Decision support-oriented intelligent analysis system in station network in data

Abstract

The invention provides a decision support-oriented intelligent analysis system in a station network in data, and relates to the field of electric digital data processing, the decision support-oriented intelligent analysis system comprises a data storage management module, an application decision interaction module, a data analysis retrieval module and an intelligent analysis support module, the data storage management module is used for storing data information of all service systems, and the application decision interaction module is used for interacting with the data analysis retrieval module; the application decision interaction module is used for logging in an account and carrying out interaction operation, the data analysis retrieval module is used for analyzing application content and retrieving to obtain corresponding service data, and the intelligent analysis support module is used for analyzing and processing the service data to obtain decision support information; the system can analyze the decision in combination with the data information of a plurality of service parts, and can assist in making a more reasonable decision.

Description

Technical Field

[0001] The present invention relates to the field of electrical digital data processing, and more particularly to an intelligent analysis system for decision support in a data middle platform network. Background Art

[0002] In the context of the rapid development of current big data and artificial intelligence, the demand for data-driven decision support in various industries is increasing day by day. Traditional decision support systems mainly rely on static reports and manual experience analysis, and there are problems such as complex data storage and management, low data retrieval efficiency, and insufficient intelligent analysis capabilities. Therefore, a system is needed to provide auxiliary analysis for decision-making content and improve the rationality of decision-making.

[0003] The foregoing discussion of the background art is only intended to facilitate the understanding of the present invention. This discussion does not recognize or admit that any of the materials mentioned is part of common general knowledge.

[0004] Now, many decision analysis systems have been developed. After a large amount of retrieval and reference, it is found that existing decision analysis systems such as the system disclosed in the publication number CN118365248B generally include a data acquisition module, a weight assignment module, a current purchase quantity determination module, a supply quantity determination module, and a purchase order generation module; the data acquisition module divides historical time into multiple historical action time periods and acquires historical data; the weight assignment module calculates the correlation degree between the current action time period and the historical action time periods and assigns weights to multiple historical action time periods; the current purchase quantity determination module inputs the historical data and weights of multiple historical action time periods into a purchase quantity determination model, and the model outputs the current purchase quantity; the supply quantity determination module uses a genetic algorithm to allocate the current purchase quantity to multiple suppliers for supply to obtain a target supply plan; however, this system only conducts decision analysis based on the data of a single department, without integrating the data information of relevant parts, and the decision analysis information obtained is incomplete. Summary of the Invention

[0005] The object of the present invention is to propose an intelligent analysis system for decision support in a data middle platform network in view of the existing deficiencies.

[0006] The present invention adopts the following technical solutions:

[0007] An intelligent analysis system for decision support in a data middle platform network, comprising a data storage and management module, an application decision interaction module, a data parsing and retrieval module, and an intelligent analysis support module;

[0008] The data storage and management module is used to store the data information of all business systems. The application decision-making interaction module is used to log in to an account and perform interaction operations. The data parsing and retrieval module parses the application content and retrieves the corresponding business data. The intelligent analysis and support module is used to analyze and process the business data to obtain decision support information;

[0009] The data storage and management module includes a data information storage unit, a data index retrieval unit, and a data permission management unit. The data information storage unit is used to store specific data content. The data index retrieval unit is used to build an index framework to accelerate the retrieval speed. The data permission management unit is used to verify and manage the access permissions of the data;

[0010] The application decision-making interaction module includes an account information management unit, an interaction interface processing unit, and a business process logic unit. The account information management unit is used to manage the information of registered users. The interaction interface processing unit is used to provide an interaction interface for users to perform operation processing. The business process logic unit is used to perform feedback analysis on the operation content;

[0011] The data parsing and retrieval module includes a business parsing unit, a type mapping unit, and a retrieval task output unit. The business parsing unit is used to analyze the requirements of the user's business information. The type mapping unit is used to map the parsing results to the corresponding data tags. The retrieval task output unit is used to output the retrieval task;

[0012] The intelligent analysis and support module includes a statistical analysis unit, a predictive analysis unit, and a decision optimization unit. The statistical analysis unit is used to perform statistical processing on the business data to obtain basic information. The predictive analysis unit is used to predict the trend of business decisions. The decision optimization unit is used to perform optimization analysis on the decision content based on the prediction results.

