An information data processing recommendation system

Abstract

The present invention discloses an information data processing recommendation system, which belongs to the field of data processing technology. The system resource management module is responsible for monitoring the node resource usage, allocating and managing tasks, and dividing resources in queues, monitoring the task completion, and generating new docking tasks and screening tasks; the task management module receives the instruction distribution program and structure file of the system resource management module to the task calculation module, and reports the task completion and resource utilization to the system resource management module; the task calculation module uses the fusion knowledge graph reasoning to construct the relationship graph between drug data, extracts key features from the processed drug data, and based on the extracted features, uses machine learning or deep learning algorithms to realize the recommendation of drug combinations, and stores the processed results in a database and feeds back to the task management module. The invention has the advantage of improving the scheduling and allocation efficiency of computing resources.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to an information data processing recommendation system. Background Art

[0002] When various materials or items are combined to form a new composition, their respective functions can cooperate with each other, thus showing many unique effects and effects. This phenomenon is widely manifested and applied in many fields, such as drug combination, scientific research and other fields. In the field of drug combination, by combining different drug ingredients, more significant therapeutic effects can be produced than using each drug alone. This is because the functions of different drugs can cooperate with each other to achieve better therapeutic effects. For example, combining a drug that can lower blood pressure with another drug that can dilate blood vessels can more effectively control hypertension. In addition, in many other fields, the combination of different materials or items can also bring many innovative effects. For example, in artistic creation, artists often use different materials and techniques to combine to create unique works of art. In architectural design, by combining different building materials and structures, buildings with unique styles and functions can be created.

[0003] When various materials or items are combined to form a new combination, the field of data processing faces many challenges, one of which is the huge and complex amount of data. Data combination is an important part of data processing, and the amount of data involved is usually very large, which requires a lot of computing resources. Therefore, there is an urgent need for an information data processing recommendation system that can improve the scheduling and allocation efficiency of computing resources. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention provides an information data processing recommendation system, which has the advantage of improving the scheduling and allocation efficiency of computing resources and solves the problems of the prior art.

[0005] The present invention is implemented as follows: an information data processing recommendation system includes a central processing module, wherein the central processing module is signal-connected to a system resource management module, a task management module, and a task calculation module;

[0006] System resource management module, responsible for monitoring node resource usage, allocating and managing tasks, dividing resources into queues, monitoring task completion, and generating new docking tasks and screening tasks;

[0007] The task management module receives the instruction distribution program and structure file from the system resource management module to the task calculation module, and reports the task completion status and resource utilization status to the system resource management module;

[0008] The task calculation module uses fused knowledge graph reasoning to construct a relationship graph between drug data, extracts key features from the processed drug data, and based on the extracted features, uses machine learning or deep learning algorithms to recommend drug combinations. The processed results are stored in the database and fed back to the task management module, and the database data is displayed on the display.

[0009] As a preferred embodiment of the present invention, the system resource management module includes the following units:

[0010] Resource monitoring unit, responsible for real-time monitoring of resource usage of each node in the system, including CPU, memory, disk space, and network bandwidth;

[0011] The task allocation unit allocates tasks to appropriate nodes according to the current system resource status and task priority to ensure efficient execution of tasks;

[0012] The queue management unit divides and manages tasks in queues, implements batch scheduling and flexible control of resources, and meets the resource requirements of different tasks;

[0013] Resource sharing unit, used for sharing resources between nodes. When a node is short of resources, it can borrow resources from other nodes.

[0014] Flexible scheduling unit dynamically adjusts task allocation and scheduling strategies according to task execution status and system resource changes to ensure system stability and smooth completion of tasks;

[0015] The task monitoring unit monitors the completion of tasks in real time, including task progress, execution time, and result output, so that tasks are executed as planned and achieve the expected results;

[0016] The task generation and screening unit generates new docking tasks and screening tasks according to system requirements and business logic, and screens and optimizes tasks through algorithms.

[0017] As a preferred embodiment of the present invention, the task allocation unit is connected to a database signal, and the task allocation process is optimized in combination with calculation results in the database.

