A method and system for measuring cognitive gap based on DIKWP model and adaptive interaction gap filling

Abstract

The application discloses a cognitive gap measurement and adaptive interaction bridging method and system based on a DIKWP model, and relates to the field of doctor-patient interaction. The application obtains objective data flow and subjective data flow of a target object, constructs objective semantic space and subjective semantic space, extracts a target event from the objective data flow, maps the target event into objective information nodes and subjective information nodes, and establishes alignment correlation, associates corresponding knowledge layer nodes, wisdom layer nodes and purpose layer nodes along a DIKWP mapping relationship, obtains an event objective semantic subgraph and an event subjective semantic subgraph, respectively generates an objective reference information set and a subjective cognitive information set, calculates information asymmetry components, knowledge misunderstanding components and intention conflict components, synthesizes a cognitive gap index, and when the cognitive gap index exceeds a threshold value, discriminates a main contradiction type according to weighted contribution values of the components, generates an interaction strategy according to the main contradiction type, and executes an interaction action. The application realizes quantitative measurement and adaptive bridging of doctor-patient cognitive differences.

Description

Technical Field

[0001] This invention relates to the field of doctor-patient interaction technology, and in particular to a cognitive gap measurement and adaptive interaction bridging method and system based on the DIKWP model. Background Technology

[0002] In continuous management scenarios such as high-risk pregnancies, existing digital systems often emphasize "risk prediction / stratification," but the following problems frequently arise when implemented in the real world: (1) Existing systems usually output risk prediction results in the form of risk scores or risk levels, but they have not established technical means to incorporate objective medical information and the subjective cognition of the target object into the same semantic framework for structured description and comparison. Risk interpretation cannot accurately match the actual cognitive state of the target object. (2) In the existing technology, the cognitive differences between doctors and patients can only be judged through subjective questionnaires or qualitative experience. There is a lack of technical indicators to distinguish, quantify and uniformly measure information asymmetry, knowledge misconceptions and intention conflicts. The system cannot objectively judge whether the cognitive state of the target object meets the requirements for action execution. (3) Existing interaction methods usually use fixed educational templates, which cannot automatically select targeted strategies according to the type of cognitive bias, nor can they prioritize the dominant contradiction that has the greatest impact on decision-making when multiple cognitive biases coexist, resulting in low interaction efficiency. Summary of the Invention

[0003] In view of the above-mentioned prior art, the present invention provides a cognitive gap measurement and adaptive interaction bridging method and system based on the DIKWP model, which mainly solves the technical problems existing in the background art.

[0004] To achieve the above objectives, the technical solution of this invention is implemented as follows: In a first aspect, the present invention provides a cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model, the method comprising the following steps: S1: Obtain the objective and subjective data streams of the target object, and construct the objective and subjective semantic spaces based on the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer and purpose layer; S2: Extract the target event to be processed from the objective data stream, map it to objective information nodes in the objective semantic space and subjective information nodes in the subjective semantic space, and establish alignment associations to obtain event alignment relationships; S3: Based on the event alignment relationship, associate the knowledge layer node, wisdom layer node and target layer node of the target event in the objective semantic space and the subjective semantic space respectively with the DIKWP model mapping relationship to obtain the event objective semantic subgraph and the event subjective semantic subgraph; S4: Generate an objective reference information set based on the objective semantic subgraph of the event, and generate a subjective cognitive information set based on the subjective semantic subgraph of the event; S5: Calculate the information asymmetry component, knowledge misclassification quantity, and intention conflict component based on the objective reference information set and the subjective cognitive information set; S6: Obtain a cognitive gap index based on the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, determine the main contradiction type based on the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. S7: Generate an interaction strategy based on the main contradiction type, output interaction content and execute interaction actions according to the interaction strategy.

[0005] As a preferred embodiment of the present invention, step S1 specifically includes: The objective data stream includes at least one or more of the following: electronic medical record data, laboratory test data, medical text or image report data; The subjective data stream includes at least one or more of the following: voice inquiry data, questionnaire data, behavioral feedback data, or emotional feedback data.

[0006] As a preferred embodiment of the present invention, step S2 specifically includes: Based on a preset event ontology, events are extracted from the objective data stream to identify the event type, event occurrence time, and event associated object of each event in the objective data stream, thereby obtaining the target event; The target event is mapped to the information layer of the objective semantic space according to the preset event ontology, thereby generating the objective information node; Based on the event type and associated object of the target event, candidate matching is performed in the information layer of the subjective semantic space to calculate the semantic similarity between the target event and each candidate subjective information node. The candidate subjective information node with the highest semantic similarity is determined as the subjective information node corresponding to the target event, and the subjective information node is obtained. The objective information node and the subjective information node are used to establish an alignment association between the target event in the objective semantic space and the subjective semantic space, thereby obtaining the event alignment relationship.

[0007] As a preferred embodiment of the present invention, step S3 specifically includes: Starting from the objective information node, the objective semantic space is traversed along the DIKWP model mapping relationship, and the objective knowledge layer node, objective wisdom layer node and objective purpose layer node corresponding to the target event are respectively associated to obtain the objective semantic subgraph of the event. Starting from the subjective information node, the subjective semantic space is traversed along the DIKWP model mapping relationship, and the subjective knowledge layer node, subjective wisdom layer node, and subjective purpose layer node corresponding to the target event are associated respectively to obtain the event subjective semantic subgraph.

