A method and system for identifying smart government information based on big data

Through principal component analysis and multi-layer classification technology, the problem that traditional methods are difficult to identify complex government information is solved, high-precision classification and management of government information is achieved, and classification efficiency and accuracy are improved.

CN118885926BActive Publication Date: 2024-11-29SHANDONG ZHENGTU INFORMATION POLYTRON TECH INC
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411345753.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-26
Publication Date
2024-11-29
Estimated Expiration
2044-09-26

AI Technical Summary

Technical Problem

It is difficult for traditional classification methods to accurately identify and manage complex government information, especially government information on the same topic is divided due to large distance differences, and government information on different topics is divided together due to small correlation.

Method used

Using a smart government information identification method based on big data, through principal component analysis and multi-layer classification, using the degree of category segmentation and principal component direction, select appropriate classification characteristics for government information classification until the degree of category segmentation is less than the preset threshold. Form a classification model that is completed with the fitting, and identify and process the newly collected government information.

Benefits of technology

It improves the classification accuracy and accuracy of government information, realizes more detailed classification management, simplifies classification operations, and improves classification efficiency and accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118885926B_ABST
    Figure CN118885926B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing, and in particular to a method and system for intelligent government information identification based on big data. The method includes the following steps: obtaining a historical government information sequence and a number of corresponding classification features; using a classification feature to divide all historical government information sequences into two categories; calculating the degree of category segmentation; using the classification feature with the largest degree of category segmentation to divide all government information data sequences into two categories and record them as target categories; obtaining the classification feature with the largest degree of category segmentation of the target category, and using it to complete the classification operation of all government information data sequences of the target category; and so on, until the degree of category segmentation is less than a preset threshold, the classification operation is stopped to obtain a fitted classification model; and the newly collected government information sequence is classified and identified using the fitted classification model. The present invention achieves accurate classification of government information by accurately measuring the classification effect of government information.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method and system for identifying smart government information based on big data. Background Art

[0002] Government information plays an important role in modern society. Its effective management and release is an important means to improve government governance, promote social development and enhance public welfare. Government information is complex and diverse. In order to facilitate the management of government information, it is necessary to identify government information of the same type and store it together for future query and retrieval.

[0003] In order to facilitate the classification and management of government information, it is necessary to separate government information of the same subject together and separate government information of different subjects. Traditional classification methods generally classify data based on the distance relationship between data, separating data with close distances together and separating data with far distances apart; however, sometimes the government information data of the same subject are highly correlated, but their distance difference is also large, at this time it is easy to separate the government information of the same subject through the distance relationship; at the same time, sometimes the government information data of different subjects are less correlated, but their distance difference is also small, at this time it is easy to separate government information of different subjects together through the distance relationship; therefore, it is not possible to classify and manage government information well using traditional classification methods.

[0004] The patent application document with publication number CN116644328A discloses a kmeans log classification method based on adaptive radius center selection. The method in the application document classifies logs based on distance relationships. This classification method is not suitable for government information classification management, and therefore this method cannot be used for government information management. Summary of the invention

[0005] In order to solve the problem of how to accurately classify government information, the present invention provides a smart government information identification method and system based on big data.

[0006] In the first aspect, the present invention provides a method for identifying smart government information based on big data, which adopts the following technical solutions:

[0007] A method for identifying smart government information based on big data, comprising the steps of:

[0008] Use a classification feature to divide all historical government information sequences into two categories; calculate the degree of category segmentation: , perform principal component analysis on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component direction corresponding to each eigenvalue, , Respectively represent one of the categories eigenvalues, the largest eigenvalue, and the The characteristic values , , Respectively represent another category eigenvalues, the largest eigenvalue, and the eigenvalues, Indicates that one of the categories The principal component direction corresponding to the eigenvalue of another category The angle between the principal component directions corresponding to the eigenvalues, M represents the number of eigenvalues ​​for each category, Indicates the preset anti-zero coefficient. represents the hyperbolic tangent function; using the classification feature with the largest category segmentation degree to divide all government information data sequences into two categories and record them as target categories; obtaining the classification feature with the largest category segmentation degree of the target category, and using it to complete the classification operation of all government information data sequences of the target category; and so on, until the category segmentation degree is less than the preset threshold, the classification operation is stopped to obtain the fitted classification model;

[0009] The fitted classification model is used to classify and identify the newly collected government information sequence.

[0010] The present invention classifies and analyzes all historical government information sequences in more detail through multi-layer classification, thereby improving the classification accuracy; further, in each layer of classification, the historical government information sequences are reasonably classified by selecting appropriate classification features, thereby improving the classification accuracy of the historical government information sequences; further, when selecting classification feature values, the classification results corresponding to each classification feature are measured, so that the selected classification features can more accurately classify the historical government information sequences; further, when measuring the classification results corresponding to the classification features, the correlation of the same category and the orthogonality of different categories are accurately measured by introducing information such as the principal component direction and the eigenvalue, thereby more accurately reflecting the classification effect of each classification feature, and providing a basis for accurate classification feature selection.

