Enhanced data communication capabilities: Visualized data analysis results can be used not only for individual analysis, but also for data communication and discussion between teams. Through intuitive visualization charts, team members can understand the data more clearly and jointly analyze and solve problems.

[0222] To this end, the above processing process provides the following exemplary code for product implementation (using Python as an example, using matplotlib and seaborn libraries):

[0223]

[0224]

[0225] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A data analysis method based on brain map, characterized in that: include: Obtain the user's interactive operation behavior events on the mind map; Parsing the mind map based on the interactive operation behavior events to generate an interactive data chain; Performing entity feature and semantic feature analysis on the interactive data link to extract entity feature words and semantic feature words; Constructing a plurality of data sections according to the entity feature words and the semantic feature words; Encoding each of the data slices to obtain a coding vector; Generate a visualized data analysis result according to the encoding vector to form a user portrait; The step of obtaining the user's interactive operation behavior event on the mind map includes: Based on the front-end monitoring component embedded in each node in the mind map, the user's interactive operation on the mind map is monitored to generate the interactive operation behavior event; The front-end monitoring component embedded in each node in the mind map monitors the user's interactive operation on the mind map to generate the interactive operation behavior event, including: Configure an HTML element for each node in the mind map, and render the HTML element to the corresponding node to form an interactive node structure; Embedding a JavaScript event listener in the interactive node structure and encapsulating it as a DOM element; Based on the DOM element, the user's interactive operation on the mind map is monitored to generate the interactive operation behavior event.

2. A method for analyzing data based on brain maps according to claim 1, characterized in that: The step of configuring an HTML element for each node in the mind map and rendering the HTML element to the corresponding node to form an interactive node structure includes: Traversing the mind map data structure and adding interactive event logic to recursively iterate each node in the mind map to dynamically create a corresponding HTML element for each node and add it to a specified container; Performing interactive logic binding on the designated container to render the HTML element to a corresponding node to obtain a virtual dynamic tree; Assigning interactive event delegates to the virtual dynamic tree to form an interactive node structure; The step of embedding a JavaScript event listener in the interactive node structure and encapsulating it as a DOM element includes: Create a virtual DOM node, and mount the virtual DOM node on a specified container to form a virtual DOM node tree; The virtual DOM node tree is mapped to the interactive node structure, so as to embed a JavaScript event listener in the interactive node structure and encapsulate it as a DOM element.

3. A method for analyzing data based on brain maps according to claim 2, characterized in that: The step of mapping the virtual DOM node tree to the interactive node structure so as to embed a JavaScript event listener in the interactive node structure and encapsulate the interactive node structure as a DOM element includes: Instantiating an event handler to map the virtual DOM node tree to the interactive node structure to obtain an enhanced interactive node structure; Based on the enhanced interactive node structure, the event attribute of each node is bound and initialized, so as to embed the JavaScript event listener in the interactive node structure and encapsulate it as a DOM element; Among them, the enhanced interactive node structure is the interactive node structure after being instantiated by the event processor. Each node contains the original data and structural information, and is also attached with event listening logic for processing user interactions. The event attribute binding and initialization processing of each node refers to the predefined event type on each node, associating it with the corresponding event processing function, and ensuring that when these events are triggered, they can respond to the user's interactive behavior, and finally convert these nodes with event listeners into actual DOM elements through the framework's rendering mechanism.

4. The method for analyzing data based on brain map according to claim 3, characterized in that: The step of monitoring the user's interactive operation on the mind map based on the DOM element to generate the interactive operation behavior event includes: Parsing the event object of the DOM element to monitor the user's interactive operation on the mind map; Generate the corresponding InteractionEvent object based on the monitored interactive operations; Performing validity verification on the InteractionEvent object to obtain a valid verification measurement value; According to the valid verification metric value, context analysis is performed on the InteractionEvent object to generate the interactive operation behavior event; The parsing of the event object of the DOM element to monitor the user's interactive operation on the mind map includes: Performing callback processing on the event object of the DOM element to determine the event type and event object; Determine a trajectory movement diagram of the user interaction in the mind map according to the event type and event object; The user's interactive operation on the mind map is monitored according to the trajectory movement diagram of the user interaction in the mind map.

5. The method for data analysis based on brain map according to claim 1, characterized in that: The step of parsing the mind map based on the interactive operation to generate an interactive data chain includes: Based on the interactive operation, extracting the node objects selected on the mind map to generate a user behavior sequence; Performing a breadth-first search on the user behavior sequence to determine a context graph between selected node objects; The interactive data link is generated according to the context graph between the node objects.

6. The method for analyzing data based on brain map according to claim 1, characterized in that: The entity feature and semantic feature analysis are performed on the interactive data link to extract entity feature words and semantic feature words; Performing entity feature analysis on the interactive data link based on a set topic model to extract entity feature words therefrom; Performing dependency syntactic analysis on the interactive data link based on a set topic model to extract semantic feature words therefrom; Performing sentiment analysis on the interactive data link based on a set sentiment analysis model to extract sentiment feature words therefrom; Based on the sentiment feature words, the feature distribution correlation between the entity feature words and the semantic feature words is determined.

7. The method for analyzing data based on brain map according to claim 6, characterized in that: The step of constructing a plurality of data sections according to the entity feature words and the semantic feature words includes: Based on the feature distribution relevance, multi-scale fusion is performed on the entity feature words and the semantic feature words to generate a multi-scale semantic link description; According to the multi-scale semantic link description, several data slices are constructed.

8. The method for data analysis based on brain mapping according to claim 1, characterized in that: The step of encoding each of the data slices to obtain a coding vector includes: Performing feature cross-analysis between the data slices to determine dependency preferences between the slices; Based on the dependency preference between the slices, the data slices are grouped to obtain several groups of data slices; Perform principal component analysis on each set of data slices to obtain the principal components of the coded objects; The encoding object principal component is mapped to a set encoding space to obtain an encoding vector.

9. The method for data analysis based on brain mapping according to claim 1, characterized in that: Generating a visualized data analysis result according to the encoding vector includes: Determine the dimension of the encoding vector, load the corresponding visualization model according to the size of the dimension, and draw a visualized data analysis result, wherein the visualized data analysis result includes at least one of a scatter plot, a parallel coordinate plot, and a heat map.

Images