Intelligent prediction and decision support device based on big data analysis
Through intelligent prediction and decision support devices based on big data analysis, the problems of low data transmission efficiency, insufficient storage space and insufficient processing capabilities in traditional data processing methods are solved, and intelligent planning and automated management of data transmission and storage are realized, improving the System performance and user experience.
Patent Information
- Application Number
- CN202411120004.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-12-06
AI Technical Summary
Traditional data processing methods cannot effectively deal with large-scale and high-speed data flows, resulting in low data transmission efficiency, insufficient storage space and inability to keep up with the demand, lack of intelligent prediction and decision-making support functions, resulting in unreasonable resource utilization and data Inefficient processing.
Using intelligent prediction and decision support device based on big data analysis, through data reception, preprocessing, storage, transmission and processing modules, data analysis is performed using association rules and machine learning algorithms, predict data transmission and storage requirements, and automatically adjust resource allocation strategies. , dynamically optimize data transmission and storage paths to improve system performance and stability.
It realizes intelligent planning for data transmission and storage, improves resource utilization efficiency, reduces operating costs, enhances system response speed and flexibility, ensures data timeliness and security, supports multiple data formats, and has good capabilities. Scalability and data compatibility improve user experience and data management efficiency.
Smart Images

Figure CN119088294A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of information technology, and specifically is an intelligent prediction and decision support device based on big data analysis. Background Art
[0002] With the advent of the big data era, data has become one of the most valuable resources in today's society. However, the explosive growth of data volume has put forward higher requirements for data transmission, storage and processing. Traditional data processing methods, such as using simple storage architecture and manual resource management, are often unable to effectively cope with such large-scale, high-speed data flows. These problems are mainly reflected in the following aspects:
[0003] First, data transmission is inefficient. Without intelligent prediction and optimization, data transmission may encounter problems such as network congestion and slow transmission speed, resulting in data not reaching its destination in time. Second, insufficient storage space is also a common problem. As data continues to grow, limited storage space will soon be filled up, and traditional storage expansion methods often involve expensive hardware upgrades and complex configuration processes. Furthermore, data processing capabilities cannot keep up with demand growth. When the amount of data surges, traditional data processing systems may not be able to process the data in a timely manner, resulting in data backlogs and delays.
[0004] Although there are some data transmission and storage solutions on the market, most of them focus on hardware upgrades, network optimization, etc., and most of them can only solve problems in one aspect, but lack a comprehensive solution. More importantly, these solutions often lack intelligent prediction and decision support functions. They cannot dynamically adjust resource allocation strategies based on real-time data, nor can they predict future data transmission and storage needs. This leads to unreasonable resource utilization and low data processing efficiency.
[0005] Based on the above background, the present invention proposes an intelligent prediction and decision support device based on big data analysis, aiming to solve many problems in traditional data processing methods through intelligent prediction and decision support. Summary of the invention
[0006] The present invention proposes an intelligent prediction and decision support device based on big data analysis, which realizes intelligent prediction of data transmission demand and automatically adjusts data storage and transmission strategies according to the prediction results, thereby optimizing resource utilization and improving the overall performance and stability of the system.
[0007] In order to achieve the above-mentioned purpose, the technical scheme of the present invention is as follows: an intelligent prediction and decision support device based on big data analysis, comprising: a data acquisition module composed of a data receiving unit, a data preprocessing unit and a data storage unit, the data receiving unit is used to receive data from various data sources in real time or regularly, the data preprocessing unit is used to clean, convert and standardize the received raw data, and the data storage unit is used to integrate an intelligent caching mechanism and dynamically adjust the data storage location;
[0008] A data transmission module composed of a data transmission unit, wherein the data transmission unit is used to transmit data from a source address to a target address based on a dynamic path;
[0009] The data analysis module is composed of a data processing unit, a model training unit, a prediction analysis unit and a decision support unit. The data processing unit is used to extract key information and features from the transmitted data. The key information and features include but are not limited to the timestamp of the data, key indicator values and abnormal data identification. The model training unit is used to train the preset correlation analysis algorithm using association rules, historical data and real-time data.
[0010] The prediction analysis unit is used to use a preset correlation analysis algorithm to analyze whether the data in the data storage unit is associated with the uploaded data, and if so, to perform subsequent predictions; predict the expected transmission time period of the associated data, and predict whether the storage space is sufficient to store the data to be transmitted;
[0011] The decision support unit is used to reserve or allocate channel resources required for data transmission in advance according to the transmission time prediction results; and automatically adjust the data storage strategy according to the storage space prediction results.
[0012] The following beneficial effects are achieved by adopting the above scheme:
[0013] 1. The prediction and analysis unit can determine the relevance of new data to uploaded data and make subsequent predictions, such as predicting the transmission time period and storage space requirements, thereby realizing intelligent planning of data transmission and storage. The decision support unit allocates channel resources and adjusts storage strategies in advance based on the prediction results, which not only improves the utilization efficiency of resources, but also ensures smooth data transmission and storage.