[0013] Furthermore, the statistical analysis unit includes an anomaly detection processor, an association rule miner, and a statistical counting processor. The anomaly detection processor is used to identify outlier data points and mark them as abnormal data. The association rule miner is used to mine the implicit relationships between data items. The statistical counting processor is used to perform statistical processing on the retrieved data;

[0014] The anomaly detection processor calculates the sensitivity index D of the data point according to the following formula:

[0015]

[0016] where X i represents the i-th data point, X is this data point, μ is the data point mean, σ is the data point standard deviation, n is the number of data points, and λ is the outlier coefficient;

[0017] When the sensitivity index is greater than the threshold, the corresponding data points are marked as abnormal data and will not be subject to statistical processing.

[0018] Furthermore, the prediction analysis unit includes a feature calculation processor, an uncertainty quantizer, and a prediction output processor. The feature calculation processor extracts feature data based on statistical data. The uncertainty quantizer is used to simulate and quantify the confidence radius. The prediction output processor is used to output the prediction result.

[0019] The uncertainty quantizer calculates the confidence radius R according to the following formula:

[0020]

[0021] where P i is the median of the i-th prediction, E(P) is the mean prediction, m st is the number of stable predictions, and m is the total number of predictions.

[0022] Furthermore, the prediction output processor processes the feature data to obtain the median prediction P, and outputs the prediction region [P - R, P + R] in combination with the confidence radius. The prediction region needs to be within the range of [0, 100]. If it exceeds, the exceeded part is truncated.

[0023] The median prediction is calculated according to the following formula:

[0024]

[0025] where V i represents the value of the i-th feature item, k i is the conversion coefficient of the i-th feature item, and T is the number of feature items.

[0026] Furthermore, the optimization decision unit includes an optimization judgment processor, a weight assignment processor, and a decision adjustment outputter. The optimization judgment processor is used to judge whether to optimize the decision. The weight assignment processor is used to assign weights to decision items. The decision adjustment outputter is used to output the adjustment information of decision items.

[0027] The optimization judgment processor calculates the optimization judgment value Q according to the following formula:

[0028]

[0029] where α is the actual conversion coefficient, β is the optimization ratio coefficient, r is the base radius, W(i) is the actual effect value of the i-th decision, and N is the number of decisions.

[0030] When Q is less than 0, optimization is required.

[0031] The beneficial effects achieved by the present invention are as follows:

[0032] This system supports scientific decision-making through intelligent analysis, enhances business insight, realizes automatic identification of abnormal data based on an anomaly detection processor combined with a dynamic anomaly sensitivity factor, improves data quality, achieves accurate prediction through an uncertainty quantizer, provides a confidence interval, improves the credibility of prediction, enhances business value through decision optimization, strengthens the intelligence of the system, automatically adjusts the decision-making plan, and improves the efficiency of resource allocation.

[0033] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the provided drawings are only for reference and illustration, and are not used to limit the present invention. Brief Description of the Drawings

[0034] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0035] Figure 2 It is a schematic diagram of the composition of the data storage and management module of the present invention;

[0036] Figure 3 It is a schematic diagram of the composition of the application decision interaction module of the present invention;

[0037] Figure 4 It is a schematic diagram of the composition of the data parsing and retrieval module of the present invention;

[0038] Figure 5 It is a schematic diagram of the composition of the intelligent analysis support module of the present invention;

[0039] Figure 6 It is a comparison table of the data effects between the present invention and a general system. Detailed Embodiments

[0040] The following are specific embodiments to illustrate the implementation manners of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Additionally, the drawings of the present invention are only simple schematic illustrations and are not drawn according to actual sizes, which is hereby stated in advance. The following embodiments will further detail the related technical content of the present invention, but the disclosed content is not used to limit the protection scope of the present invention.

