A disease risk early warning system and method based on digital behavior analysis

Abstract

The application provides a disease risk early warning system and method based on digital behavior analysis, and relates to the technical field of disease risk early warning.The system comprises a data collection layer, a feature extraction layer, a risk analysis layer and an early warning output layer.By collecting browsing behavior data, mouse behavior data, search behavior data and time pattern data when a user browses a page, time dimension features, content dimension features and behavior dimension features when the user browses the page are extracted, the time dimension features, the content dimension features and the behavior dimension features are normalized respectively, time feature values, content feature values and behavior feature values are obtained, the time feature values, the content feature values and the behavior feature values are input into a preset risk assessment model, and a risk score is obtained.Finally, disease risk early warning information of the user is determined according to the risk score.The application can realize continuous monitoring and early warning of potential health risks of the user, and improve the timeliness and accuracy of health risk early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of disease risk warning technology, and in particular to a disease risk warning system and method based on digital behavior analysis. Background Art

[0002] In recent years, with the rapid development of internet technology, people's daily lives and work have become increasingly dependent on the internet, generating a large amount of digital behavior data. The digital behavior characteristics corresponding to this data can often reflect the user's physical and mental health status, providing a new research direction for health risk early warning.

[0003] Currently, early warnings of user health risks primarily rely on regular physical examinations and wearable device monitoring. These methods require users to actively cooperate or wear specialized testing equipment, which can be inconvenient to use and lead to poor compliance. Furthermore, traditional health monitoring methods often only capture intermittent data, failing to provide continuous, real-time monitoring of user health status and making it easy to miss important health risk signals. Furthermore, existing health risk warning systems primarily analyze physiological indicators and fail to fully utilize users' daily digital behavior characteristics, which often contain important health risk information. Summary of the Invention

[0004] The purpose of the present invention is to provide a disease risk warning system and method based on digital behavior analysis, which is used to solve the problems existing in the existing health risk warning methods, such as the need for active cooperation of users, discontinuous data collection, and untimely warnings. It can realize continuous monitoring and early warning of users' potential health risks without the need for users to wear any equipment or change their daily behavior habits, and can improve the timeliness and accuracy of health risk warnings.

[0005] In order to achieve the above-mentioned objectives, in the first aspect, the present invention provides a disease risk warning system based on digital behavior analysis, including a data acquisition layer, a feature extraction layer, a risk analysis layer and a warning output layer. The data acquisition layer is used to collect browsing behavior data, mouse behavior data, search behavior data and time pattern data when the user browses the page; the feature extraction layer is used to extract the time dimension characteristics, content dimension characteristics and behavior dimension characteristics when the user browses the page based on the browsing behavior data, mouse behavior data, search behavior data and time pattern data; the risk analysis layer is used to normalize the time dimension characteristics, content dimension characteristics and behavior dimension characteristics respectively to obtain time feature values, content feature values ​​and behavior feature values, and then input the time feature values, content feature values ​​and behavior feature values ​​into a preset risk assessment model to obtain a risk score; the warning output layer is used to determine the user's disease risk warning information based on the risk score.

[0006] According to a disease risk early warning system based on digital behavior analysis provided by the present invention, browsing behavior data includes URL address, page title, access timestamp, source page, dwell time, page content update and HTTP request response; mouse behavior data includes mouse movement trajectory, mouse movement speed, mouse clicks and mouse scrolling; search behavior data includes search keywords, search time, search frequency and search result clicks; time pattern data includes the time distribution of visited pages.

[0007] A data collection layer provided according to the present invention is specifically used to: collect and record URL addresses, page titles, access timestamps, source pages, and dwell time through browser plug-ins, and obtain page content updates and HTTP request responses through DOM change monitoring.

[0008] According to a disease risk warning system based on digital behavior analysis provided by the present invention, the time dimension features include the time distribution pattern of page visits; the content dimension features include the document health relevance of the page; and the behavior dimension features include the switching behavior and attention distribution of the visited pages.

[0009] According to a disease risk warning system based on digital behavior analysis provided by the present invention, the feature extraction layer is specifically used to calculate the document health relevance of the page based on the deep semantic analysis method.