[0008] As a preferred embodiment of the present invention, the specific process of generating an objective reference information set based on the objective semantic subgraph of the event in step S4 includes: The objective knowledge constraint set is obtained based on the rule conditions, causal relationships, and threshold conditions contained in the knowledge layer nodes of the objective semantic subgraph of the event; Based on the prerequisites of the rule conditions contained in the knowledge layer nodes, the decision conditions of the wisdom layer nodes, and the management objectives of the purpose layer nodes in the objective semantic subgraph of the event, the necessary information items are traced back to obtain the set of necessary information items. An objective goal vector is generated based on the management goals of the goal layer nodes in the objective semantic subgraph of the event; The objective reference information set is constructed based on the set of necessary information items, the set of objective knowledge constraints, and the objective objective vector.

[0009] As a preferred embodiment of the present invention, the specific process of generating a set of subjective cognitive information based on the subjective semantic subgraph of the event in step S4 includes: Based on the factual statements contained in the information layer nodes of the event subjective semantic subgraph, a set of known information items is obtained; Based on the explanatory and attributional propositions contained in the knowledge layer nodes of the event's subjective semantic subgraph, a set of subjective belief propositions is obtained; Based on the preferences and action intentions contained in the wisdom layer nodes and the purpose layer nodes in the subjective semantic subgraph of the event, a subjective intention vector is obtained; The set of subjective cognitive information is constructed based on the set of known information items, the set of subjective belief propositions, and the subjective intention vector.

[0010] As a preferred embodiment of the present invention, step S5 specifically includes: Information asymmetry components are calculated based on the semantic differences between the objective reference information set and the subjective cognitive information set; The knowledge misclassification score is calculated based on the knowledge proposition conflict difference between the objective reference information set and the subjective cognitive information set. The intention conflict component is calculated based on the difference in purpose direction between the objective reference information set and the subjective cognitive information set.

[0011] As a preferred embodiment of the present invention, step S6 specifically includes: The cognitive gap index is obtained by weighted summing of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component are calculated respectively. When the weighted contribution value exceeds the preset contribution threshold of the corresponding component type, and the difference between the weighted contribution value and the preset contribution threshold of the corresponding component type is not less than the preset dominant difference, the component type corresponding to the weighted contribution value is determined as the main contradiction type.

[0012] As a preferred embodiment of the present invention, step S7 specifically includes: Candidate strategies are matched from the strategy library according to the main contradiction type, the optimal candidate strategy is selected as the interaction strategy according to the score of each candidate strategy, and the interaction content is generated according to the interaction strategy. When the main contradiction type is information asymmetry, the interaction content includes at least content used to fill in missing information; When the main contradiction type is a knowledge misconception type, the interactive content shall at least include content for correcting knowledge biases; When the primary conflict type is an intent conflict, the interactive content shall include at least content for aligning intent conflicts.

[0013] Secondly, the present invention also provides a cognitive gap measurement and adaptive interaction bridging system based on the DIKWP model, wherein the system is applied to the method described, and the system includes: The data acquisition module is used to acquire the objective data stream and subjective data stream of the target object; The semantic space construction module is used to construct an objective semantic space based on the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer and purpose layer, according to the objective data flow, and to construct a subjective semantic space based on the subjective data flow. The event alignment module is used to extract the target event to be processed from the objective data stream, map the target event to objective information nodes in the objective semantic space and subjective information nodes in the subjective semantic space and establish alignment associations to obtain the event alignment relationship. The semantic subgraph generation module is used to associate the knowledge layer nodes, wisdom layer nodes and target layer nodes of the target event in the objective semantic space and the subjective semantic space respectively according to the event alignment relationship and the DIKWP model mapping relationship, so as to obtain the objective semantic subgraph and the subjective semantic subgraph of the event. The information set generation module is used to generate an objective reference information set based on the objective semantic subgraph of the event, and to generate a subjective cognitive information set based on the subjective semantic subgraph of the event. The gap component calculation module is used to calculate the information asymmetry component, knowledge misclassification quantity, and intention conflict component based on the objective reference information set and the subjective cognitive information set. The main contradiction identification module is used to obtain a cognitive gap index based on the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, the main contradiction type is identified based on the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. The interaction strategy generation and execution module is used to generate an interaction strategy based on the main contradiction type, output interaction content based on the interaction strategy, and execute interaction actions.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention constructs a dual semantic space of objective and subjective information and aligns it with the same target event to achieve a calculable measurement of cognitive differences between doctors and patients; it decomposes the cognitive gap into three interventionable sub-indicators: information asymmetry, knowledge misconceptions, and intention conflicts, making the selection of interaction strategies based on evidence; it adaptively generates interaction strategies according to the type of primary contradiction, effectively improving the pertinence of explanation and the understanding of the target event; it effectively solves the cognitive differences between doctors and patients, and at the same time effectively improves patients' action compliance. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating the steps of a cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model. Figure 2 This is a schematic diagram of a cognitive gap measurement and adaptive interactive bridging system based on the DIKWP model. Detailed Implementation

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. In the following description, the expression "some embodiments" refers to a subset of all possible embodiments; however, it should be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with each other without conflict.

[0017] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the invention can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described in order to avoid obscuring the invention.

[0018] It should be understood that the present invention can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein. Rather, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of the invention to those skilled in the art. Furthermore, the terminology used herein is intended only to describe particular embodiments and is not intended to limit the invention. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “compose” and / or “comprising,” when used in this specification, identify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.

[0019] It should also be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "inner," "outer," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0020] To fully understand this invention, a detailed structure will be presented in the following description to illustrate the technical solution proposed by this invention. Optional embodiments of the invention are described in detail below; however, in addition to these detailed descriptions, the invention may have other embodiments.

[0021] Firstly, this invention provides a cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model, please refer to the appendix. Figure 1 The method includes the following steps: S1: Obtain the objective and subjective data streams of the target object, and construct the objective and subjective semantic spaces based on the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer and purpose layer.

[0022] In this embodiment, the objective data stream includes at least one or more of electronic medical record data, laboratory test data, medical text or image report data.