[0041] Example 1.

[0042] This example provides an intelligent analysis system for decision support in a data middle platform network, combined with Figure 1, including a data storage management module, an application decision-making interaction module, a data parsing and retrieval module, and an intelligent analysis support module;

[0043] The data storage management module is used to store the data information of all business systems. The application decision-making interaction module is used to log in to an account and perform interaction operations. The data parsing and retrieval module parses the application content and retrieves the corresponding business data. The intelligent analysis support module is used to analyze and process the business data to obtain decision support information;

[0044] The data storage management module includes a data information storage unit, a data index retrieval unit, and a data permission management unit. The data information storage unit is used to store specific data content. The data index retrieval unit is used to build an index framework to accelerate the retrieval speed. The data permission management unit is used to verify and manage the access permissions of the data;

[0045] The application decision-making interaction module includes an account information management unit, an interaction interface processing unit, and a business process logic unit. The account information management unit is used to manage the information of registered users. The interaction interface processing unit is used to provide an interaction interface for users to perform operation processing. The business process logic unit is used to perform feedback analysis on the operation content;

[0046] The data parsing and retrieval module includes a business parsing unit, a type mapping unit, and a retrieval task output unit. The business parsing unit is used to analyze the requirements of the user's business information. The type mapping unit is used to map the parsing result to the corresponding data label. The retrieval task output unit is used to output the retrieval task;

[0047] The intelligent analysis support module includes a statistical analysis unit, a predictive analysis unit, and a decision optimization unit. The statistical analysis unit is used to perform statistical processing on the business data to obtain basic information. The predictive analysis unit is used to predict the trend of business decisions. The decision optimization unit is used to optimize and analyze the decision content based on the prediction result.

[0048] The statistical analysis unit includes an anomaly detection processor, an association rule miner, and a statistical counting processor. The anomaly detection processor is used to identify outlier data points and mark them as abnormal data. The association rule miner is used to mine the implicit relationships between data items. The statistical counting processor is used to perform statistical processing on the retrieved data;

[0049] The anomaly detection processor calculates the sensitivity index D of the data point according to the following formula:

[0050]

[0051] where X iDenote the i-th data point as \(X\), where \(\mu\) is the mean of the data points, \(\sigma\) is the standard deviation of the data points, \(n\) is the number of data points, and \(\lambda\) is the outlier coefficient;

[0052] When the sensitivity index is greater than the threshold, the corresponding data points are marked as abnormal data and will not be statistically processed.

[0053] The prediction analysis unit includes a feature calculation processor, an uncertainty quantizer, and a prediction output processor. The feature calculation processor extracts feature data based on statistical data. The uncertainty quantizer is used to simulate and quantify the confidence radius. The prediction output processor is used to output the prediction result;

[0054] The uncertainty quantizer calculates the confidence radius \(R\) according to the following formula:

[0055]

[0056] where \(P\) i is the median of the i-th prediction, \(E(P)\) is the mean of the predictions, \(m\) st is the number of stable predictions, and \(m\) is the total number of predictions.

[0057] The prediction output processor processes the feature data to obtain the prediction median \(P\), and outputs the prediction region \([P - R, P + R]\) in combination with the confidence radius. The prediction region needs to be within the range of \([0, 100]\). If it exceeds, the exceeding part is truncated;

[0058] The prediction median is calculated according to the following formula:

[0059]

[0060] where \(V\) i represents the value of the i-th feature item, \(k\) i is the conversion coefficient of the i-th feature item, and \(T\) is the number of feature items.