[0010] According to a disease risk early warning system based on digital behavior analysis provided by the present invention, the formula for calculating the document health relevance of a page is:

[0011] H(d)=∑(wi·sim(ti,M))

[0012] Among them, sim(ti,M)=max{cos(v_ti,v_m)|m∈M}, wi=tf(ti)·log(N / df(ti));

[0013] Where H(d) is the health relevance of the document, wi is the weight, ti is the keyword term in the document, M is the medical dictionary set, m is the medical concept, v_ti is the word vector representation of the keyword term, v_m is the word vector representation of the medical concept, sim(ti,M) is the semantic similarity between the keyword term and the medical concept, tf(ti) is the frequency of the keyword term, df(ti) is the number of documents containing the keyword term, and N is the total number of documents.

[0014] According to a disease risk warning system based on digital behavior analysis provided by the present invention, the risk analysis layer is specifically used for: when the time distribution pattern of users' access to pages is manifested as the more frequently the users access, the larger the time feature value output by the risk assessment model; when the health relevance of the document of the page visited by the user is higher, the content feature value output by the risk assessment model is larger; when the user switches between access pages more frequently and the attention is more focused, the behavior feature value output by the risk assessment model is larger.

[0015] According to a disease risk early warning system based on digital behavior analysis provided by the present invention, the risk score is calculated as follows:

[0016] Risk_base = w1×F1+w2×F2+w3×F3

[0017] Where Risk_base is the risk score, w1 is the time feature weight, F1 is the time feature value, w2 is the content feature weight, F2 is the content feature value, w3 is the behavior feature weight, and F3 is the behavior feature value.

[0018] According to a disease risk warning system based on digital behavior analysis provided by the present invention, the disease risk warning information includes the disease risk level; the warning output layer is specifically used to: when the risk score is larger, the output disease risk level is higher, indicating that the user is more likely to suffer from the disease.

[0019] In a second aspect, the present invention provides a disease risk early warning method based on digital behavior analysis, using the disease risk early warning system based on digital behavior analysis of the first aspect, the method comprising:

[0020] Step 1: The data collection layer collects browsing behavior data, mouse behavior data, search behavior data, and time pattern data when users browse pages;

[0021] Step 2: The feature extraction layer extracts the time dimension features, content dimension features, and behavior dimension features of the user browsing the page based on the browsing behavior data, mouse behavior data, search behavior data, and time pattern data;

[0022] Step 3: The risk analysis layer normalizes the time dimension features, content dimension features, and behavior dimension features to obtain time feature values, content feature values, and behavior feature values. The time feature values, content feature values, and behavior feature values ​​are then input into a preset risk assessment model to obtain a risk score.

[0023] Step 4: The warning output layer determines the user's disease risk warning information based on the risk score.

[0024] The technical solution of the present invention has at least the following technical effects:

[0025] The present invention provides a disease risk warning system and method based on digital behavior analysis, which relates to the field of disease risk warning technology. The system includes a data acquisition layer, a feature extraction layer, a risk analysis layer, and a warning output layer. By collecting browsing behavior data, mouse behavior data, search behavior data, and time pattern data when users browse pages, the time dimension features, content dimension features, and behavior dimension features of the users when browsing pages are extracted. The time dimension features, content dimension features, and behavior dimension features are normalized respectively to obtain time feature values, content feature values, and behavior feature values. The time feature values, content feature values, and behavior feature values ​​are then input into a preset risk assessment model to obtain a risk score; finally, the user's disease risk warning information is determined based on the risk score. The present invention can achieve continuous monitoring and early warning of users' potential health risks without the need for users to wear any equipment or change their daily behavior habits, thereby improving the timeliness and accuracy of health risk warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0027] In the attached figure:

[0028] Figure 1 This is a structural block diagram of the disease risk early warning system based on digital behavior analysis of the present invention;

[0029] Figure 2 An early warning curve diagram for case verification of the present invention;

[0030] Figure 3 This is a warning curve diagram for group verification of the present invention. DETAILED DESCRIPTION

[0031] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0032] The following will describe some embodiments of the present invention in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0033] See also Figure 1An embodiment of the present invention provides a disease risk warning system based on digital behavior analysis, including a data collection layer, a feature extraction layer, a risk analysis layer and a warning output layer. The data collection layer is used to collect browsing behavior data, mouse behavior data, search behavior data and time pattern data when the user browses the page; the feature extraction layer is used to extract the time dimension features, content dimension features and behavior dimension features when the user browses the page based on the browsing behavior data, mouse behavior data, search behavior data and time pattern data; the risk analysis layer is used to normalize the time dimension features, content dimension features and behavior dimension features respectively to obtain time feature values, content feature values ​​and behavior feature values, and then input the time feature values, content feature values ​​and behavior feature values ​​into a preset risk assessment model to obtain a risk score; the warning output layer is used to determine the user's disease risk warning information based on the risk score.