[0018] As a preferred embodiment of the present invention, the task management module includes the following units:

[0019] Instruction receiving and distributing unit: responsible for receiving instructions from the system resource management module, and distributing the corresponding programs and structure files to the task computing module according to the instruction content;

[0020] Task status monitoring unit: monitors the execution status of tasks in real time, including task progress and resource usage. Through real-time feedback, the system resource management module can keep abreast of the latest developments of tasks.

[0021] Resource utilization reporting unit: collects resource utilization data of the task computing module, compiles the data into reports, and sends them to the system resource management module regularly.

[0022] As a preferred embodiment of the present invention, the task calculation module is connected to a database, and data in the database is retrieved during task calculation.

[0023] As a preferred embodiment of the present invention, the task calculation module is refined to include the following units:

[0024] The knowledge graph fusion unit is responsible for integrating and fusing drug information from multiple sources. Through data cleaning and integration algorithms, the scattered drug information is aggregated into a unified graph structure.

[0025] Drug data processing unit, which receives and processes raw drug data, including data cleaning, format conversion and standardization;

[0026] Feature extraction unit: identifies and quantifies key features and extracts key features from processed drug data;

[0027] Drug combination recommendation unit: Based on the extracted features, combined with the drug relationships and feature information in the knowledge graph, using machine learning or deep learning algorithms, trains the model to predict the effect of drug combinations and implements drug combination recommendations;

[0028] The first database storage and feedback unit: stores the processed results in the database and feeds them back to the task management module.

[0029] As a preferred embodiment of the present invention, the task status monitoring unit signal is connected to the knowledge graph fusion unit;

[0030] The knowledge graph fusion unit signal is connected to the drug combination recommendation unit;

[0031] The drug combination recommendation unit signal is connected to the task status monitoring unit.

[0032] As a preferred embodiment of the present invention, the unexecuted task data of the system resource management module can be sent to a database, and the task calculation module can call the unexecuted task data.

[0033] As a preferred embodiment of the present invention, the task calculation module includes:

[0034] Data preprocessing unit: collects raw drug data from hospital information systems and performs data cleaning, deduplication, and standardization;

[0035] Knowledge graph construction unit: Based on the preprocessed drug data, a drug knowledge graph is constructed. The nodes in the graph represent drugs, and the edges represent the relationship between drugs.

[0036] Bipartite graph matrix construction unit: Based on the knowledge graph, the drugs and their relationships are mapped into a bipartite graph. The nodes on one side of the graph represent the drugs, and the nodes on the other side represent the relationship types. By constructing a bipartite graph matrix, the relationship strength between the drugs is quantified to provide data support for the subsequent recommendation algorithm;

[0037] Matrix feature mapping unit: uses machine learning technology to map the nodes and edges in the bipartite graph matrix into a low-dimensional vector space;

[0038] Recommendation algorithm unit: Based on the results of matrix feature mapping, for a given drug combination, the algorithm calculates other drug combinations similar to the combination, and sorts and recommends them based on the similarity;

[0039] The second data storage and feedback unit stores the processed results in the database and feeds them back to the task management module.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] First, the present invention can significantly improve the utilization rate of computing resources and reduce computing costs through resource sharing and flexible scheduling of the system resource management module. Secondly, the real-time reporting function of the task management module enables the system to monitor the task completion and resource utilization in real time, providing decision support for the system resource management module. Finally, the task calculation module uses the fusion of knowledge graph reasoning and machine learning algorithms to more accurately extract the key features of drug data, achieve more accurate drug combination recommendations, and provide strong support for the field of pharmaceutical research and development. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a system block diagram of an information data processing recommendation system provided by an embodiment of the present invention;

[0043] Figure 2 is a system block diagram of a system resource management module provided by an embodiment of the present invention;

[0044] Figure 3 It is a system block diagram of a system resource management module connected to a task allocation unit and a database provided by an embodiment of the present invention;

[0045] Figure 4 is a system block diagram of a task management module provided by an embodiment of the present invention;

[0046] Figure 5is a system block diagram of a task computing module provided by an embodiment of the present invention;

[0047] Figure 6 is a flowchart of an information data processing recommendation system provided by an embodiment of the present invention;

[0048] Figure 7 It is a system block diagram of a task computing module provided by another embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to further understand the content, features and effects of the present invention, the following embodiments are given as examples and described in detail with reference to the accompanying drawings.