[0014] Through automated decision support, the need for human intervention is reduced, operating costs are reduced, and the system's responsiveness and flexibility are improved.
[0015] 2. The data receiving unit receives data from multiple data sources in real time or regularly, ensuring the timeliness and diversity of the data. The data preprocessing unit cleans, converts and standardizes the data, effectively improving the data quality and consistency, and providing accurate basic data for subsequent analysis. The data storage unit uses local or cloud storage, providing a flexible and scalable storage solution while ensuring data security.
[0016] 3. The data transmission unit can efficiently transmit data from the source address to the target address, supporting data exchange within the device and between the device and the external system.
[0017] 4. The data processing unit extracts key information and features, providing strong support for model training and predictive analysis. The model training unit uses historical data and real-time data to train the prediction model, so that the model can continuously adapt to data changes and improve the accuracy of prediction.
[0018] 5. The design of the device of the present invention has good scalability and can be flexibly adjusted as the amount of data increases and business needs change. The loose coupling design between the units enables the device to adapt to different application scenarios and data source types.
[0019] 6. The intelligent cache mechanism can dynamically adjust the storage location according to the actual usage of the data, thereby optimizing the overall performance of the system. By reducing unnecessary data transmission and storage operations, the intelligent cache mechanism can reduce the energy and resource consumption of the system. Storing frequently accessed data in the cache can significantly reduce data reading time and improve the response speed of the system. Fast access and response to data can enhance the user experience and increase user satisfaction.
[0020] 7. By using the extracted features, historical data, and real-time data, combined with association rules for model training, the complex relationship between data can be captured more accurately. This method not only considers the co-occurrence frequency between data items, but also considers the conditional probability and lift, thereby improving the accuracy of the prediction. At the same time, association rule mining helps to discover potential connections between data and can provide valuable information for decision makers.
[0021] In summary, this intelligent prediction and decision support device based on big data analysis provides enterprises and organizations with powerful data-driven capabilities through efficient data collection, transmission, analysis and decision support, which helps to achieve more accurate and efficient operation management.
[0022] Furthermore, the data format received by the data receiving unit includes but is not limited to structured data, semi-structured data and unstructured data.
[0023] Beneficial effects: The ability to receive and process multiple data formats enables the device to have a wider range of data compatibility and meet the needs of different data sources and data types. Whether it is structured data from a database, or semi-structured or unstructured data from web pages, log files, etc., it can be effectively received and processed. Due to the diversity of data formats, the device has greater flexibility in handling different business scenarios and needs. The ability to receive multiple data formats promotes the fusion of different types of data and helps to discover the correlation and hidden patterns between data. This data fusion can provide enterprises with a more comprehensive and in-depth perspective and support more complex analysis and decision-making processes.
[0024] Furthermore, the data storage unit is also used to automatically generate an actual database and a virtual database according to the uploaded data, the actual database includes a specific storage path and a specific storage space, and the virtual database includes a related storage path and a related storage space;
[0025] When the prediction and analysis unit determines that the real-time data received by the data receiving unit is consistent with the uploaded data, the real-time data is stored according to the specific storage path and specific storage space in the database;
[0026] When the prediction and analysis unit determines that the real-time data received by the data receiving unit is related to the uploaded data, the real-time data is stored according to the relevant storage path and the relevant storage space in the database;
[0027] When real-time data is stored, the decision support unit will make real-time judgments and corrections on the storage path and storage space.
[0028] Beneficial effects: By automatically generating actual databases and virtual databases, the storage path and storage space of data can be automatically managed, greatly improving the efficiency and accuracy of data management. This automated management reduces manual intervention, reduces the risk of human error, and makes data management more convenient and efficient. According to the characteristics and relationships of the data, the data is allocated to specific storage paths and spaces, realizing the optimal configuration of storage resources. The combined use of actual databases and virtual databases not only ensures the orderly physical storage of data, but also facilitates the rapid retrieval and analysis of data through the logical organization of the virtual database. The rapid judgment of the real-time data by the predictive analysis unit ensures that the data can be stored in the correct path and space. When the real-time data is consistent or related to the uploaded data, the system can respond quickly and store the data in the corresponding location, ensuring the real-time and accuracy of the data. The decision support unit makes real-time judgments and corrections during the data storage process, and can promptly discover and correct the selection errors of the storage path or space, further ensuring the accuracy and reliability of data storage.
[0029] Further, the model training unit utilizes machine learning algorithms, including but not limited to linear regression, decision trees, random forests, and neural networks.
[0030] Beneficial effects: The model training unit provides a variety of machine learning algorithms, allowing the device to adapt to different types of data sets and problem scenarios. This diverse selection helps find the algorithm that best suits a specific problem, thereby improving the accuracy of predictions. Different machine learning algorithms have their own advantages when processing different types of data. For example, linear regression is suitable for exploring the linear relationship between independent variables and dependent variables; decision trees and random forests perform well in handling classification problems; and neural networks are able to handle complex nonlinear patterns. Therefore, the diverse algorithms of the model training unit make the device widely applicable to various data analysis needs.