[0034] Specifically, the data collection layer uses a multi-source data collection mechanism to collect browsing behavior data when users browse pages. Browsing behavior data includes URL address, page title, access timestamp, source page, duration of stay, page content updates, and HTTP request responses.

[0035] The data collection layer uses a browser plug-in to collect information such as the URL address, page title, access timestamp, source page, and duration of user browsing. This browser plug-in utilizes an event-driven collection architecture, registering listeners for page lifecycle events to continuously track user behavior data. The data collection layer utilizes a layered data caching mechanism, establishing a lightweight local database for initial storage and preprocessing of collected browsing behavior data. The data cache utilizes a three-tiered architecture: in-memory cache for real-time data, local storage for short-term historical data, and cloud storage for long-term data archiving.

[0036] The data collection layer also records user browsing behavior data through browser plug-ins to build a complete page access profile, which not only records basic metadata such as the URL address, page title, access timestamp, source page, and duration of stay. For dynamically loaded content, it also obtains page content updates and HTTP request responses through DOM change monitoring, achieving real-time tracking and recording the time point and specific changes of each page content update. By analyzing HTTP request and response header information, it can identify the page's cache status, redirect history, and server response characteristics, thereby building a complete page access link map.

[0037] The data collection layer also features an adaptive mouse behavior data collection mechanism, capturing mouse movement trajectory, mouse movement speed, and interaction events such as mouse clicks and mouse scrolling. For trajectory sampling, the data collection layer employs a state-based multi-level sampling strategy, categorizing mouse movement speed into five levels: stationary state (0-50 pixels / second), fine manipulation state (50-200 pixels / second), normal movement state (200-800 pixels / second), fast movement state (800-2000 pixels / second), and extreme movement state (greater than 2000 pixels / second). These speed thresholds are derived from natural groupings derived through cluster analysis of mouse operation data from 10,000 users. The corresponding sampling intervals are 2 milliseconds, 5 milliseconds, 10 milliseconds, 20 milliseconds, and 40 milliseconds, respectively. The data collection layer uses a Kalman filter to predict mouse movement in real time and adjust sampling parameters in advance to ensure a balance between sampling accuracy and system performance.

[0038] Specifically, search behavior data includes search keywords, search time, search frequency, and search result clicks; time pattern data includes the time distribution of visited pages.

[0039] The feature extraction layer uses a multi-layered page activity calculation method and monitors the foreground / background state of a page using the Page Visibility API. Combined with actual user interaction events, the feature extraction layer accurately distinguishes between active browsing time and background dwell time, and then extracts temporal features, including the temporal distribution of page visits. In interaction event analysis, the feature extraction layer employs an adaptive interaction threshold adjustment mechanism to balance detection sensitivity and stability. The initial interaction threshold is set to 2 seconds, a median value derived from analyzing large-scale user behavior data, which effectively balances interaction detection sensitivity and stability. The feature extraction layer dynamically adjusts the interaction threshold by analyzing the user's operation frequency distribution in real time. When a user is detected to be in a high-frequency interaction state (e.g., more than 3 interactions per second), the interaction threshold is lowered to 1 second. When a user is in a low-frequency interaction state (e.g., less than 0.5 interactions per second), the interaction threshold is raised to 3 seconds. This adaptive mechanism has demonstrated excellent performance in experiments, accurately capturing the interaction patterns of diverse users. At the same time, the feature extraction layer establishes an interaction intensity calculation model based on a sliding time window. The size of the time window is automatically adjusted according to the user's operation density. For example, it can dynamically change between 10 seconds and 60 seconds.

[0040] In the session analysis mechanism, the feature extraction layer accurately captures user multitasking characteristics through fine-grained behavior tracking. Based on an adaptive session segmentation algorithm, the feature extraction layer dynamically adjusts the session time window to adapt to different users' browsing habits, thereby extracting behavioral characteristics such as attention distribution and page switching behavior. The initial session time window is set to 30 minutes, determined based on large-scale user behavior research and covering 85% of common browsing session patterns. The feature extraction layer dynamically adjusts the session time window size by analyzing users' tab switching behavior. When users exhibit frequent page switching (e.g., an average tab switching interval of less than 1 minute), the session time window is reduced to 15 minutes. When users exhibit prolonged focus (e.g., an average tab switching interval of more than 5 minutes), the session time window is expanded to 45 minutes. This dynamic adjustment mechanism significantly improved session identification accuracy in experiments, from 78% with a traditional fixed session time window to 91%.