[0061] The optimization decision unit includes an optimization judgment processor, a weight assignment processor, and a decision adjustment outputter. The optimization judgment processor is used to judge whether to optimize the decision. The weight assignment processor is used to assign weights to the decision items. The decision adjustment outputter is used to output the adjustment information of the decision items;

[0062] The optimization judgment processor calculates the optimization judgment value \(Q\) according to the following formula:

[0063]

[0064] where \(\alpha\) is the actual conversion coefficient, \(\beta\) is the optimization ratio coefficient, \(r\) is the base radius, \(W(i)\) is the actual effect value of the i-th decision, and \(N\) is the number of decisions;

[0065] When Q is less than 0, optimization is required.

[0066] Embodiment 2.

[0067] This embodiment includes all the contents of Embodiment 1 and provides an intelligent analysis system for decision support in a data middle platform network, including a data storage and management module, an application decision interaction module, a data parsing and retrieval module, and an intelligent analysis support module;

[0068] The data storage and management module is used to store the data information of all business systems, the application decision interaction module is used to log in to an account and perform interaction operations, the data parsing and retrieval module parses the application content and retrieves the corresponding business data, and the intelligent analysis support module is used to analyze and process the business data to obtain decision support information;

[0069] Combined with Figure 2 , the data storage and management module includes a data information storage unit, a data index retrieval unit, and a data permission management unit. The data information storage unit is used to store specific data content, the data index retrieval unit is used to build an index framework to accelerate the retrieval speed, and the data permission management unit is used to verify and manage the access permissions of the data;

[0070] Combined with Figure 3 , the application decision interaction module includes an account information management unit, an interaction interface processing unit, and a business process logic unit. The account information management unit is used to manage the information of registered users, the interaction interface processing unit is used to provide an interaction interface for users to perform operation processing, and the business process logic unit is used to perform feedback analysis on the operation content;

[0071] Combined with Figure 4 , the data parsing and retrieval module includes a business parsing unit, a type mapping unit, and a retrieval task output unit. The business parsing unit is used to parse the requirements of the user's business information, the type mapping unit is used to map the parsing result to the corresponding data label, and the retrieval task output unit is used to output the retrieval task;

[0072] Combined with Figure 5 , the intelligent analysis support module includes a statistical analysis unit, a prediction analysis unit, and a decision optimization unit. The statistical analysis unit is used to perform statistical processing on the business data to obtain basic information, the prediction analysis unit is used to predict the trend of business decisions, and the decision optimization unit is used to perform optimization analysis on the decision content based on the prediction result;

[0073] The data information storage unit includes a distributed storage cluster, a hot and cold data partitioner, and a data compression encoder. The distributed storage cluster is used to implement sharded storage and disaster recovery backup of massive data. The hot and cold data partitioner is used to identify frequently accessed data and allocate it to the high-speed storage area. The data compression encoder is used to compress and store unstructured data;

[0074] The data index retrieval unit includes an inverted index builder, a vectorized cache processor, and a load balancing scheduler. The inverted index builder is used to establish the mapping relationship between keywords and documents. The vectorized cache processor is used to convert hot indexes into memory vectors to accelerate retrieval. The load balancing scheduler is used to dynamically allocate query requests;

[0075] The data permission management unit includes a user permission registrar, a permission verification processor, and an access track tracer. The user permission registrar is used to register the permission information of all users. The permission verification processor is used to verify the user identity and grant corresponding permissions. The access track tracer is used to record the behavior logs of data access;

[0076] The account information management unit includes a user registration processor, a detailed information editor, and a user information register. The user registration processor is used to register new user information. The detailed information editor is used to edit the specific information of users. The user information register is used to store the personal information of users;

[0077] The interaction interface processing unit includes a dynamic UI component library, a multimodal rendering engine, and a response adapter. The dynamic UI component library is used to provide interaction tools. The multimodal rendering engine is used to generate an operation interface with mixed multi-components. The response adapter is used to automatically adapt to the display end resolution and operation mode;

[0078] The business process logic unit includes an interaction collection processor, a process information library, and a logic feedback processor. The interaction collection processor is used to collect the input information of the interaction interface. The process information library is used to store business process information. The logic feedback processor outputs feedback information to the interaction interface processing unit based on the collected information and process information;