[0031] Furthermore, the prediction and analysis unit can also predict the speed and stability of data transmission based on historical transmission data and real-time network conditions.
[0032] Beneficial effects: By predicting the speed of data transmission, transmission tasks can be arranged reasonably to avoid large data transmission during network congestion periods, thereby improving transmission efficiency. At the same time, based on the prediction of stability, more reliable transmission paths or protocols can be selected to reduce the risk of data loss or transmission errors. The predictive analysis unit can identify in advance the period when the network is in poor condition or problems may occur, thereby providing users with early warning. Based on the prediction of speed and stability, network resources such as bandwidth and storage capacity can be allocated more efficiently. This helps to ensure the priority of critical data transmission and avoid wasting resources on inefficient or unstable connections.
[0033] Furthermore, the decision support unit can adjust the data transmission strategy according to the prediction result of the prediction analysis unit, wherein the data transmission strategy includes but is not limited to changing the transmission protocol, optimizing the transmission path and allocating more network resources.
[0034] Beneficial effects: Based on the prediction of speed and stability, the decision support unit can dynamically select the optimal transmission protocol, such as TCP, UDP or other dedicated protocols, to adapt to different network environments and data transmission requirements. This can significantly improve the efficiency of data transmission and reduce transmission delays. In the case of poor network conditions, the decision support unit can optimize the transmission path and select a more stable and less congested network route for data transmission. This helps to reduce data loss and transmission errors and improve the stability of network connections. Based on the prediction results, the decision support unit can intelligently allocate more network resources to key data transmission tasks to ensure the timely and accurate transmission of important data. This dynamic resource allocation method can better meet the ever-changing data transmission needs.
[0035] Furthermore, the data storage unit is provided with a variety of data access interfaces and protocols, adopts a distributed storage architecture, and also has data backup and recovery functions.
[0036] Beneficial effects: Providing multiple data access interfaces and protocols means that the storage unit can seamlessly connect to different types of systems and devices. This flexibility means that data access is no longer restricted to specific technologies or platforms, thus meeting diverse data access needs. The distributed storage architecture makes data storage highly scalable, and storage capacity can be easily expanded as the amount of data grows. The data backup function ensures that important data will not be lost in the event of hardware failure, natural disasters or human errors. Backup data can be stored in remote locations, further improving data security. At the same time, the data recovery function enables rapid resumption of normal operations in the event of problems, reducing the risk of business interruption.
[0037] Furthermore, the data transmission module also includes a data encryption unit and a data decryption unit. The data encryption unit is used to encrypt the data before data transmission, and the data decryption unit is used to decrypt the data after the data reaches the target address and restore it to the original data.
[0038] Beneficial effects: By encrypting data through the data encryption unit, it can be ensured that even if the data is intercepted during transmission, it is difficult to be interpreted by unauthorized third parties. This greatly improves the security of data during transmission and protects sensitive information from being leaked. Encryption technology not only protects the confidentiality of data, but also prevents data from being tampered with during transmission to a certain extent. Through the introduction of data encryption units and data decryption units, the data transmission module has achieved significant beneficial effects in data security, privacy protection, compliance, trust improvement, and prevention of data tampering.
[0039] Furthermore, it also includes a system monitoring module for real-time monitoring of the operating status and performance of the device. When an abnormal situation is detected, it can automatically trigger an alarm mechanism and notify the administrator.
[0040] Beneficial effects: The system monitoring module continuously monitors the operating status and performance of the device. Once an abnormal situation is detected, such as equipment failure, performance degradation, or exceeding the preset safety threshold, it will immediately trigger the alarm mechanism. This ensures that problems can be discovered in time so that measures can be taken quickly to repair them and prevent them from deteriorating further. Through real-time monitoring and timely response, the system monitoring module helps maintain high stability of the device. Intervention before potential problems become actual failures can greatly reduce system downtime and ensure continuity of service.
[0041] Furthermore, it also includes a user interaction module, which allows the user to input query conditions, view prediction results and decision suggestions, and can receive manual adjustments or confirmations from the user on the strategies provided by the decision support unit.
[0042] Beneficial effects: The user interaction module provides users with an intuitive and easy-to-use interface, allowing users to easily enter query conditions, view prediction results and decision suggestions. This interactivity improves the user experience and allows users to more actively participate in the formulation and adjustment of data transmission strategies. Users can enter different query conditions according to actual needs to obtain personalized prediction results and decision suggestions. At the same time, users can also manually adjust or confirm the strategies provided by the decision support unit. This flexibility and controllability enable users to make the best decision based on actual conditions. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 It is a framework diagram of an embodiment of the intelligent prediction and decision support device based on big data analysis of the present invention.
[0044] Figure 2 The present invention is a flowchart of an embodiment of an intelligent prediction and decision support device based on big data analysis.
[0045] Figure 3 This is a class diagram of the overall system structure of the intelligent prediction and decision support device based on big data analysis of the present invention.