[0023] The subjective data stream includes at least one or more of the following: voice inquiry data, questionnaire data, behavioral feedback data, or emotional feedback data.

[0024] Furthermore, the specific process of constructing the objective semantic space is as follows: First, the objective data stream is preprocessed. Preprocessing includes de-identification to remove personally identifiable information of the target objects, format standardization to unify the data format of data from different sources, and anomaly or missing value marking to indicate outliers and missing values ​​in the data.

[0025] Secondly, entity identification, unit normalization, and time-series aggregation are performed on the objective data layer nodes to form objective information layer nodes. Entity identification refers to identifying medical entities from the objective data layer nodes, including but not limited to symptoms, signs, laboratory indicators, diagnostic names, and drug names. Unit normalization refers to converting laboratory indicator values ​​with different units of measurement into standard units. Time-series aggregation refers to integrating multiple observations of the same entity in chronological order to form time-series data.

[0026] Next, objective information layer nodes are matched and associated with rule conditions, causal relationships, and threshold conditions in the medical rule base to form objective knowledge layer nodes. The medical rule base includes rule-based knowledge from clinical guidelines, treatment pathways, and medical knowledge graphs. For example, for the information layer node "elevated blood pressure," it can be matched with the rule condition "systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg must be retested within 24 hours," the causal relationship "elevated blood pressure increases the risk of preeclampsia," and the threshold condition "systolic blood pressure ≥160 mmHg or diastolic blood pressure ≥110 mmHg indicates severe hypertension."

[0027] Then, based on the state, risk level, and resource constraints of the target event, contextualized reasoning is performed on the objective knowledge layer nodes to form objective wisdom layer nodes. Contextualized reasoning refers to a comprehensive judgment of the applicability of the knowledge layer nodes by combining the individual characteristics of the target object (such as gestational age, age, and comorbidities), the current risk level (high / medium / low risk), and available medical resources.

[0028] Finally, objective constraints are applied to the objective intelligence layer nodes based on safety objectives, management stages, and medical order requirements to form objective purpose layer nodes. The safety objectives refer to the core objective of ensuring maternal and infant safety; the management stages refer to the current management stage (e.g., pregnancy management, intrapartum management, postpartum management); and the medical order requirements refer to the specific medical orders given by clinicians. Following the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, intelligence layer, and purpose layer, the objective data layer nodes, objective information layer nodes, objective knowledge layer nodes, objective intelligence layer nodes, and objective purpose layer nodes are sequentially associated through generating relational edges to obtain the objective semantic space.

[0029] Furthermore, the specific process of constructing the subjective semantic space is as follows: First, the subjective data stream undergoes semantic parsing, referential resolution, emotion recognition, preference recognition, and statement confidence assessment to obtain subjective data layer nodes with semantic and confidence labels. Semantic parsing refers to converting speech or text data into a structured semantic representation. Referential resolution refers to identifying and parsing the specific objects referred to by pronouns (such as "it," "that," etc.) in the target object's statement. Emotion recognition refers to identifying the emotional states (such as anxiety, calmness, resistance, etc.) implied in the target object's statement. Preference recognition refers to identifying the implicit preference tendencies in the target object's statement.

[0030] Secondly, fact extraction and subjective interpretation extraction are performed on the subjective data layer nodes to form subjective information layer nodes. Fact extraction refers to extracting objective factual information from the description of the target event (e.g., "I measured my blood pressure yesterday"). Subjective interpretation extraction refers to extracting the target event's subjective understanding and interpretation of one's own situation (e.g., "I think my blood pressure is high because I didn't sleep well").

[0031] Next, belief extraction and attribution identification are performed on the subjective information layer nodes to form subjective knowledge layer nodes. Here, belief extraction refers to extracting the target event's belief statements regarding the cause, risk, threshold, consequences, or necessity of action for the disease (e.g., "occasional elevation does not require treatment"). Attribution identification refers to identifying the target event's attribution methods for the event's causes.

[0032] Then, preference trade-offs and value rankings are performed on the subjective knowledge layer nodes to form subjective wisdom layer nodes. The preference trade-offs refer to identifying the priority of a target event across multiple target dimensions (such as security, economic cost, time cost, privacy protection, family participation, etc.).

[0033] Finally, action intention identification and target extraction are performed on the subjective wisdom layer nodes to form subjective purpose layer nodes. Action intention identification refers to identifying whether the target object has a clear willingness or refusal to act. Target extraction refers to extracting the core objectives of the target event in decision-making. Then, according to the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer, and purpose layer, the subjective data layer nodes, subjective information layer nodes, subjective knowledge layer nodes, subjective wisdom layer nodes, and subjective purpose layer nodes are sequentially associated through generating relational edges to obtain the subjective semantic space.

[0034] S2: Extract the target event to be processed from the objective data stream, map it to the objective information node in the objective semantic space and the subjective information node in the subjective semantic space, and establish an alignment association to obtain the event alignment relationship.

[0035] In this embodiment, step S2 specifically includes: Based on a preset event ontology, events are extracted from the objective data stream to identify the event type, event occurrence time, and event associated object of each event in the objective data stream, thereby obtaining the target event; Specifically, the event types include, but are not limited to, symptom events (such as headache, edema), sign events (such as elevated blood pressure, decreased fetal movement), laboratory events (such as positive urine protein), and diagnostic events (such as preeclampsia). The event occurrence time refers to the point in time or time period during which the event occurs or is observed. The event-associated object refers to the medical entity associated with the event (such as specific symptoms, specific laboratory indicators). Through event extraction, candidate events and their attribute information are identified from the objective data stream to obtain the target event to be processed.