[0050] The structure of the present invention is described in detail below in conjunction with the accompanying drawings.

[0051] See also Figure 1-Figure 7 , an information data processing recommendation system provided by an embodiment of the present invention comprises a central processing module, wherein the central processing module is signal-connected to a system resource management module, a task management module and a task calculation module;

[0052] System resource management module, responsible for monitoring node resource usage, allocating and managing tasks, dividing resources into queues, monitoring task completion, and generating new docking tasks and screening tasks;

[0053] The task management module receives the instruction distribution program and structure file from the system resource management module to the task calculation module, and reports the task completion status and resource utilization status to the system resource management module;

[0054] The task calculation module uses fused knowledge graph reasoning to construct a relationship graph between drug data, extracts key features such as efficacy, safety, side effects, etc. from the processed drug data, and uses machine learning or deep learning algorithms based on the extracted features to recommend drug combinations. The processed results are stored in the database and fed back to the task management module, and the database data is displayed on the display.

[0055] In the above setup, the information data processing recommendation system works as follows:

[0056] 1. Initialization and resource monitoring:

[0057] After the system is started, the central processing module first activates the system resource management module and starts monitoring the resource usage of the nodes. The system resource management module divides computing resources into queues to ensure effective utilization and flexible scheduling of resources.

[0058] 2. Task allocation:

[0059] The system resource management module allocates and manages tasks according to the task queue and current resource conditions. For new tasks, the system resource management module generates new docking tasks and screening tasks to ensure the timeliness and accuracy of tasks.

[0060] 3. Task distribution and receipt:

[0061] The task management module receives instructions from the system resource management module and distributes related programs and structure files to the task calculation module. At the same time, the task management module is also responsible for receiving task completion and resource utilization reports from the task calculation module.

[0062] 4. Data processing and calculation:

[0063] After receiving the task, the task computing module uses the fusion knowledge graph reasoning to construct a relationship graph between drug data. Key features such as efficacy, safety, side effects, etc. are extracted from the constructed relationship graph. Based on these extracted features, the task computing module uses machine learning or deep learning algorithms to recommend drug combinations.

[0064] 5. Result storage and feedback:

[0065] The task calculation module stores the processed drug combination recommendation results in the database. At the same time, it feeds the results back to the task management module so that it can report the task completion status to the system resource management module.

[0066] 6. Data shows:

[0067] The task computing module or other parts of the system (such as the data visualization module) displays the data in the database to the user through a display, making it convenient for the user to view and analyze.

[0068] 7. Cycle monitoring and adjustment:

[0069] The system resource management module continuously monitors the completion of tasks and resource utilization, and redistributes and adjusts tasks as needed. If there are new tasks or requirements, the system will repeat the above steps to ensure timely processing of tasks and effective utilization of resources.

[0070] Working principle of this system: The system resource management module first monitors the resource usage of the node to ensure the reasonable allocation and efficient use of resources. When there is a new task, the system resource management module will divide the resources according to the queue method and generate corresponding docking tasks and screening tasks. After receiving the instruction, the task management module will distribute the program and structure files to the task calculation module for calculation. The task calculation module uses the fusion knowledge graph reasoning to construct the drug data relationship map, extract key features, and recommend drug combinations through machine learning or deep learning algorithms.

[0071] Beneficial effects of this system: First, through the resource sharing and flexible scheduling of the system resource management module, the utilization rate of computing resources can be significantly improved and the computing cost can be reduced. Secondly, the real-time reporting function of the task management module enables the system to monitor the task completion and resource utilization in real time, providing decision support for the system resource management module. Finally, the task calculation module uses the fusion of knowledge graph reasoning and machine learning algorithms to more accurately extract the key features of drug data, achieve more accurate drug combination recommendations, and provide strong support for the field of medical research and development.

[0072] Specifically, the system resource management module includes the following units:

[0073] Resource monitoring unit, responsible for real-time monitoring of resource usage of each node in the system, including CPU, memory, disk space, and network bandwidth;

[0074] The task allocation unit allocates tasks to appropriate nodes according to the current system resource status and task priority to ensure efficient execution of tasks;

[0075] The queue management unit divides and manages tasks in queues, implements batch scheduling and flexible control of resources, and meets the resource requirements of different tasks;

[0076] Resource sharing unit, used for sharing resources between nodes. When a node is short of resources, it can borrow resources from other nodes.