[0046] Figure 4 It is an ER diagram of the system database in the intelligent prediction and decision support device based on big data analysis of the present invention.
[0047] Figure 5 The present invention is a flow chart of data storage in an intelligent prediction and decision support device based on big data analysis. DETAILED DESCRIPTION
[0048] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.
[0049] The following is further described in detail through specific implementation methods:
[0050] Embodiment 1, basically as attached Figures 1 to 4As shown: An intelligent prediction and decision support device based on big data analysis, including: a data acquisition module composed of a data receiving unit, a data preprocessing unit and a data storage unit, the data receiving unit is used to receive data from various data sources in real time or regularly, and the data format received by the data receiving unit includes but is not limited to structured data, semi-structured data and unstructured data. The data preprocessing unit is used to clean, convert and standardize the received raw data, and the data preprocessing unit is used to automatically identify and process abnormal values, missing values and duplicate values in the data. The data storage unit is used to integrate an intelligent cache mechanism and dynamically adjust the data storage location (that is, automatically store the data in the cache according to the frequency and importance of data access), so as to optimize the overall performance of the system and reduce the energy consumption and resource consumption of the system; the data storage unit is provided with a variety of data access interfaces and protocols, adopts a distributed storage architecture, and also has data backup and recovery functions.
[0051] The data transmission module is composed of a data transmission unit, and the data transmission unit is used to transmit data from a source address to a target address based on a dynamic path (i.e., dynamically selecting the best path to transmit data based on multiple factors such as network conditions, load conditions and cost considerations); the data transmission module also includes a data encryption unit and a data decryption unit, the data encryption unit is used to encrypt data before data transmission, and the data decryption unit is used to decrypt data after the data reaches the target address and restore it to the original data.
[0052] The data analysis module consists of a data processing unit, a model training unit, a prediction analysis unit and a decision support unit. The data processing unit is used to extract key information and features from the transmitted data. The key information and features include but are not limited to the timestamp of the data, key indicator values and abnormal data identification. The model training unit is used to train the preset correlation analysis algorithm using association rules, historical data and real-time data. The model training unit uses machine learning algorithms, including but not limited to linear regression, decision trees, random forests and neural networks.
[0053] The prediction and analysis unit is used to use a preset correlation analysis algorithm to analyze whether the data in the data storage unit is associated with the uploaded data. If so, subsequent predictions are made; the expected transmission time period of the associated data is predicted, and whether the storage space is sufficient to store the data to be transmitted. The prediction and analysis unit can also predict the speed and stability of data transmission based on historical transmission data and real-time network conditions.
[0054] The decision support unit is used to reserve or allocate channel resources required for data transmission in advance according to the transmission time prediction results; and automatically adjust the data storage strategy according to the storage space prediction results. The decision support unit can adjust the data transmission strategy according to the prediction results of the prediction analysis unit, where the data transmission strategy includes but is not limited to changing the transmission protocol, optimizing the transmission path, and allocating more network resources.
[0055] The specific implementation process is as follows:
[0056] Data collection: The data receiving unit starts working and receives data in real time or periodically from various data sources (such as databases, log files, sensor data streams, and social media APIs, etc.). This data may include structured data (such as database tables), semi-structured data, or unstructured data (such as text, images, and videos).
[0057] The data enters the data preprocessing unit for cleaning, conversion and standardization. During the preprocessing process, the unit automatically identifies and processes outliers, missing values and duplicate values in the data to ensure the accuracy and consistency of the data. The preprocessed data is stored in the data storage unit, which can monitor the access frequency and importance of each storage unit in real time. By analyzing the usage pattern of the data, it is predicted which data may be frequently accessed in the future. Based on these predictions, it is possible to identify which data is "hot data" (frequently accessed data) and which is "cold data" (infrequently accessed data). For hot data, it is moved from the original storage location (such as hard disk or SSD) to the cache to speed up access. For cold data, it is moved out of the cache and stored in a more economical and slower storage medium to save valuable cache resources. The data storage unit supports multiple data access interfaces and protocols and adopts a distributed storage architecture to improve data reliability and scalability. At the same time, it also has data backup and recovery functions to ensure data security.
[0058] Data transmission: When the data storage unit needs to transfer data to another location, the data transmission unit will select the optimal transmission path to transmit the data based on the built-in intelligent routing algorithm and the real-time network status. This process is dynamic and will be adjusted according to changes in network conditions. Before data transmission, the data encryption unit encrypts the data to ensure the security of the data during transmission. The encrypted data is transmitted from the source address to the destination address through the network (which may be a local area network, a wide area network or the Internet). After the data arrives at the destination address, the data decryption unit decrypts the data and restores it to the original data for subsequent analysis or use.