[0036] The objective information node and the subjective information node are used to establish an alignment association between the target event in the objective semantic space and the subjective semantic space, and the association identifier, alignment confidence and alignment method of the alignment association are recorded to obtain the event alignment relationship.

[0037] S3: Based on the event alignment relationship, associate the target event with the knowledge layer node, wisdom layer node and target layer node corresponding to the objective semantic space and the subjective semantic space respectively along the DIKWP model mapping relationship to obtain the event objective semantic subgraph and the event subjective semantic subgraph.

[0038] In this embodiment, step S3 specifically includes: Starting with the objective information node, the objective semantic space is traversed along the DIKWP model mapping relationship. Objective knowledge layer nodes, objective wisdom layer nodes, and objective purpose layer nodes corresponding to the target event are associated respectively, resulting in the objective semantic subgraph of the event. The traversal method is as follows: In the objective semantic space, a knowledge layer node with a direct mapping relationship to the objective information node is searched; then, a wisdom layer node with a direct mapping relationship to that knowledge layer node is searched; and finally, a purpose layer node with a direct mapping relationship to that wisdom layer node is searched. The objective information node, along with the associated objective knowledge layer nodes, objective wisdom layer nodes, and objective purpose layer nodes, are then linked together, starting from the subjective information node, to obtain the objective semantic subgraph of the event.

[0039] Similarly, starting from the subjective information node, the subjective semantic space is traversed along the mapping relationship of the DIKWP model, and the subjective knowledge layer node, subjective wisdom layer node, and subjective purpose layer node corresponding to the target event are associated respectively to obtain the event subjective semantic subgraph.

[0040] S4: Generate an objective reference information set based on the event-level objective semantic subgraph, and generate a subjective cognitive information set based on the event-level subjective semantic subgraph.

[0041] In this embodiment, the specific process of generating an objective reference information set based on the objective semantic subgraph of the event includes: The objective knowledge constraint set is obtained based on the rule conditions, causal relationships, and threshold conditions contained in the knowledge layer nodes of the objective semantic subgraph of the event; Based on the prerequisites of the rule conditions contained in the knowledge layer nodes, the decision conditions of the wisdom layer nodes, and the management objectives of the purpose layer nodes in the objective semantic subgraph of the event, the necessary information items are traced back to obtain the set of necessary information items. An objective goal vector is generated based on the management goals of the goal layer nodes in the objective semantic subgraph of the event; The objective reference information set is constructed based on the set of necessary information items, the set of objective knowledge constraints, and the objective objective vector.

[0042] In this embodiment, the specific process of generating a set of subjective cognitive information based on the event subjective semantic subgraph includes: Based on the factual statements contained in the information layer nodes of the event subjective semantic subgraph, a set of known information items is obtained; Based on the explanatory and attributional propositions contained in the knowledge layer nodes of the event's subjective semantic subgraph, a set of subjective belief propositions is obtained; Based on the preferences and action intentions contained in the wisdom layer nodes and the purpose layer nodes in the subjective semantic subgraph of the event, a subjective intention vector is obtained; The set of subjective cognitive information is constructed based on the set of known information items, the set of subjective belief propositions, and the subjective intention vector.

[0043] S5: Calculate the information asymmetry component, knowledge misclassification component, and intention conflict component based on the objective reference information set and the subjective cognitive information set.

[0044] In this embodiment, step S5 specifically includes: The information asymmetry component is calculated based on the semantic differences between the objective reference information set and the subjective cognitive information set.

[0045] Specifically, firstly, based on the set of necessary information items... The importance of each essential information item in the risk assessment is assigned a comprehensive weight. .

[0046] Secondly, for each necessary information item Combine it with the set of known information items The corresponding known information items are matched. The matching criteria are ontology coding consistency, time window overlap, and value range consistency. The coverage is determined based on the matching results. If the necessary information item has a completely matching item in the set of known information items, then If a partial match exists (e.g., the time has expired or the value is inaccurate), then Take the midpoint between 0 and 1; if completely missing, then... .

[0047] Then, combine this with the confidence level of the subjective statement or feedback. Corrections are made. The confidence level is determined by the confidence level assignment during subjective data stream preprocessing. Then, the information asymmetry component... The calculation formula is as follows:

[0048] in, This represents the effective coverage after confidence level correction. Indicates the degree of missing information. This represents the overall weight. For each necessary information item... Based on comprehensive weight Weight the degree of its missingness, then divide by the total weight. Normalization is performed to obtain the weighted average information missing rate. This is achieved when all necessary information items are fully and with high confidence covered. Approaching 0; when all necessary information items are completely missing, Approaching 1, Indicates the target event.

[0049] The knowledge misclassification score is calculated based on the knowledge proposition conflict difference between the objective reference information set and the subjective cognitive information set.

[0050] Specifically, firstly, regarding the set of objective knowledge constraints... Each rule constraint in In the set of subjective belief propositions Search and rule constraints The corresponding belief propositions of the conclusion Then, the rules are constrained. With belief propositions Perform semantic alignment and determine whether the two relationships are "supportive", "partially consistent", "conflicting", or "irrelevant".

[0051] The judgment criterion is as follows: if the subjective belief proposition and the objective knowledge constraint are consistent with each other in the core conclusion and have no substantial difference, then it is judged as "supporting", that is, the conflict coefficient is high. If the core conclusions are consistent but there are discrepancies in degree, scope, or conditions, it is judged as "partially consistent," and the conflict coefficient is [not specified]. The coefficient is determined by linear interpolation between 0.3 and 0.6 based on the degree of deviation; if there is substantial contradiction in the core conclusions, it is judged as "conflict", and the conflict coefficient is... The score is assigned between 0.7 and 1.0 based on the severity of the conflict; if the two involve different semantic categories and are not comparable, they are judged as "irrelevant" and are not included in the knowledge misclassification score. The calculation of the conflict coefficient. For knowledge conflicts related to hard safety constraints, the conflict coefficient... Take 1 directly.