[0011] Preferably, the method uses a classification feature to classify all historical government information sequences into two categories, including:

[0012] The preset classification boundary of each classification feature is obtained, and all historical government information sequences are divided into two categories using the classification boundary.

[0013] The present invention performs classification by presetting classification boundaries, and its classification operation is relatively simple, so the classification efficiency is relatively high.

[0014] Preferably, the step of obtaining the classification feature with the largest category segmentation degree of the target category and using it to complete the classification operation of all government information data sequences of the target category includes:

[0015] All historical government information sequences of each target category are divided into two categories using one of all classification features, the category segmentation degree is calculated, and the classification feature with the largest category segmentation degree is selected from all classification features to divide all historical government information sequences of the target category into two categories.

[0016] The present invention performs classification analysis again on the classification results obtained by the previous classification, thereby realizing hierarchical processing of the classification operation, improving the meticulousness of the classification operation, and further improving the accuracy of the classification operation.

[0017] Preferably, the step of stopping the classification operation until the category segmentation degree is less than a preset threshold to obtain a fitted classification model includes:

[0018] Any bottom-level category is recorded as the category to be analyzed, and a classification feature is selected from all classification features to divide the category to be analyzed into two categories. The category segmentation degree is calculated, and the category segmentation degree of each classification feature is obtained. If the category segmentation degree of all classification features is less than the preset threshold, the classification operation of the category to be analyzed is stopped; the classification cutoff judgment is performed on other bottom-level categories until each bottom-level category meets the classification cutoff judgment condition, and a fitted classification model is obtained.

[0019] The present invention uses a threshold value to stop the classification operation, which is simpler and more convenient to implement and improves the efficiency of the classification operation.

[0020] Preferably, the method of using the fitted classification model to classify and identify the newly collected government information sequence includes:

[0021] Obtain the subject information of each bottom-level category in the fitted classification model; input the newly collected government information sequence into the fitted classification model to obtain the category to which the newly collected government information sequence belongs, and use the subject information of the category as the subject information of the newly collected government information sequence.

[0022] The present invention can obtain the corresponding classification results by inputting the newly collected government information sequence into the fitted classification model. This classification method is simpler to operate and more efficient to implement.

[0023] Preferably, the principal component analysis of all historical government information sequences of each category to obtain a number of eigenvalues ​​and the principal component direction corresponding to each eigenvalue includes:

[0024] The principal component analysis algorithm is used to perform principal component analysis on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component direction corresponding to each eigenvalue.

[0025] Preferably, the acquisition of historical government affairs information sequences and corresponding classification features includes:

[0026] Acquire some historical government information, convert non-data information in the historical government information into data information to obtain converted historical government information, and record a sequence of the converted historical government information as a historical government information sequence;

[0027] Obtain several classification features of historical government information.

[0028] In the second aspect, the present invention provides a smart government information identification system based on big data, which adopts the following technical solutions:

[0029] A smart government information identification system based on big data includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned smart government information identification method based on big data is implemented.

[0030] By adopting the above technical solution, the above-mentioned method for intelligent government information identification based on big data is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is made according to the memory and the processor for easy use.

[0031] The present invention has the following technical effects:

[0032] The present invention classifies and analyzes all historical government information sequences in a more detailed manner through multi-layer classification, thereby improving the classification accuracy;

[0033] Furthermore, when classifying at each level, the historical government information sequence is reasonably classified by selecting appropriate classification features, thereby improving the classification accuracy of the historical government information sequence;

[0034] Furthermore, when selecting classification feature values, by measuring the classification results corresponding to each classification feature, the selected classification feature can more accurately classify the historical government information sequence;

[0035] Furthermore, when measuring the classification results corresponding to the classification features, by introducing information such as the principal component direction and eigenvalue, the correlation between the same category and the orthogonality of different categories are accurately measured, thereby more accurately reflecting the classification effect of each classification feature and providing a basis for accurate classification feature selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0037] Figure 1 It is a flow chart of a method in a method for identifying smart government information based on big data in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0039] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0040] The embodiment of the present invention discloses a method for identifying smart government information based on big data, referring to Figure 1 , comprising steps S1 to S3:

[0041] S1: Obtain the historical government information sequence and the corresponding classification features.