[0041] In document health content analysis, the feature extraction layer calculates the document health relevance of the page through deep semantic analysis, that is, extracts content dimension features. For document d, its health relevance H(d) is calculated using the following formula:

[0042] H(d)=∑(wi·sim(ti,M))

[0043] Where ti is the keyword in the document, wi is the weight, m ​​is the medical concept, M is the medical dictionary set, and sim(ti,M) is the semantic similarity between the keyword and the medical concept. The calculation formula is:

[0044] sim(ti,M)=max{cos(v_ti,v_m)|m∈M}

[0045] Here, v_ti and v_m are word vector representations of keywords and medical concepts, respectively. By using these word vector representations, a semantic bridge is established between general text and specialized medical concepts. This approach transcends the limitations of traditional keyword matching and can identify content related to potential health risks.

[0046] The weight wi is calculated using the improved TF-IDF method:

[0047] wi = tf(ti)·log(N / df(ti))

[0048] Here, tf(ti) is the frequency of the keyword term, df(ti) is the number of documents containing the keyword term, and N is the total number of documents. This improved weight calculation method has shown better results than traditional TF-IDF in medical text analysis, improving accuracy by 15% on the test set.

[0049] Specifically, the risk analysis layer is used to: The more frequently a user visits a page, the larger the time feature value output by the risk assessment model; the more health-related the documents on the page the user visits, the larger the content feature value output by the risk assessment model; and the more frequently a user switches between pages and the more focused their attention, the larger the behavioral feature value output by the risk assessment model. For example, if a user frequently visits a page, browsing related pages in the middle of the night or at short intervals (i.e., the user's time distribution shows frequent visits, and the content of the documents on the pages is highly relevant to "colds"), this indicates that the user is concerned about their health and frequently visits the web due to concerns. Furthermore, the more frequently they switch pages and the more focused their attention, the more urgent their attention is and they don't want to miss any information on the page. This can indicate that the user may have symptoms related to "colds" and be at a higher risk of contracting a cold.

[0050] Specifically, the risk score is calculated as follows:

[0051] Risk_base = w1×F1+w2×F2+w3×F3

[0052] In the formula, Risk_base is the risk score, w1 is the time feature weight, F1 is the time feature value, w2 is the content feature weight, F2 is the content feature value, w3 is the behavior feature weight, and F3 is the behavior feature value. The initial configuration of weights is based on expert knowledge and large-scale data analysis, with a content feature weight of 0.4, a time feature weight of 0.3, and a behavior feature weight of 0.3. Furthermore, to take into account the influence of timing factors, a timing adjustment term is introduced into the risk score calculation formula, resulting in the final risk score:

[0053] Risk_final=Risk_base·(1+β·T_factor)

[0054] Where β is the timing adjustment coefficient and T_factor is the timing adjustment term.

[0055] Preferably, β=0.3. In a long-term tracking experiment of 1,000 users, this value amplifies abnormal behavior signals while maintaining a low false alarm rate (less than 5%).

[0056] Disease risk warning information includes disease risk levels, which can be categorized as high, medium, and low, corresponding to a specific risk score range. The warning output layer specifically outputs a higher disease risk level for higher risk scores, indicating a greater likelihood of the user contracting the disease.

[0057] Based on the same inventive concept, another embodiment of the present invention provides a disease risk early warning method based on digital behavior analysis, which uses the disease risk early warning system based on digital behavior analysis of the above embodiment, and includes:

[0058] Step 1: The data collection layer collects browsing behavior data, mouse behavior data, search behavior data, and time pattern data when users browse pages;

[0059] Step 2: The feature extraction layer extracts the time dimension features, content dimension features, and behavior dimension features of the user browsing the page based on the browsing behavior data, mouse behavior data, search behavior data, and time pattern data;

[0060] Step 3: The risk analysis layer normalizes the time dimension features, content dimension features, and behavior dimension features to obtain time feature values, content feature values, and behavior feature values. The time feature values, content feature values, and behavior feature values ​​are then input into a preset risk assessment model to obtain a risk score.