[0077] Flexible scheduling unit dynamically adjusts task allocation and scheduling strategies according to task execution status and system resource changes to ensure system stability and smooth completion of tasks;

[0078] The task monitoring unit monitors the completion of tasks in real time, including task progress, execution time, and result output, so that tasks are executed as planned and achieve the expected results;

[0079] The task generation and screening unit generates new docking tasks and screening tasks according to system requirements and business logic, and screens and optimizes tasks through algorithms.

[0080] The above settings have the following functions:

[0081] When tasks in the queue are blocked due to insufficient resources, the resource sharing unit can automatically borrow resources from other nodes to ensure that the tasks in the queue can continue to execute. This interaction achieves dynamic balance and efficient utilization of resources.

[0082] The task allocation unit allocates tasks to nodes according to the current resource status, while the flexible scheduling unit dynamically adjusts the allocation strategy according to the task execution status and resource changes. This interaction enables the system to adaptively respond to various changes and maintain an efficient and stable state.

[0083] The task monitoring unit monitors the completion of tasks in real time, while the task generation and screening unit generates new tasks or screens existing tasks based on the monitoring results. This interaction enables the system to automatically adjust the number and priority of tasks to maintain consistency with business needs.

[0084] Furthermore, the task allocation unit is connected to the database signal and combines the calculation results in the database to optimize the task allocation process.

[0085] The above settings have the following functions:

[0086] Intelligent allocation: The task allocation unit can monitor system resources in real time, such as CPU, memory, storage, etc., and intelligently allocate tasks to currently idle or low-load nodes based on task priority. This allocation method ensures maximum utilization of system resources and reduces resource waste.

[0087] Database signal connection: The task allocation unit is connected to the database signal and can obtain the calculation results in the database in real time. These results not only provide important reference for task allocation, but also help optimize the entire allocation process.

[0088] Traditional resource allocation methods are often based on the estimated amount of computation of the task, which has certain deviations because the estimated amount of computation may not be accurate. However, by combining the speed at which the calculation results in the database increase, the actual needs of the task can be evaluated more accurately, thereby allocating resources more efficiently. In this system, when a task starts to execute, the task allocation unit monitors the speed at which its calculation results increase within a specific time period. Based on this speed, the amount of resources required for the task can be judged more accurately, and dynamic resource allocation can be performed. This dynamic allocation method can not only better meet the actual needs of the task, but also reduce performance degradation or resource waste caused by insufficient or excessive resources. Furthermore, the calculation results in the database can be retrieved to pre-screen out tasks that require repeated calculations.

[0089] Specifically, the task management module includes the following units:

[0090] Instruction receiving and distributing unit: responsible for receiving instructions from the system resource management module, and distributing the corresponding programs and structure files to the task computing module according to the instruction content.

[0091] Task status monitoring unit: monitors the execution status of tasks in real time, including task progress, resource usage, etc. Through real-time feedback, the system resource management module can timely understand the latest dynamics of tasks and provide decision-making basis for resource allocation and task scheduling;

[0092] Resource utilization reporting unit: collects resource utilization data of task computing modules, compiles these data into reports, and sends them to the system resource management module regularly;

[0093] The above settings have the following functions:

[0094] When the instruction receiving and distributing unit receives the instruction from the system resource management module, it works in conjunction with the task status monitoring unit. After the instruction receiving and distributing unit distributes the task to the task computing module, the task status monitoring unit immediately starts to monitor the execution status of the task. This real-time status monitoring can ensure that when an abnormal situation occurs in the task, the system can respond quickly to avoid resource waste and task delays.

[0095] The resource utilization reporting unit regularly sends resource utilization reports to the system resource management module, and the system resource management module optimizes resource allocation and task scheduling based on these reports. Through this interaction, the system can use computing resources more intelligently and improve overall processing efficiency and response speed.