[0059] Data analysis and prediction: The data processing unit further processes the transmitted data, extracts key information and features, and prepares for subsequent model training and predictive analysis. For example, first parse the data to identify the fields, formats, and structure information. Directly extract timestamp information from the data field and convert it to a unified time format. Aggregate, calculate, or transform the original data according to business needs and data characteristics to extract key indicator values. Use statistical methods to identify and mark outliers or abnormal behaviors in the data. The model training unit uses the extracted features and historical data, combined with real-time data and association rules (specifically as follows: 1. It represents the frequency of item sets X and Y appearing in the transaction database at the same time, that is, the proportion of transactions containing both X and Y to the total number of transactions. It measures the universality of the rule. 2. It represents the conditional probability that item set Y also appears in the transaction where item set X appears, that is, P(Y|X). It measures the credibility of the rule. 3. It represents the ratio of the probability of Y appearing under the condition that X appears to the probability of Y appearing itself, that is, P(Y|X) / P(Y). When the lift is greater than 1, it means that X and Y are positively correlated; when it is equal to 1, it means that X and Y are independent; when it is less than 1, it means that X and Y are negatively correlated), to train the preset correlation analysis algorithm. A variety of machine learning algorithms can be used here, such as linear regression, decision tree, random forest or neural network, etc., and the appropriate algorithm can be selected according to the specific application scenario and data characteristics. The prediction analysis unit uses the trained model to analyze the data in the data storage unit. First, determine whether the newly received data is associated with the uploaded data. If there is a correlation, subsequent predictions are made, that is, predicting the expected transmission time period of the associated data, predicting whether the storage space is sufficient, etc. That is: first, it is necessary to select a suitable correlation analysis algorithm, such as a neural network or a decision tree, according to business needs and data characteristics, so that it can identify frequent item sets or dependencies between data. Using the selected correlation analysis algorithm, the data is mined to discover the association rules between data items. These rules describe the frequency and pattern of simultaneous occurrence of data items, such as "patients with heart disease are often accompanied by high blood pressure" or "patients often experience specific side effects after taking a certain drug". When new data is received and stored in the data storage unit, the prediction analysis unit will determine whether the data is associated with the uploaded data by comparing the characteristics of the new data with the characteristics of the association rules in the historical data. If the new data is associated with the uploaded data, the prediction analysis unit will use the time series prediction model or the analysis based on the historical transmission records to predict the expected transmission time period of the associated data. At the same time, it will also evaluate the current capacity and future growth trend of the data storage unit to predict whether the storage space is sufficient to accommodate the upcoming data. In addition, the prediction analysis unit can also predict the speed and stability of data transmission based on historical transmission data and real-time network conditions.
[0060] Decision support: The decision support unit makes corresponding decisions based on the prediction results of the prediction analysis unit. For example, according to the transmission time prediction results, the channel resources required for data transmission are reserved or allocated in advance; according to the storage space prediction results, the data storage strategy is automatically adjusted (such as compressing data, deleting old data, or expanding storage capacity).
[0061] In addition, the decision support unit can also predict the results based on the speed and stability of data transmission and adjust the data transmission strategy. This may include changing the transmission protocol (such as switching from TCP to UDP to improve real-time performance), optimizing the transmission path (selecting a route with lower network latency) or allocating more network resources (such as increasing bandwidth).
[0062] Through the above-mentioned specific implementation process, this intelligent prediction and decision support device based on big data analysis can realize the full process automation processing from data collection, transmission to analysis and decision support, and provide efficient and accurate data-driven decision support for various application scenarios.
[0063] Based on the above embodiments, the system of the present invention is applied in the medical industry, as follows:
[0064] The data receiving unit starts to capture information from a variety of data sources. These data sources may include the hospital's information system database, log files generated by medical devices, and sensor data streams that monitor patients' vital signs in real time, or even API data on disease discussions and drug reactions on social media. The received data comes in a variety of forms, including structured electronic medical records, semi-structured medical reports (such as XML or JSON format), and unstructured medical images and scanned images of doctors' handwritten notes.
[0065] Once the data is received, it enters the data preprocessing unit to remove errors or invalid information, such as unreasonable data values and format errors. At the same time, the missing parts of the data will be estimated and filled, and duplicate records will be merged or deleted to ensure the integrity and accuracy of the data set. The preprocessed data is then sent to the data storage unit, and the data storage location is dynamically adjusted according to the frequency and importance of data access.
[0066] Data transmission: When data needs to be transferred from the data storage unit to another location, the data transmission unit will select the optimal transmission path to transmit the data based on the built-in intelligent routing algorithm and the real-time network status. Before the data is sent, the data encryption unit will use advanced encryption algorithms to protect the data to ensure that even if the data is intercepted during the transmission process, it cannot be interpreted by unauthorized personnel. The encrypted data packet is then transmitted through the network, which can be a local area network within the hospital, a wide area network between cities, or even the Internet covering the world. After the data arrives at the destination, the data decryption unit will decrypt it and restore it to the original, readable data format, ready for subsequent analysis and processing.
[0067] Data analysis and prediction: The data processing unit further processes the received data and extracts key information features and patterns. This information may be timestamp information: consultation time, examination time, medication time, etc., and converts these timestamps into a unified time format. Key indicator values: patients' average blood sugar level, blood pressure change trend, frequency of use of specific drugs, etc. These indicators are important for assessing patients' health status and monitoring disease progression. Abnormal value detection: abnormal physiological indicators of patients, improper medication or potential complications, etc. It is crucial for subsequent disease prediction and treatment plan formulation.