[0079] The business analysis unit includes a natural language processor, a demand classifier, and a context analyzer. The natural language processor is used to analyze the semantic requirements input by users. The demand classifier is used to identify the query intent and classify the demands. The context analyzer is used to establish a session memory library to achieve multi-round interaction context understanding;

[0080] The type mapping unit includes a requirement receiving processor, a mapping information register, and a requirement mapping processor. The requirement receiving processor is used to receive the identified requirement type. The mapping information register is used to store the mapping relationship between the requirement type and the data label. The requirement mapping processor is used to map the requirement type to the data label and output it to the retrieval task output unit;

[0081] The retrieval task output unit includes a data compounding processor, a compounding verification processor, and a task generation processor. The compounding processor is used to compound different data labels. The compounding verification processor is used to verify the compounded data label set. The task generation processor generates a corresponding search task based on the verified data label set;

[0082] The statistical analysis unit includes an anomaly detection processor, an association rule miner, and a statistical counting processor. The anomaly detection processor is used to identify outlier data points and mark them as abnormal data. The association rule miner is used to mine the implicit relationships between data items. The statistical counting processor is used to perform statistical processing on the retrieved data;

[0083] The anomaly detection processor calculates the sensitivity index D of the data point according to the following formula:

[0084]

[0085] where, X i represents the i-th data point, X is this data point, μ is the data point mean, σ is the data point standard deviation, n is the number of data points, and λ is the outlier coefficient;

[0086] When the sensitivity index is greater than the threshold, the corresponding data point is marked as abnormal data and will not be statistically processed;

[0087] The prediction analysis unit includes a feature calculation processor, an uncertainty quantizer, and a prediction output processor. The feature calculation processor extracts feature data based on the statistical data. The uncertainty quantizer is used to simulate and quantify the confidence radius. The prediction output processor is used to output the prediction result;

[0088] The uncertainty quantizer calculates the confidence radius R according to the following formula:

[0089]

[0090] where, P i is the prediction median of the i-th time, E(P) is the prediction mean, m st is the number of stable predictions, and m is the total number of predictions;

[0091] The uncertainty quantifier records the information of each prediction and statistically analyzes the prediction effects according to different prediction types. When the actual decision-making effect is within the prediction confidence interval, the corresponding prediction is called a stable prediction;

[0092] The prediction output processor processes the feature data to obtain the prediction median P, and outputs the prediction region [P - R, P + R] in combination with the confidence radius. The prediction region needs to be within the range of [0, 100]. If it exceeds, the exceeded part is truncated;

[0093] The prediction median is calculated according to the following formula:

[0094]

[0095] where V i represents the value of the i-th feature item, k i is the conversion coefficient of the i-th feature item, and T is the number of feature items;

[0096] The optimization decision unit includes an optimization judgment processor, a weight allocation processor, and a decision adjustment outputter. The optimization judgment processor is used to judge whether to optimize the decision, the weight allocation processor is used to allocate weights to decision items, and the decision adjustment outputter is used to output the adjustment information of decision items;

[0097] The optimization judgment processor calculates the optimization judgment value Q according to the following formula:

[0098]

[0099] where α is the actual conversion coefficient, β is the optimization ratio coefficient, r is the base radius, W(i) is the actual effect value of the i-th decision, and N is the number of decisions;

[0100] When Q is less than 0, optimization is required;

[0101] Both i and j appearing in the above text are ordinals used to represent serial numbers and have no actual meaning.

[0102] Some code information of this system is as follows:

[0103]

[0104]

[0105] The content disclosed above is only the preferred feasible embodiment of the present invention, and does not limit the protection scope of the present invention. Therefore, all equivalent technical changes made by using the content of the specification and drawings of the present invention are included in the protection scope of the present invention. In addition, the elements therein can be updated with the development of technology.