[0052] Then, based on the severity weight of the rule constraints. Apply weighting. Severity weighting. The severity weight is determined by the level of evidence constrained by this rule; the higher the level of evidence, the greater the weight of severity. Therefore, the knowledge misclassification metric... The calculation formula is as follows:

[0053] in, The belief confidence level corresponding to a subjective belief proposition is determined when the set of subjective belief propositions is generated. For each rule constraint, a severity weight is assigned. The product of the conflict coefficient and the belief confidence level is weighted and then divided by the total severity weight. Normalization is performed to obtain the weighted average knowledge conflict rate. This is achieved when all subjective beliefs are consistent with objective knowledge constraints. Approaching 0; when all subjective beliefs severely conflict with objective knowledge constraints and their confidence level is high. Approaching 1.

[0054] The intention conflict component is calculated based on the difference in purpose direction between the objective reference information set and the subjective cognitive information set.

[0055] Specifically, firstly, the objective target vector With subjective intention vector Projected onto the same preset target dimension space.

[0056] Then, calculate the objective target vector. With subjective intention vector The degree of difference between them. The calculation of the degree of difference includes two parts: (1) Directional difference: The directional deviation between two vectors is calculated using the cosine distance, i.e. The range of values ​​is ; (2) Safety hard constraint violation indication: Determine the subjective intent vector Does it violate hard safety constraints? If so, then the binary indicator function... The value is 1; otherwise, it is 0. Specific situations violating hard safety constraints include: intentional actions that may cause vital signs to exceed safe ranges (e.g., systolic blood pressure ≥180 mmHg or ≤80 mmHg, heart rate ≥140 bpm or ≤40 bpm, blood oxygen saturation ≤90%), and intentional actions that may cause irreversible organ damage (e.g., delaying medical attention beyond the "golden time window" recommended by clinical guidelines). Therefore, the intention conflict component... The calculation formula is:

[0057] in, , The weight for directional differences is set to 0.6 by default. The default weight for safety hard constraints is 0.4. The function truncates the calculation results to the interval [0,1]. The further the subjective intent deviates from the objective requirements in the target dimension, or the more it violates hard safety constraints, the more likely it is to fail. The higher. When the subjective intent and objective purpose are completely consistent and do not violate hard safety constraints. Approaching 0; when the subjective intention is completely contrary to the objective purpose or violates hard safety constraints, Approaching 1.

[0058] S6: Obtain a cognitive gap index based on the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, determine the main contradiction type based on the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component.

[0059] In this embodiment, step S6 specifically includes: The cognitive gap index is obtained by weighted summing of the information asymmetry component, the knowledge misclassification component, and the intention conflict component, and its specific formula is as follows:

[0060] in, These are the weights of the information asymmetry component, the knowledge misclassification component, and the intention conflict component, respectively. In this embodiment, the default value is... The adjustment can be dynamically adjusted based on the risk level of the target event and the data quality. The specific adjustment method is as follows: (1) When the risk level of the target event is "high risk", increase Up to 0.5, and Each is reduced to 0.25; when the risk level is "medium risk", the default weight is used; when the risk level is "low risk", the weight is increased. Up to 0.4, It dropped to 0.3. It dropped to 0.3; (2) When the overall confidence level of the subjective data stream is lower than the preset confidence threshold (default 0.6), increase the confidence level. Prioritize information completion while reducing and ; Cognitive Gap Indicators The value range is [0,1]. The larger the value, the greater the cognitive difference between the objective semantic space and the subjective semantic space.

[0061] Furthermore, when the cognitive gap indicator Exceeding the preset threshold At that time, the determination of the main contradiction type is triggered. The default value is set according to the risk level of the target event: high-risk events, preset threshold. Medium-risk events, preset threshold Low-risk events, preset threshold .

[0062] The specific process for identifying the type of principal contradiction is as follows: First, calculate the information asymmetry components separately. Knowledge Misunderstanding Quantity Conflict component of intent The specific formulas for calculating the weighted contribution values ​​of each are as follows:

[0063]

[0064]

[0065] in, The weighted contribution value of the information asymmetry component. The weighted contribution value for knowledge misclassification. The weighted contribution value of the intention conflict component.

[0066] Then, to Sort the data and determine the type of principal contradiction according to the following rules: (1) If the maximum weighted contribution value The corresponding component type meets the following conditions: If the contribution exceeds the preset contribution threshold corresponding to this component type (default value is 0.30), and The difference between the second largest weighted contribution value and the second largest weighted contribution value shall not be less than the preset dominant difference value. (The default value is 0.15), then the component type corresponding to the maximum weighted contribution value is determined as a single-type principal contradiction. That is: when When the principal contradiction is maximized and meets the above conditions, it is of the information asymmetry type; when When the principal contradiction is maximized and meets the above conditions, it is a knowledge misconception type; when When the maximum value is achieved and the above conditions are met, the main contradiction type is intentional conflict.

[0067] (2) If the maximum weighted contribution value The contribution threshold for the corresponding component type or the maximum weighted contribution value is not exceeded. The difference between the second largest weighted contribution value and the second largest contribution value is less than the preset dominant difference. If the weighted contribution values ​​of at least two component types exceed their respective preset contribution thresholds, then the main contradiction type will be classified as a composite type.

[0068] In complex cases, the processing order of each component type is determined according to the priority of the hard safety constraints: if the hard safety constraints are violated or there is a risk of them being violated, then the component with conflict of intent is processed first; if the lack of necessary information makes it impossible to reliably determine the knowledge conflict, then the component with information asymmetry is processed first; if erroneous knowledge directly leads to high-risk behavior, then the component with knowledge misconception is processed first.