[0042] Specifically, the government information collected in the past is obtained and recorded as historical government information; the text in the historical government information is converted into data using the Word2Vec method to obtain the converted historical government information, and the sequence formed by the converted historical government information is recorded as the historical government information sequence;

[0043] Obtain several classification characteristics of historical government information. The classification characteristics of historical government information include but are not limited to the following aspects: policy, administration, finance, social services, security and emergency information, and government reports.

[0044] S2: Use a classification feature to divide all historical government information sequences into two categories; calculate the degree of category segmentation; use the classification feature with the largest degree of category segmentation to divide all government information data sequences into two categories and record them as target categories; obtain the classification feature with the largest degree of category segmentation of the target category, and use it to complete the classification operation of all government information data sequences of the target category; and so on, until the degree of category segmentation is less than the preset threshold, stop the classification operation, and obtain the fitted classification model.

[0045] It should be noted that there are many classification features in historical government information. Some classification features can separate government information of different themes to a large extent and separate government information of the same theme together, while some classification features cannot separate government information of different themes to a large extent and separate government information of the same theme together. Therefore, when performing classification processing, it is necessary to evaluate the segmentation effect of each classification feature, so as to select the classification features that can better separate the government information from all classification features for classification processing.

[0046] S20: Use a classification feature to divide all historical government information sequences into two categories.

[0047] It should be noted that, since it is not known which classification feature can better classify the historical government information sequence, we can only randomly select one for classification, determine its classification effect, and then select the appropriate classification feature based on the classification effect.

[0048] Preferably, as an example, a classification feature is used to classify all historical government information sequences into two categories, including:

[0049] The preset classification boundary of each classification feature is obtained, and all historical government information sequences are divided into two categories using the classification boundary.

[0050] For example, the arbitrarily selected classification feature is policy, and the preset classification boundary of the policy classification feature is: whether it belongs to cultural policy, among which the historical government information sequences that belong to cultural policy are classified into one category, and the historical government information sequences that do not belong to cultural policy are classified into another category.

[0051] S21: Calculate the degree of category segmentation.

[0052] It should be noted that in order to select appropriate classification features, the classification effect of the classification features needs to be evaluated. When the classification effect of the classification features is good, the historical government information sequences of different topics will be separated to a greater extent, and the historical government information sequences of the same topic will be separated together. Therefore, it can be evaluated based on this.

[0053] It should be further explained that the correlation between historical government information sequences of the same subject is relatively large, while the correlation between historical government information sequences of different subjects is relatively poor, and therefore the classification effect of the classification feature is evaluated based on this.

[0054] Preferably, as an example, calculating the category segmentation degree includes:

[0055] ;

[0056] Among them, principal component analysis is performed on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component direction corresponding to each eigenvalue. Indicates that one of the categories eigenvalues, Indicates that one of the categories eigenvalues, Indicates another category eigenvalues, Indicates another category The larger the eigenvalue, the stronger the distribution characteristics of the historical government information sequence in the direction of the principal component corresponding to the eigenvalue. In other words, more historical government information sequences are distributed in the direction of the principal component corresponding to the eigenvalue. Indicates that one of the categories The principal component direction corresponding to the eigenvalue of another category The closer the angle is to the principal component direction corresponding to the eigenvalues, the , indicating that the category is distributed in The information on the principal component direction corresponding to the eigenvalue is related to the information on the other category distribution in The greater the orthogonality of the information in the principal component direction corresponding to the eigenvalue, in other words, the more likely the category is distributed in the The information on the principal component direction corresponding to the eigenvalue is related to the information on the other category distribution in The correlation between the information in the principal component direction corresponding to the eigenvalue is small. The smaller the correlation between information of different categories, the greater the degree to which the classification feature separates information of different topics. Therefore, the classification effect of the classification feature is better. As weights to accurately reflect the orthogonality of the two categories of historical government information sequences. represents the maximum eigenvalue of one of the categories, represents the maximum eigenvalue of another category, It reflects the relevance of information in the category. The larger the value, the greater the degree to which the classification feature divides the information of the same topic together. M represents the number of feature values ​​in each category. Indicates the preset anti-zero coefficient, which is used to prevent the denominator from being zero. represents the hyperbolic tangent function, which is used for normalization. represents the absolute value symbol, Indicates the degree of category segmentation of the classification feature. This value is used to evaluate the classification effect of the classification feature.

[0057] It should be added that principal component analysis is performed on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component directions corresponding to each eigenvalue, including:

[0058] The principal component analysis algorithm is used to perform principal component analysis on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component direction corresponding to each eigenvalue.

[0059] S22: Using the classification feature with the largest degree of category segmentation, all government information data sequences are divided into two categories and recorded as target categories respectively; obtaining the classification feature with the largest degree of category segmentation of the target category, and using it to complete the classification operation of all government information data sequences of the target category; and so on, until the degree of category segmentation is less than the preset threshold, the classification operation is stopped to obtain the fitted classification model.