[0068] The model training unit uses these features and past data records, combined with the latest real-time data, to train and optimize the prediction model, based on specific application requirements and data characteristics, such as using linear regression to predict the incidence of a disease, or using neural networks to identify abnormal lesions in medical images.
[0069] The predictive analysis unit conducts in-depth analysis of the data in the data storage unit. The predictive analysis unit selects appropriate association analysis algorithms (such as cluster analysis, association rule mining, or deep learning models) that can identify frequent item sets or dependencies in medical data. Using the selected association analysis algorithm, the data is mined in depth to discover association rules between data items. That is, concurrent relationships between different diseases, drug interactions, or specific characteristics of patient groups. For example, by analyzing the patient's medical history, diagnosis results, and medication records, the co-occurrence patterns between certain diseases or the correlation between treatment plans can be discovered.
[0070] When new data is received and stored in the data storage unit, the prediction and analysis unit will quickly extract features from the new data and compare and analyze it with the established association rules in the historical data. Through this process, it is determined whether the new data is associated with the uploaded medical data. If the new data is highly associated with the old data, the prediction and analysis unit will further use the time series prediction model or the analysis method based on the historical transmission record to predict the transmission demand of medical data in the future. This includes predicting the expected transmission time period of the associated data and evaluating the urgency of the transmission of these data. At the same time, the prediction and analysis unit will also comprehensively consider the current capacity of the data storage unit, the historical growth trend, and the expected growth of medical data in the future, and predict whether the storage space is sufficient to accommodate the upcoming medical data. This is crucial for timely expansion of storage capacity and avoiding data loss. In addition to storage space prediction, the prediction and analysis unit can also evaluate the speed and stability of data transmission based on historical transmission data and real-time network conditions. This is of great significance to ensure the timely and accurate transmission of medical data, especially in scenarios such as telemedicine and emergency rescue.
[0071] Decision support: The decision support unit is the "brain" of the entire system. It makes smart decisions based on the results provided by the predictive analysis unit. For example, if it predicts that the demand for data transmission will surge in a certain period of time in the future, it will reserve or allocate more network channel resources in advance to ensure smooth data transmission. If it predicts that storage space will be insufficient, it may automatically start the data compression program, delete expired data, or send an alert to the system administrator to expand storage capacity.
[0072] In addition, the decision support unit will adjust the transmission strategy based on the speed and stability of data transmission. If it predicts that the network condition is poor, it may choose to change the transmission protocol to improve the real-time nature of the data, or choose a transmission path with lower network latency, or even increase network resource allocation to improve transmission efficiency.
[0073] The device of the present invention can monitor the data flow of various departments in the hospital in real time, ensuring that the patient's electronic medical records, diagnostic reports, medical images and other data can be quickly and safely shared between different departments and doctors. When it is predicted that a department is about to receive a large number of new patients, the device can automatically adjust the data storage and transmission strategy to ensure that doctors can obtain the patient's complete medical record information in a timely manner, so as to make accurate diagnosis and treatment plans. At the same time, the device can also use historical data and machine learning algorithms to predict disease incidence and epidemic trends, and provide hospitals with targeted resource allocation and prevention and control strategy recommendations. In this way, the hospital can operate more efficiently and provide better medical services to patients.
[0074] Embodiment 2 is different from the above embodiments in that it also includes a system monitoring module for real-time monitoring of the operating status and performance of the device. When an abnormal situation is detected, it can automatically trigger an alarm mechanism and notify the administrator.
[0075] The specific implementation process is as follows: when the device of the present invention is started, the system monitoring module is also started and initialized and configured. The configuration content includes setting monitoring indicators (such as CPU usage, memory usage, network bandwidth, etc.), alarm thresholds, and alarm notification methods (such as email, SMS or system notification).
[0076] The system monitoring module collects the operation data of each module of the device regularly (such as every second or every minute), including but not limited to CPU usage, memory consumption, disk I / O, and network transmission speed, etc. These data are stored in the monitoring data warehouse for subsequent analysis and query.
[0077] The system monitoring module analyzes the collected data in real time to determine whether each indicator is within the normal range. If an indicator exceeds the preset alarm threshold, it is considered an abnormal situation. Once an abnormal situation is detected, the system monitoring module will automatically trigger the alarm mechanism. According to the notification method set during the initial configuration, the system monitoring module will send an alarm notification to the administrator. The notification can include a detailed abnormal report and recommended solutions. The abnormal report includes information such as abnormal indicators, abnormal time, and possible impact.
[0078] All detected abnormal situations and corresponding alarm records will be saved in the abnormal record database. The administrator can conduct subsequent analysis based on these records to find out the cause of the abnormality and take measures to prevent similar situations from happening again.