[0096] When the instruction receiving and distributing unit is combined with the resource utilization reporting unit, they can jointly produce a new function: adaptive task scheduling. By analyzing the resource requirements of tasks and the resource status of the system in real time, the system can automatically adjust the priority and allocation strategy of tasks to ensure that critical tasks can get sufficient resource support first, thereby improving the overall performance of the system and user satisfaction.

[0097] In an optional embodiment, the task calculation module is refined to include the following units:

[0098] The knowledge graph fusion unit is responsible for integrating and fusing drug information from multiple sources. Through data cleaning and integration algorithms, scattered drug information such as efficacy, safety, and side effects are aggregated into a unified graph structure.

[0099] Drug data processing unit, which receives and processes raw drug data, including data cleaning, format conversion and standardization;

[0100] Feature extraction unit: Use feature engineering methods, such as text mining, statistical analysis, and domain expert knowledge, to identify and quantify key features, and extract key features such as efficacy, safety, side effects, etc. from processed drug data.

[0101] Drug combination recommendation unit: Based on the extracted features, combined with the drug relationships and feature information in the knowledge graph, using machine learning or deep learning algorithms, train the model to predict the effects of drug combinations and make recommendations on drug combinations.

[0102] Database storage and feedback unit: Store the processed results in the database and feed them back to the task management module. Use database management system (DBMS) technology to ensure the security and accessibility of data storage. At the same time, send the results to the task management module in real time or regularly through API or message queue.

[0103] The above settings have the following functions:

[0104] Enhanced data quality: The information provided by the knowledge graph fusion unit can help the drug data processing unit to clean and standardize data more accurately, thereby improving the quality of the data.

[0105] Personalized recommendation: The key features extracted by the feature extraction unit provide rich input information for the drug combination recommendation unit, enabling it to generate more personalized drug combination recommendations.

[0106] Real-time feedback and optimization: The database storage and feedback unit feeds back the recommendation results to the task management module, and collects feedback information to optimize the model of the drug combination recommendation unit, achieving real-time feedback and continuous improvement.

[0107] Preferably, the task status monitoring unit signal is connected to the knowledge graph fusion unit;

[0108] The knowledge graph fusion unit signal is connected to the drug combination recommendation unit;

[0109] The drug combination recommendation unit is signal-connected to the task status monitoring unit;

[0110] The three units (task status monitoring unit, knowledge graph fusion unit, and drug combination recommendation unit) are cycled and integrated multiple times to have the following functions:

[0111] 1. Real-time drug combination recommendation: Combining the task status monitoring unit and the knowledge graph fusion unit, the drug combination recommendation unit can learn and reason immediately when new drug data is added to the knowledge graph, and update the recommendation results in real time. This enables medical professionals to immediately obtain the latest drug combination recommendations to meet changing medical needs.

[0112] 2. Predictive resource planning: By analyzing historical task data, the task status monitoring unit can predict resource requirements in the future. Combining this prediction with the drug information from the knowledge graph fusion unit, the system can predict the development trend of drug research or treatment strategies, thereby performing resource planning and task scheduling in advance.

[0113] 3. Adaptive drug combination recommendation model: The drug combination recommendation unit can use the feedback provided by the task status monitoring unit to continuously adjust and optimize its recommendation model. For example, when a drug combination shows better results in actual applications, the model can automatically update its weights and parameters to improve the accuracy of future recommendations.

[0114] 4. Anomaly detection and fault prediction: The task status monitoring unit can monitor the abnormal conditions of tasks, such as long execution time, abnormal resource usage, etc. When an anomaly is detected, the system can trigger the fault prediction mechanism, use the drug information and historical data in the knowledge graph to predict the possible cause of the fault, and intervene and repair in advance.

[0115] 5. Intelligent task scheduling: Combined with the task status monitoring unit and the resource management module, the system can realize intelligent task scheduling. For example, when a drug combination recommendation task requires high-performance computing resources, the system can automatically schedule it to execute on a computing node with these resources to improve execution efficiency.

[0116] In another optional embodiment, the task calculation module includes:

[0117] Data preprocessing unit: collects original drug data from the hospital information system, including drug name, type, use, and side effects, and performs data cleaning, deduplication, and standardization;

[0118] Knowledge graph construction unit: Based on the preprocessed drug data, a drug knowledge graph is constructed. The nodes in the graph represent drugs, and the edges represent the relationships between drugs, such as similar effects, interactions, contraindications, etc. The knowledge graph can intuitively display the complex relationships between drugs.