Claims

1. An intelligent analysis system for decision support in a data middle platform network, characterized in that, It includes a data storage management module, an application decision-making interaction module, a data parsing and retrieval module, and an intelligent analysis support module; The data storage management module is used to store the data information of all business systems. The application decision-making interaction module is used to log in to an account and perform interaction operations. The data parsing and retrieval module parses the application content and retrieves the corresponding business data. The intelligent analysis support module is used to analyze and process the business data to obtain decision support information; The data storage management module includes a data information storage unit, a data index retrieval unit, and a data permission management unit. The data information storage unit is used to store specific data content. The data index retrieval unit is used to build an index framework to accelerate the retrieval speed. The data permission management unit is used to verify and manage the access permissions of the data; The application decision-making interaction module includes an account information management unit, an interaction interface processing unit, and a business process logic unit. The account information management unit is used to manage the information of registered users. The interaction interface processing unit is used to provide an interaction interface for users to perform operation processing. The business process logic unit is used to perform feedback analysis on the operation content; The data parsing and retrieval module includes a business parsing unit, a type mapping unit, and a retrieval task output unit. The business parsing unit is used to analyze the requirements of the user's business information. The type mapping unit is used to map the parsing result to the corresponding data label. The retrieval task output unit is used to output the retrieval task; The intelligent analysis support module includes a statistical analysis unit, a prediction analysis unit, and a decision optimization unit. The statistical analysis unit is used to perform statistical processing on the business data to obtain basic information. The prediction analysis unit is used to predict the trend of business decisions. The decision optimization unit is used to perform optimization analysis on the decision content based on the prediction result.

2. The intelligent analysis system for decision support in the data middle platform network according to claim 1, characterized in that, The statistical analysis unit includes an anomaly detection processor, an association rule miner, and a statistical counting processor. The anomaly detection processor is used to identify outlier data points and mark them as abnormal data. The association rule miner is used to mine the implicit relationships between data items. The statistical counting processor is used to perform statistical processing on the retrieved data; The anomaly detection processor calculates the sensitivity index D of the data point according to the following formula: Among them, X i represents the i-th data point, X is this data point, μ is the mean of the data points, σ is the standard deviation of the data points, n is the number of data points, and λ is the outlier coefficient; When the sensitivity index is greater than the threshold, the corresponding data point is marked as abnormal data and will not be statistically processed.

3. The intelligent analysis system for decision support in the data middle platform network according to claim 2, characterized in that, The prediction analysis unit includes a feature calculation processor, an uncertainty quantizer, and a prediction output processor. The feature calculation processor extracts feature data based on the statistical data. The uncertainty quantizer is used to simulate and quantify the confidence radius. The prediction output processor is used to output the prediction result; The uncertainty quantizer calculates the confidence radius R according to the following formula: Among them, P i is the predicted median value of the i-th time, E(P) is the predicted mean value, m st is the number of stable prediction times, and m is the total number of prediction times.

4. The intelligent analysis system for decision support in the data middle platform network according to claim 3, characterized in that, The prediction output processor processes the feature data to obtain the prediction median P, and outputs the prediction region [P - R, P + R] in combination with the confidence radius. The prediction region needs to be within the range of [0, 100]. If it exceeds, the exceeded part is truncated; The prediction median is calculated according to the following formula: Among them, V i represents the value of the i-th feature item, and k i is the conversion coefficient of the i-th feature item, and T is the number of feature items.

5. The intelligent analysis system for decision support in the data middle platform network according to claim 4, characterized in that, The optimization decision-making unit includes an optimization judgment processor, a weight allocation processor, and a decision adjustment outputter. The optimization judgment processor is used to judge whether to optimize the decision. The weight allocation processor is used to allocate weights to decision items. The decision adjustment outputter is used to output the adjustment information of decision items; The optimization judgment processor calculates an optimization judgment value Q according to the following formula: where α is the actual conversion coefficient, β is the optimization ratio coefficient, r is the base radius, W(i) is the actual effect value of the i-th decision, and N is the number of decisions; When Q is less than 0, optimization is required.

Images