[0079] By adding a system monitoring module, the intelligent prediction and decision support device of the present invention adds the ability to monitor the device's own status and performance in real time during the entire process of data collection, transmission, analysis and decision support. This not only improves the reliability of the device of the present invention, but also ensures that it can respond quickly and restore services when problems occur, thereby providing users with more stable and efficient data-driven decision support.
[0080] Embodiment 3 differs from the above embodiments in that it also includes a user interaction module, which allows the user to input query conditions, view prediction results and decision suggestions, and can receive manual adjustments or confirmations from the user on the strategies provided by the decision support unit.
[0081] The specific implementation process is as follows: users can enter query conditions on the interface, such as time range, data type, and keywords. The user interaction module retrieves relevant data from the data storage unit based on the query conditions entered by the user. At the same time, the prediction results generated by the prediction analysis unit and the decision suggestions of the decision support unit can be displayed on the interface. Users can clearly see various prediction data and decision plans to make quick judgments.
[0082] Users can provide feedback on prediction results and decision suggestions, such as adjusting the parameters of the prediction model, optimizing the decision strategy, etc. Users can also manually adjust or confirm the strategy provided by the decision support unit to ensure that the decision meets actual needs.
[0083] By introducing the user interaction module, the intelligent prediction and decision support device of the present invention not only realizes the automation of the whole process, but also gives users more participation and control. Users can intuitively view the prediction results and decision suggestions, and manually adjust or confirm them according to actual needs, thereby improving the flexibility and practicality of the system. This user-friendly design helps to promote effective interaction between users and the device, thereby improving overall work efficiency and decision quality.
[0084] Embodiment 4 is different from the above embodiment in that, as shown in the attached Figure 5 As shown: the data storage unit is also used to automatically generate an actual database and a virtual database according to the uploaded data. The actual database includes a specific storage path and a specific storage space, and the virtual database includes a related storage path and a related storage space.
[0085] When the prediction and analysis unit determines that the real-time data received by the data receiving unit is consistent with the uploaded data, the real-time data is stored according to the specific storage path and specific storage space in the database; when the prediction and analysis unit determines that the real-time data received by the data receiving unit is related to the uploaded data, the real-time data is stored according to the relevant storage path and relevant storage space in the database; when the real-time data is stored, the decision support unit will make real-time judgments and corrections on the storage path and storage space.
[0086] The specific implementation process is as follows: the data storage unit builds an actual database based on the uploaded data. When the real-time data is the same as the uploaded data, the real-time data is stored according to the specific storage path and specific storage space in the actual database. When the real-time data is different from the uploaded data but related, the real-time data can be stored according to the related storage path and related storage space in the virtual data. During the storage process, the specific storage path and specific storage space (or related storage path and related storage space) are judged in real time.
[0087] Users upload data through the data receiving unit. This data can be various types of information, such as text, images, videos, etc. After receiving the uploaded data, the data storage unit first parses the data content and identifies key information such as data type, source, and relevance. Based on the specific storage requirements and characteristics of the data (such as file size, access frequency, and data type, etc.), a specific storage path is generated and the corresponding storage space is allocated to build an actual database. At the same time, according to the relevance between data (such as metadata, tags, and timestamps, etc.), a virtual database is built to record the relevant storage paths and storage spaces for quick retrieval and association analysis.
[0088] The data receiving unit continuously receives real-time data from various sources. The prediction analysis unit uses a pre-set algorithm or model to quickly analyze the received real-time data to determine its matching degree or relevance with the uploaded data.
[0089] If the real-time data completely matches a certain uploaded data, it is prepared to be stored according to the specific storage path and storage space in the actual database. If the real-time data is not exactly the same as the uploaded data but is related (such as belonging to the same topic, category or adjacent time series), it is prepared to be stored according to the relevant storage path and storage space in the virtual database.
[0090] According to the judgment result of the prediction analysis unit, the data storage unit sends the real-time data to the designated storage path and storage space. During the data storage process, the decision support unit monitors the storage operation in real time, including the selection of storage path, the utilization rate of storage space, etc. If it is found that the storage path or storage space is improperly selected (such as insufficient space, wrong path, etc.), the decision support unit will immediately start the correction mechanism: automatically adjust the storage path to a more appropriate location. Specific examples are supplemented as follows:
[0091] Assume that a user uploads a set of time series data through a data receiving unit. This set of data represents the value of a monitoring indicator at different time points. The specific data is as follows: 1.1 1.1 1.1. The uploaded data is simply represented by timestamps (T1, T2, T3) and corresponding data values (1.1, 1.1, 1.1).
[0092] The data storage unit builds an actual database based on this set of uploaded data. The specific storage path and storage space may be allocated according to the system's internal rules or user settings, for example: / data / timeseries / T1_T2_T3 / (specific storage path); sufficient disk space is allocated to store this set of data (specific storage space).
[0093] In practical applications, there may be complex correlations between data. According to the time series characteristics of the data, a virtual database is constructed to record the storage path and storage space of related data. However, since this set of data itself is very similar, the construction of the virtual database may focus more on advanced features such as the continuity of timestamps or the changing trend of data values.