[0061] Step 4: The warning output layer determines the user's disease risk warning information based on the risk score.

[0062] The disease risk early warning system and method based on digital behavior analysis provided by the present invention have demonstrated excellent early warning performance after large-scale experimental verification: the overall prediction accuracy is approximately 55%, obvious abnormalities are approximately 65%, and minor abnormalities are 40%.

[0063] In summary, the disease risk warning system and method based on user digital behavior analysis proposed in the present invention have significant technical advantages and practical application effects. In terms of timeliness, by analyzing the digital behavior characteristics of users, an early warning can be issued an average of 10 days before the actual symptoms appear. Experimental data show that the predicted risk value curve shows an upward trend before the actual risk value rises significantly. This early warning time is 3 to 5 days earlier than the traditional method based on physiological indicator monitoring. This early warning capability provides users and medical institutions with a sufficient time window for preventive intervention, effectively reducing the risk of disease development.

[0064] In terms of early warning accuracy, continuous optimization of the deep learning model has achieved significant performance improvements. As shown in experimental results, the system's early warning accuracy has gradually increased from an initial 75% to 87%, and verification accuracy has increased from 72% to 85%. This high accuracy significantly reduces false alarms, avoids unnecessary waste of medical resources, and ensures the reliability of early warnings.

[0065] Compared to traditional health monitoring methods, this invention uses a non-invasive approach that doesn't require users to wear any equipment or change their daily habits. By analyzing users' daily digital behaviors, health risk warnings are provided, and the data obtained is highly consistent and authentic. This approach not only increases user willingness to use, but also overcomes the issues of large sampling intervals and discontinuous data in traditional methods.

[0066] This invention offers personalized early warning capabilities. By establishing a behavioral baseline for each user, the warning threshold can be dynamically adjusted based on their individual characteristics. Experimental data shows that this personalized early warning mechanism reduces the false alarm rate by 30% and the missed alarm rate by 25% compared to traditional fixed threshold methods. The stable early warning effect is particularly evident across user groups of different age groups, demonstrating its excellent adaptability.

[0067] This invention also has broad applicability. In tests with users aged 20 to 50, it demonstrated stable early warning effectiveness. Furthermore, the design architecture supports expansion of early warning models for a variety of common diseases and can be easily integrated with existing medical systems. This scalability and compatibility lays the foundation for the system's widespread application.

[0068] To verify the practicality and effectiveness of the present invention, a three-month experimental study was conducted from April to July 2024. After obtaining informed consent, a 25-year-old male volunteer A was selected as a typical case for in-depth analysis, and a group verification was performed on an experimental group of 50 people.

[0069] In the case study, volunteer A developed cold symptoms during the experiment. By analyzing his daily digital behavior, early warning was successfully achieved. Figure 2 As shown, nine days before the onset of symptoms, the user's risk score began to rise significantly. This was manifested by increased nighttime activity and frequent searches and browsing of cold-related content. When the risk score exceeded the warning threshold of 0.7, the system promptly issued a health risk warning. Subsequent tracking confirmed that this warning closely matched the actual onset of symptoms, demonstrating the system's early warning capabilities.

[0070] In the group validation study, the present invention selected 50 volunteers aged between 20 and 50 for follow-up study. Figure 3 As shown, the system demonstrated stable early warning effectiveness across different age groups, with an overall accuracy rate of approximately 85%. During the three-month experiment, a total of 35 health events were recorded, of which the system successfully issued early warnings for 30, with only five missed alerts and three false alarms. Particularly noteworthy was the system's average lead time of 10.5 days, providing ample time for timely intervention.

[0071] Experimental data demonstrates that this invention has excellent early warning performance and practical value. While maintaining a high accuracy rate (85.3%), it also achieves a high recall rate (82.7%). Statistical analysis shows that users of different age groups have positive feedback on the system, believing that the early warning information it provides is practically helpful in their health management. The system's average response time is controlled within 1 second, and its stability reaches 99.9%, fully meeting practical application requirements.

[0072] This example demonstrates the feasibility and effectiveness of the present invention in providing early warning of disease risks through specific experimental cases and data analysis. The experimental results demonstrate that the present invention can timely identify potential health risks by analyzing users' digital behavior characteristics, providing a scientific basis for preventive health interventions.