[0119] Bipartite graph matrix construction unit: Based on the knowledge graph, the drugs and their relationships are mapped into a bipartite graph. The nodes on one side of the graph represent the drugs, and the nodes on the other side represent the relationship types. By constructing a bipartite graph matrix, the relationship strength between the drugs is quantified to provide data support for the subsequent recommendation algorithm;

[0120] Matrix feature mapping unit: Using machine learning techniques such as deep learning or graph embedding algorithms, the nodes and edges in the bipartite graph matrix are mapped into a low-dimensional vector space. These vectors represent the location of the drug in a specific feature space and help discover potential relationships between drugs.

[0121] Recommendation algorithm unit: Based on the results of matrix feature mapping, for a given drug combination, the algorithm calculates other drug combinations similar to the combination, and sorts and recommends them based on similarity. In addition, the algorithm can also consider factors such as drug side effects and contraindications to ensure the rationality of the recommendation.

[0122] Database storage and feedback unit: stores the processed results in the database and feeds them back to the task management module.

[0123] Furthermore, the task calculation module is connected to a database to retrieve data from the database during task calculation. This setting can improve the calculation speed.

[0124] Furthermore, the unexecuted task data of the system resource management module can be sent to a database, and the task calculation module can call the unexecuted task data.

[0125] When the task calculation module calculates task A, it needs to call the processed task data in the database (such as task B), but there is also a need to call unexecuted tasks (such as task C). All unexecuted tasks (including task C) are first stored in the database, and the call can be allowed. Since the result of task C has not been calculated, task C can be transferred to the task calculation module through the database, and task C can be sorted before task A. The task management module is notified to interrupt the calculation of task A and calculate task C first. When the calculation of task C is completed, it is stored in the database, and then the calculation of task A continues. At this time, task A can call tasks B and C, thereby improving the calculation speed.

[0126] It also has the following functions: when the calculation is completed, the calculation result of a task can be matched with the unexecuted task, and the unexecuted task can be automatically deleted after matching. This setting can prevent a task from being missed. And the execution time of a task can be calculated at the same time (the time difference between the unexecuted task entering the database and the unexecuted task being automatically deleted).

[0127] Furthermore, it also has the following functions:

[0128] Data persistence and security: Tasks are first sent to the database for storage to ensure data persistence. Even if the system fails or restarts, task data will not be lost. The database provides data integrity and security guarantees, such as backup, recovery, and permission control, to ensure the security of task data.

[0129] Task tracking and monitoring: The database serves as an intermediate storage point, which can easily track and monitor tasks. The system can record task status, execution time, results and other information to facilitate subsequent task management and analysis.

[0130] Error handling and recovery: If an error or exception occurs during the calculation of a task, the system can recover the task from the database and recalculate it. This method can improve the fault tolerance and stability of the system and ensure the reliability of the task.

[0131] Working principle of the present invention:

[0132] When in use, the system resource management module first monitors the resource usage of the node to ensure the reasonable allocation and efficient use of resources. When there is a new task, the system resource management module will divide the resources in a queue manner and generate corresponding docking tasks and screening tasks. After receiving the instruction, the task management module will distribute the program and structure files to the task calculation module for calculation. The task calculation module uses the fusion knowledge graph reasoning to construct the drug data relationship map, extract key features, and recommend drug combinations through machine learning or deep learning algorithms.

[0133] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0134] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An information data processing recommendation system, comprising a central processing module, characterized in that: The central processing module is signal-connected to a system resource management module, a task management module and a task calculation module; System resource management module, responsible for monitoring node resource usage, allocating and managing tasks, dividing resources into queues, monitoring task completion, and generating new docking tasks and screening tasks; The task management module receives the instruction distribution program and structure file from the system resource management module to the task calculation module, and reports the task completion status and resource utilization status to the system resource management module; The task calculation module uses fused knowledge graph reasoning to construct a relationship graph between drug data, extracts key features from the processed drug data, and based on the extracted features, uses machine learning or deep learning algorithms to recommend drug combinations. The processed results are stored in the database and fed back to the task management module, and the database data is displayed on the display.