[0094] Real-time data 1 is completely matched, that is, when the real-time data receiving unit receives the following data: 1.1 1.1 1.1, the prediction analysis unit will determine that this set of real-time data is completely consistent with the uploaded data. Therefore, the data storage unit stores this set of real-time data according to the specific storage path / data / timeseries / T1_T2_T3 / in the actual database.
[0095] Real-time data 2 related data, that is, when the real-time data receiving unit receives another set of data: 1.0 1.1 1.1, this set of real-time data is not exactly the same as the uploaded data, but there is a correlation (they all belong to the same time series and the data values are similar). The prediction analysis unit will recognize this correlation and instruct the data storage unit to store it according to the relevant storage path recorded in the virtual database (if any, or generate a new relevant path according to the preset rules). Assume that we store this set of data in the path / data / timeseries / T0_T1_T2 / according to the continuity of the time series.
[0096] During the data storage process, the decision support unit continuously monitors the storage operation. If any storage path or storage space problem (such as insufficient space, etc.) is found, it will immediately initiate the correction mechanism. For example, if the storage space under the / data / timeseries / T0_T1_T2 / path is insufficient, the decision support unit may automatically adjust the storage path to another location with sufficient space, such as / data / timeseries_backup / T0_T1_T2 / , and ensure that the integrity and consistency of the data are not affected.
[0097] The above is only an embodiment of the present invention, and the common knowledge such as the known specific structure and / or characteristics in the scheme is not described in detail here. It should be pointed out that for those skilled in the art, several deformations and improvements can be made without departing from the structure of the present invention, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.
Claims
1. An intelligent prediction and decision support device based on big data analysis, characterized in that: include: A data acquisition module consisting of a data receiving unit, a data preprocessing unit and a data storage unit. The data receiving unit is used to receive data from various data sources in real time or periodically. The data preprocessing unit is used to clean, convert and standardize the received raw data. The data storage unit is used to integrate an intelligent caching mechanism and dynamically adjust the data storage location. A data transmission module composed of a data transmission unit, wherein the data transmission unit is used to transmit data from a source address to a target address based on a dynamic path; The data analysis module is composed of a data processing unit, a model training unit, a prediction analysis unit and a decision support unit. The data processing unit is used to extract key information and features from the transmitted data. The key information and features include but are not limited to the timestamp of the data, key indicator values and abnormal data identification. The model training unit is used to train the preset correlation analysis algorithm using association rules, historical data and real-time data. The prediction analysis unit is used to analyze whether the data of the data storage unit is associated with the uploaded data according to the correlation analysis algorithm, and if so, to make subsequent predictions; predict the expected transmission time period of the associated data, and predict whether the storage space is sufficient to store the data to be transmitted; The decision support unit is used to reserve or allocate channel resources required for data transmission in advance according to the transmission time prediction result; Automatically adjust data storage strategies based on storage space prediction results.
2. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The data format received by the data receiving unit includes but is not limited to structured data, semi-structured data and unstructured data.
3. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The data storage unit is also used to automatically generate an actual database and a virtual database according to the uploaded data, the actual database includes a specific storage path and a specific storage space, and the virtual database includes a related storage path and a related storage space; When the prediction and analysis unit determines that the real-time data received by the data receiving unit is consistent with the uploaded data, the real-time data is stored according to the specific storage path and specific storage space in the database; When the prediction and analysis unit determines that the real-time data received by the data receiving unit is related to the uploaded data, the real-time data is stored according to the relevant storage path and the relevant storage space in the database; When real-time data is stored, the decision support unit will make real-time judgments and corrections on the storage path and storage space.
4. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The model training unit utilizes machine learning algorithms including but not limited to linear regression, decision trees, random forests, and neural networks.
5. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The predictive analysis unit can also predict the speed and stability of data transmission based on historical transmission data and real-time network conditions.
6. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The decision support unit can adjust the data transmission strategy according to the prediction result of the prediction analysis unit, wherein the data transmission strategy includes but is not limited to changing the transmission protocol, optimizing the transmission path and allocating more network resources.
7. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The data storage unit is equipped with multiple data access interfaces and protocols, adopts a distributed storage architecture, and also has data backup and recovery functions.
8. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: The data transmission module also includes a data encryption unit and a data decryption unit. The data encryption unit is used to encrypt the data before data transmission, and the data decryption unit is used to decrypt the data after the data reaches the target address and restore it to the original data.
9. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: It also includes a system monitoring module for real-time monitoring of the operating status and performance of the device. When an abnormal situation is detected, it can automatically trigger an alarm mechanism and notify the administrator.
10. The intelligent prediction and decision support device based on big data analysis according to claim 1, characterized in that: It also includes a user interaction module, which allows the user to input query conditions, view prediction results and decision suggestions, and can receive manual adjustments or confirmations from the user on the strategies provided by the decision support unit.
Citation Information
Cited By
Data format conversion method, system and equipment and storage medium
CN119336830A