[0069] S7: Generate an interaction strategy based on the main contradiction type, output interaction content and execute interaction actions according to the interaction strategy.

[0070] In this embodiment, firstly, candidate strategies are matched from the strategy library based on the type of primary conflict. The strategy library is a pre-configured database containing multiple candidate strategies, each of which is configured with applicable primary conflict type, triggering prerequisites, security constraints, and content generation rules.

[0071] Based on the type of primary contradiction, suitable candidate strategies are selected from the strategy library. Then, the candidate strategies are further selected based on the risk level of the target event, the profile characteristics of the target object (such as health literacy level, language preference) and available interaction channels (such as mobile terminals, SMS, telephone, family terminals, etc.) to obtain a set of candidate strategies.

[0072] Then, the optimal candidate strategy is selected as the interaction strategy based on the scores of each candidate strategy. The formula for calculating the score is as follows:

[0073] in, Candidate strategies Cognitive gap indicators The expected decline Candidate strategies The interaction burden assessment value (including estimated interaction duration and operation complexity). Candidate strategies Misunderstanding risk assessment value (based on the degree of matching between content complexity and the target audience's health literacy level). Candidate strategies The degree to which hard safety constraints are met. The weights for each dimension are set by default to 0.40, 0.20, 0.25, and 0.15, and can be dynamically adjusted according to the event risk level. The candidate strategy that meets the security constraints and has the highest strategy score is selected as the interaction strategy. The interaction strategy must include at least the strategy type, target component, content slot, interaction channel, verification method, stopping condition, and switching condition.

[0074] Furthermore, interactive content is generated based on the interaction strategy. Different types of primary contradictions correspond to different interactive content, as detailed below: (1) When the main contradiction type is information asymmetry, the interaction content shall at least include content for filling in the missing information, specifically including the meaning of the missing information item, the current status or trend, the threshold triggering condition, the information supplementation or retesting method, the completion time limit and the submission channel; (2) When the main contradiction type is knowledge misconception, the interactive content should at least include content for correcting knowledge bias, specifically including professional explanations, popular explanations, action suggestions and consequence comparisons, and update the knowledge layer nodes of the subjective semantic space through restatement verification or situation selection verification. (3) When the main conflict type is intention conflict, the interaction content shall at least include content for aligning intention conflicts, specifically including common goal declaration, conflict dimension identification, set of optional solutions that meet security hard constraints, and minimum acceptable action set; (4) When the main contradiction type is composite, the interactive content corresponding to each component type is generated in sequence according to the processing order, and the number of content slots in a single round of interaction does not exceed the preset cognitive load limit (the default is no more than 5 content slots).

[0075] Finally, the interactive action is executed through the designated interactive channels based on the interactive content and strategy. These interactive channels include, but are not limited to, mobile terminal push notifications, SMS messages, telephone voice calls, family member collaboration, and offline face-to-face communication.

[0076] Furthermore, after a complete interaction round, the system collects feedback data from the target audience on the interactive actions, including comprehension, acceptance, action completion, emotional state, and response time.

[0077] The subjective data stream is updated based on feedback data. New facts are written into the information layer nodes of the subjective semantic space, the restatement or selection results are updated into the knowledge layer nodes of the subjective semantic space, the solution weighing results are updated into the wisdom layer nodes of the subjective semantic space, and the action commitment or rejection results are updated into the purpose layer nodes of the subjective semantic space.

[0078] Based on the updated subjective data stream, steps S1 to S6 are re-executed to recalculate the cognitive gap index. The three components were used to obtain the updated cognitive gap index.

[0079] Calculate the rate of decrease of the updated cognitive gap index relative to the cognitive gap index before this round of interaction. :

[0080] in, This is an indicator of the cognitive gap prior to this round of interaction. For the updated cognitive gap indicator, It should be a very small positive number (default is 0.01) to prevent division by zero errors.

[0081] When the rate of decline Not lower than the preset decline rate threshold (Default value is 0.10) and when the target component continues to decrease, maintain the current strategy or gradually reduce the interaction intensity.

[0082] When the rate of decline Below the preset decline rate threshold At that time, the reasons for failure are determined based on the recalculated weighted contribution values ​​of each component and feedback data: if the target component still dominates but has low comprehension, switch to a more accessible content template; if comprehension is high but acceptance or action completion is low, switch the strategy type (e.g., switch from knowledge correction to intent alignment); if the dominant component changes, switch to the strategy type corresponding to the new dominant component; if the same strategy is used consecutively... Wheel (default) If the decline is insufficient, the channels are inaccessible, or emotional resistance increases, then upgrade the interaction channels.

[0083] Record the reasons for selection and residual diagnosis results for each hold, switch, downgrade, or upgrade.

[0084] The stopping conditions include at least one of the following: cognitive gap indicators Furthermore, all mandatory safety requirements have been met; the target component is below the corresponding threshold and is continuous. Wheel (default) Stable; reaches the maximum number of interaction rounds (default is set to 3~15 rounds depending on the urgency of the event).

[0085] The following specific application example illustrates the implementation effect of the method and system of the present invention.

[0086] For example, consider the "elevated blood pressure" event in high-risk pregnancy management: In step S1, the objective data stream (including blood pressure measurement records in electronic medical records and prenatal examination reports) and subjective data stream (including voice inquiry records such as "I think it only rises occasionally and I don't need to do anything" and questionnaire feedback) of the target object are obtained, and the objective semantic space and subjective semantic space are constructed respectively.