[0073] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the embodiments disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and variations can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A disease risk early warning system based on digital behavior analysis, characterized in that: It includes a data collection layer, a feature extraction layer, a risk analysis layer, and an early warning output layer. The data collection layer is used to collect browsing behavior data, mouse behavior data, search behavior data, and time pattern data when users browse pages; The feature extraction layer is used to extract the time dimension features, content dimension features and behavior dimension features of the user when browsing the page based on the browsing behavior data, mouse behavior data, search behavior data and time pattern data; the risk analysis layer is used to normalize the time dimension features, content dimension features and behavior dimension features respectively to obtain time feature values, content feature values ​​and behavior feature values, and then input the time feature values, content feature values ​​and behavior feature values ​​into a preset risk assessment model to obtain a risk score; The warning output layer is used to determine the user's disease risk warning information based on the risk score; The time dimension features include the time distribution pattern of page visits; the content dimension features include the document health relevance of the page; the behavior dimension features include the switching behavior and attention distribution of the page visits; The feature extraction layer is specifically used to calculate the document health relevance of the page based on the deep semantic analysis method; The calculation formula for the document health relevance of the page is: H(d)=∑(wi·sim(ti,M)) Among them, sim(ti,M)=max{cos(v_ti,v_m)|m∈M}, wi=tf(ti)·log(N / df(ti)); Where H(d) is the health relevance of the document, wi is the weight, ti is the keyword term in the document, M is the medical dictionary set, m is the medical concept, v_ti is the word vector representation of the keyword term, v_m is the word vector representation of the medical concept, sim(ti,M) is the semantic similarity between the keyword term and the medical concept, tf(ti) is the frequency of the keyword term, df(ti) is the number of documents containing the keyword term, and N is the total number of documents.

2. The disease risk early warning system based on digital behavior analysis according to claim 1 is characterized in that: The browsing behavior data includes URL address, page title, access timestamp, source page, dwell time, page content update and HTTP request response; the mouse behavior data includes mouse movement track, mouse movement speed, mouse clicks and mouse scrolling; the search behavior data includes search keywords, search time, search frequency and search result clicks; the time pattern data includes the time distribution of visited pages.

3. The disease risk early warning system based on digital behavior analysis according to claim 2 is characterized in that: The data collection layer is specifically used to collect and record the URL address, page title, access timestamp, source page, and stay duration through browser plug-ins, and obtain page content updates and HTTP request responses through DOM change monitoring.

4. The disease risk early warning system based on digital behavior analysis according to claim 1 is characterized in that: The risk analysis layer is specifically configured to: when the time distribution pattern of user page visits is such that the more frequent the user visits, the larger the time feature value output by the risk assessment model; when the document health relevance of the page visited by the user is higher, the content feature value output by the risk assessment model is larger; The more frequently the user switches access pages and the more focused his attention is, the greater the behavioral feature value output by the risk assessment model.

5. The disease risk early warning system based on digital behavior analysis according to claim 4 is characterized in that: The risk score is calculated as follows: Risk_base = w1×F1+w2×F2+w3×F3 Where Risk_base is the risk score, w1 is the time feature weight, F1 is the time feature value, w2 is the content feature weight, F2 is the content feature value, w3 is the behavior feature weight, and F3 is the behavior feature value.

6. The disease risk early warning system based on digital behavior analysis according to claim 5 is characterized in that: The disease risk warning information includes a disease risk level; the warning output layer is specifically used to: when the risk score is larger, the output disease risk level is higher, indicating that the user is more likely to suffer from the disease.

7. A disease risk early warning method based on digital behavior analysis, characterized in that: Using the disease risk early warning system based on digital behavior analysis according to any one of claims 1 to 6, the method comprises: Step 1: The data collection layer collects browsing behavior data, mouse behavior data, search behavior data, and time pattern data when users browse pages; Step 2: The feature extraction layer extracts the time dimension features, content dimension features and behavior dimension features of the user when browsing the page based on the browsing behavior data, mouse behavior data, search behavior data and time pattern data; Step 3: The risk analysis layer normalizes the time dimension features, content dimension features, and behavior dimension features to obtain time feature values, content feature values, and behavior feature values, and then inputs the time feature values, content feature values, and behavior feature values ​​into a preset risk assessment model to obtain a risk score; Step 4: The warning output layer determines the user's disease risk warning information based on the risk score.

Images