2. The information data processing recommendation system according to claim 1, characterized in that: The system resource management module includes the following units: Resource monitoring unit, responsible for real-time monitoring of resource usage of each node in the system, including CPU, memory, disk space, and network bandwidth; The task allocation unit allocates tasks to appropriate nodes according to the current system resource status and task priority to ensure efficient execution of tasks; The queue management unit divides and manages tasks in queues, implements batch scheduling and flexible control of resources, and meets the resource requirements of different tasks; Resource sharing unit, used for sharing resources between nodes. When a node is short of resources, it can borrow resources from other nodes. Flexible scheduling unit dynamically adjusts task allocation and scheduling strategies according to task execution status and system resource changes to ensure system stability and smooth completion of tasks; The task monitoring unit monitors the completion of tasks in real time, including task progress, execution time, and result output, so that tasks are executed as planned and achieve the expected results; The task generation and screening unit generates new docking tasks and screening tasks according to system requirements and business logic, and screens and optimizes tasks through algorithms.

3. An information data processing recommendation system as claimed in claim 2, characterized in that: The task allocation unit is connected to the database signal and optimizes the task allocation process in combination with the calculation results in the database.

4. The information data processing recommendation system according to claim 3, characterized in that: The task management module includes the following units: Instruction receiving and distributing unit: responsible for receiving instructions from the system resource management module, and distributing the corresponding programs and structure files to the task computing module according to the instruction content; Task status monitoring unit: monitors the execution status of tasks in real time, including task progress and resource usage. Through real-time feedback, the system resource management module can keep abreast of the latest developments of tasks. Resource utilization reporting unit: collects resource utilization data of the task computing module, compiles the data into reports, and sends them to the system resource management module regularly.

5. The information data processing recommendation system according to claim 1, characterized in that: The task calculation module is connected to a database and retrieves data from the database during task calculation.

6. The information data processing recommendation system according to claim 4, characterized in that: The task calculation module is refined to include the following units: The knowledge graph fusion unit is responsible for integrating and fusing drug information from multiple sources. Through data cleaning and integration algorithms, the scattered drug information is aggregated into a unified graph structure. Drug data processing unit, which receives and processes raw drug data, including data cleaning, format conversion and standardization; Feature extraction unit: identifies and quantifies key features and extracts key features from processed drug data; Drug combination recommendation unit: Based on the extracted features, combined with the drug relationships and feature information in the knowledge graph, using machine learning or deep learning algorithms, trains the model to predict the effect of drug combinations and implements drug combination recommendations; The first database storage and feedback unit: stores the processed results in the database and feeds them back to the task management module.

7. An information data processing recommendation system as claimed in claim 6, characterized in that: The task status monitoring unit signal is connected to the knowledge graph fusion unit; The knowledge graph fusion unit signal is connected to the drug combination recommendation unit; The drug combination recommendation unit signal is connected to the task status monitoring unit.

8. The information data processing recommendation system according to claim 1, characterized in that: The unexecuted task data of the system resource management module can be sent to a database, and the task calculation module can call the unexecuted task data.

9. The information data processing recommendation system according to claim 1, characterized in that: The task calculation module includes: Data preprocessing unit: collects raw drug data from hospital information systems and performs data cleaning, deduplication, and standardization; Knowledge graph construction unit: Based on the preprocessed drug data, a drug knowledge graph is constructed. The nodes in the graph represent drugs, and the edges represent the relationship between drugs. Bipartite graph matrix construction unit: Based on the knowledge graph, the drugs and their relationships are mapped into a bipartite graph. The nodes on one side of the graph represent the drugs, and the nodes on the other side represent the relationship types. By constructing a bipartite graph matrix, the relationship strength between the drugs is quantified to provide data support for the subsequent recommendation algorithm; Matrix feature mapping unit: uses machine learning technology to map the nodes and edges in the bipartite graph matrix into a low-dimensional vector space; Recommendation algorithm unit: Based on the results of matrix feature mapping, for a given drug combination, the algorithm calculates other drug combinations similar to the combination, and sorts and recommends them based on the similarity; The second data storage and feedback unit stores the processed results in the database and feeds them back to the task management module.

Images