[0087] In step S2, “blood pressure rise” is extracted from the objective data stream as the target event, mapped as an objective information node, and matched with the corresponding subjective information node (the statement “occasional rise does not need to be processed”) in the information layer of the subjective semantic space to establish an alignment association.

[0088] In step S3, the knowledge layer nodes, wisdom layer nodes, and purpose layer nodes corresponding to the objective and subjective sides are associated along the DIKWP mapping relationship to obtain the objective semantic subgraph and the subjective semantic subgraph of the event. The objective side knowledge layer nodes contain the rule "systolic blood pressure ≥140mmHg needs to be retested within 24 hours", and the subjective side knowledge layer nodes contain the belief "occasional elevation does not require treatment".

[0089] In step S4, an objective reference information set is generated based on the objective semantic subgraph (necessary information items include "re-measured blood pressure value" and "accompanying symptoms", objective knowledge constraints include blood pressure threshold rules, and the objective purpose vector points to "ensuring maternal and infant safety"), and a subjective cognitive information set is generated based on the subjective semantic subgraph (known information items include "blood pressure measured yesterday was 145 / 92", subjective beliefs include "occasional elevation does not require treatment", and the subjective intention vector points to "observe first and then decide").

[0090] In step S5, the information asymmetry component is calculated. (The patient knows their blood pressure value but is unaware of the retest interval), knowledge misclassification. (There is a serious conflict between the rule that "occasional elevations do not require treatment" and the rule that "≥140 requires retesting"), indicating a conflict of intent. (Objectively, timely retesting is required; subjectively, observation is preferred first.)

[0091] In step S6, the cognitive gap index is calculated. If the threshold of 0.40 for medium-risk events is exceeded, the main contradiction judgment is triggered. Weighted contribution value If the value is the largest, then the main contradiction type is identified as a knowledge misconception type.

[0092] In step S7, based on the knowledge misconception-based main contradiction generation interaction strategy, the output includes professional explanations ("Pregnancy blood pressure ≥140 / 90 needs to be retested within 24 hours"), popular explanations ("Just like a car's water temperature gauge turning red, don't wait for it to cool down on its own, but check it immediately"), action suggestions ("Please go to the community hospital to retest your blood pressure before 3 pm today") and consequence comparisons ("Timely retesting can avoid the risk of preeclampsia; delay may increase maternal and infant complications"), and is pushed to the patient via mobile terminal.

[0093] After collecting feedback, the subjective data stream is updated and the cognitive gap indicator is recalculated. If the decline is insufficient, switch strategies and continue iterating until the cognitive gap indicator is reached. It dropped below the threshold.

[0094] Secondly, this invention also provides a cognitive gap measurement and adaptive interaction bridging system based on the DIKWP model. This system is applied to the method described herein. (See appendix for details.) Figure 2 The system includes: The data acquisition module is used to acquire the objective data stream and subjective data stream of the target object; The semantic space construction module is used to construct an objective semantic space based on the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer and purpose layer, according to the objective data flow, and to construct a subjective semantic space based on the subjective data flow. The event alignment module is used to extract the target event to be processed from the objective data stream, map the target event to objective information nodes in the objective semantic space and subjective information nodes in the subjective semantic space and establish alignment associations to obtain the event alignment relationship. The semantic subgraph generation module is used to associate the knowledge layer nodes, wisdom layer nodes and target layer nodes of the target event in the objective semantic space and the subjective semantic space respectively according to the event alignment relationship and the DIKWP model mapping relationship, so as to obtain the objective semantic subgraph and the subjective semantic subgraph of the event. The information set generation module is used to generate an objective reference information set based on the objective semantic subgraph of the event, and to generate a subjective cognitive information set based on the subjective semantic subgraph of the event. The gap component calculation module is used to calculate the information asymmetry component, knowledge misclassification quantity, and intention conflict component based on the objective reference information set and the subjective cognitive information set. The main contradiction identification module is used to obtain a cognitive gap index based on the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, the main contradiction type is identified based on the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. The interaction strategy generation and execution module is used to generate an interaction strategy based on the main contradiction type, output interaction content based on the interaction strategy, and execute interaction actions.

[0095] The principle of the system is the same as that of the method, so it will not be described again.

[0096] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model, characterized in that, The method includes the following steps: S1: Obtain the objective and subjective data streams of the target object, and construct the objective and subjective semantic spaces based on the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer and purpose layer; S2: Extract the target event to be processed from the objective data stream, map it to objective information nodes in the objective semantic space and subjective information nodes in the subjective semantic space, and establish alignment associations to obtain event alignment relationships; S3: Based on the event alignment relationship, associate the knowledge layer node, wisdom layer node and target layer node of the target event in the objective semantic space and the subjective semantic space respectively with the DIKWP model mapping relationship to obtain the event objective semantic subgraph and the event subjective semantic subgraph; S4: Generate an objective reference information set based on the objective semantic subgraph of the event, and generate a subjective cognitive information set based on the subjective semantic subgraph of the event; S5: Calculate the information asymmetry component, knowledge misclassification quantity, and intention conflict component based on the objective reference information set and the subjective cognitive information set; S6: Obtain a cognitive gap index based on the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, determine the main contradiction type based on the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. S7: Generate an interaction strategy based on the main contradiction type, output interaction content and execute interaction actions according to the interaction strategy.

2. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 1, characterized in that, Step S1 specifically includes: The objective data stream includes at least one or more of the following: electronic medical record data, laboratory test data, medical text or image report data; The subjective data stream includes at least one or more of the following: voice inquiry data, questionnaire data, behavioral feedback data, or emotional feedback data.

3. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 2, characterized in that, Step S2 specifically includes: Based on a preset event ontology, events are extracted from the objective data stream to identify the event type, event occurrence time, and event associated object of each event in the objective data stream, thereby obtaining the target event; The target event is mapped to the information layer of the objective semantic space according to the preset event ontology, thereby generating the objective information node; Based on the event type and associated object of the target event, candidate matching is performed in the information layer of the subjective semantic space to calculate the semantic similarity between the target event and each candidate subjective information node. The candidate subjective information node with the highest semantic similarity is determined as the subjective information node corresponding to the target event, and the subjective information node is obtained. The objective information node and the subjective information node are used to establish an alignment association between the target event in the objective semantic space and the subjective semantic space, thereby obtaining the event alignment relationship.

4. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 3, characterized in that, Step S3 specifically includes: Starting from the objective information node, the objective semantic space is traversed along the DIKWP model mapping relationship, and the objective knowledge layer node, objective wisdom layer node and objective purpose layer node corresponding to the target event are respectively associated to obtain the objective semantic subgraph of the event. Starting from the subjective information node, the subjective semantic space is traversed along the DIKWP model mapping relationship, and the subjective knowledge layer node, subjective wisdom layer node, and subjective purpose layer node corresponding to the target event are associated respectively to obtain the event subjective semantic subgraph.

5. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 4, characterized in that, In step S4, the specific process of generating the objective reference information set based on the objective semantic subgraph of the event includes: The objective knowledge constraint set is obtained based on the rule conditions, causal relationships, and threshold conditions contained in the knowledge layer nodes of the objective semantic subgraph of the event; Based on the prerequisites of the rule conditions contained in the knowledge layer nodes, the decision conditions of the wisdom layer nodes, and the management objectives of the purpose layer nodes in the objective semantic subgraph of the event, the necessary information items are traced back to obtain the set of necessary information items. An objective goal vector is generated based on the management goals of the goal layer nodes in the objective semantic subgraph of the event; The objective reference information set is constructed based on the set of necessary information items, the set of objective knowledge constraints, and the objective objective vector.

6. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 5, characterized in that, In step S4, the specific process of generating a set of subjective cognitive information based on the subjective semantic subgraph of the event includes: Based on the factual statements contained in the information layer nodes of the event subjective semantic subgraph, a set of known information items is obtained; Based on the explanatory and attributional propositions contained in the knowledge layer nodes of the event's subjective semantic subgraph, a set of subjective belief propositions is obtained; Based on the preferences and action intentions contained in the wisdom layer nodes and the purpose layer nodes in the subjective semantic subgraph of the event, a subjective intention vector is obtained; The set of subjective cognitive information is constructed based on the set of known information items, the set of subjective belief propositions, and the subjective intention vector.

7. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 6, characterized in that, Step S5 specifically includes: Information asymmetry components are calculated based on the semantic differences between the objective reference information set and the subjective cognitive information set; The knowledge misclassification score is calculated based on the knowledge proposition conflict difference between the objective reference information set and the subjective cognitive information set. The intention conflict component is calculated based on the difference in purpose direction between the objective reference information set and the subjective cognitive information set.

8. The cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 6, characterized in that, Step S6 specifically includes: The cognitive gap index is obtained by weighted summing of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component are calculated respectively. When the weighted contribution value exceeds the preset contribution threshold of the corresponding component type, and the difference between the weighted contribution value and the preset contribution threshold of the corresponding component type is not less than the preset dominant difference, the component type corresponding to the weighted contribution value is determined as the main contradiction type.

9. A cognitive gap measurement and adaptive interaction bridging method based on the DIKWP model according to claim 8, characterized in that, Step S7 specifically includes: Candidate strategies are matched from the strategy library according to the main contradiction type, the optimal candidate strategy is selected as the interaction strategy according to the score of each candidate strategy, and the interaction content is generated according to the interaction strategy. When the main contradiction type is information asymmetry, the interaction content includes at least content used to fill in missing information; When the main contradiction type is a knowledge misconception type, the interactive content shall at least include content for correcting knowledge biases; When the primary conflict type is an intent conflict, the interactive content shall include at least content for aligning intent conflicts.

10. A cognitive gap measurement and adaptive interaction bridging system based on the DIKWP model, said system being applied to perform the method as described in any one of claims 1-9, characterized in that, The system includes: The data acquisition module is used to acquire the objective data stream and subjective data stream of the target object; The semantic space construction module is used to construct an objective semantic space based on the DIKWP model mapping relationship of the data layer, information layer, knowledge layer, wisdom layer and purpose layer, according to the objective data flow, and to construct a subjective semantic space based on the subjective data flow. The event alignment module is used to extract the target event to be processed from the objective data stream, map the target event to objective information nodes in the objective semantic space and subjective information nodes in the subjective semantic space and establish alignment associations to obtain the event alignment relationship. The semantic subgraph generation module is used to associate the knowledge layer nodes, wisdom layer nodes and target layer nodes of the target event in the objective semantic space and the subjective semantic space respectively according to the event alignment relationship and the DIKWP model mapping relationship, so as to obtain the objective semantic subgraph and the subjective semantic subgraph of the event. The information set generation module is used to generate an objective reference information set based on the objective semantic subgraph of the event, and to generate a subjective cognitive information set based on the subjective semantic subgraph of the event. The gap component calculation module is used to calculate the information asymmetry component, knowledge misclassification quantity, and intention conflict component based on the objective reference information set and the subjective cognitive information set. The main contradiction identification module is used to obtain a cognitive gap index based on the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. When the cognitive gap index exceeds a preset threshold, the main contradiction type is identified based on the weighted contribution values ​​of the information asymmetry component, the knowledge misclassification quantity, and the intention conflict component. The interaction strategy generation and execution module is used to generate an interaction strategy based on the main contradiction type, output interaction content based on the interaction strategy, and execute interaction actions.

Images