[0060] S220: Using the classification feature with the largest degree of category segmentation, all government information data sequences are divided into two categories and recorded as target categories respectively.

[0061] It should be noted that the greater the degree of category segmentation, the better the classification effect. Therefore, it is necessary to select the classification feature with the maximum degree of category segmentation to classify the government information data series.

[0062] Preferably, as an example, all government information data sequences are divided into two categories using the classification feature with the largest category segmentation degree, and are recorded as target categories respectively, including:

[0063] The preset classification boundary of the classification feature with the largest degree of category segmentation is obtained, and all government information data sequences are divided into two categories using the preset classification boundary of the classification feature with the largest degree of category segmentation.

[0064] S221: Obtain the classification feature with the largest category segmentation degree of the target category, and use it to complete the classification operation of all government information data sequences of the target category.

[0065] Preferably, as an example, obtaining the classification feature with the largest category segmentation degree of the target category and using it to complete the classification operation of all government information data sequences of the target category includes:

[0066] All historical government information sequences of each target category are divided into two categories using one of all classification features, the category segmentation degree is calculated, and the classification feature with the largest category segmentation degree is selected from all classification features to divide all historical government information sequences of the target category into two categories.

[0067] S222: The same process is repeated until the classification degree is less than a preset threshold, and the classification operation is stopped to obtain a fitted classification model.

[0068] Preferably, as an example, by analogy, until the category segmentation degree is less than a preset threshold, the classification operation is stopped to obtain a fitted classification model, including:

[0069] Any bottom-level category is recorded as the category to be analyzed, and a classification feature is selected from all classification features to divide the category to be analyzed into two categories, and the category segmentation degree is calculated to obtain the category segmentation degree of each classification feature. If the category segmentation degree of all classification features is less than the preset threshold, the classification operation of the category to be analyzed is stopped; the classification cutoff judgment is performed on other bottom-level categories until each bottom-level category meets the classification cutoff judgment condition, and a fitting classification model is obtained. This embodiment is described by taking the preset threshold value of 0.4 as an example, and other embodiments may take other values, which are not specifically limited in this embodiment.

[0070] It should be noted that, in this embodiment, the category without any sub-category is regarded as the bottom-level category.

[0071] S3: Use the fitted classification model to classify and identify the newly collected government information sequence.

[0072] Preferably, as an example, the newly collected government information sequence is classified and identified using the fitted classification model, including:

[0073] The subject information of all historical government information sequences in each bottom-level category in the fitted classification model is manually obtained; the newly collected government information sequence is input into the fitted classification model to obtain the category to which the newly collected government information sequence belongs, and the subject information of the category is used as the subject information of the newly collected government information sequence.

[0074] It can be seen that the classification model completed through fitting can quickly obtain the subject information of the government information sequence and realize the rapid identification of government information.

[0075] An embodiment of the present invention also discloses a smart government information identification system based on big data, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a smart government information identification method based on big data according to the present invention is implemented.

[0076] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0077] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory, a dynamic random access memory, a static random access memory, an enhanced dynamic random access memory, a high bandwidth memory, a hybrid storage cube, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0078] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0079] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying smart government information based on big data, characterized in that: Includes steps: Obtain historical government information sequences and corresponding classification features; All historical government information sequences are divided into two categories using a classification feature; Calculate the degree of category segmentation: , perform principal component analysis on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component direction corresponding to each eigenvalue, , Respectively represent one of the categories eigenvalues, the largest eigenvalue, and the The characteristic values , , Respectively represent another category eigenvalues, the largest eigenvalue, and the eigenvalues, Indicates that one of the categories The principal component direction corresponding to the eigenvalue of another category The angle between the principal component directions corresponding to the eigenvalues, M represents the number of eigenvalues ​​for each category, Indicates the preset anti-zero coefficient. represents the hyperbolic tangent function; using the classification feature with the largest category segmentation degree to divide all government information data sequences into two categories, each of which is recorded as the target category; obtaining the classification feature with the largest category segmentation degree of the target category, and using it to complete the classification operation of all government information data sequences of the target category; and so on, until the category segmentation degree is less than the preset threshold, the classification operation is stopped to obtain the fitted classification model; The fitted classification model is used to classify and identify the newly collected government information sequence.

2. According to the method for identifying smart government information based on big data according to claim 1, it is characterized in that: The method uses a classification feature to classify all historical government information sequences into two categories, including: The preset classification boundary of each classification feature is obtained, and all historical government information sequences are divided into two categories using the classification boundary.

3. According to the method for identifying smart government information based on big data in claim 1, it is characterized in that: The step of obtaining the classification feature with the largest classification segmentation degree of the target category and using it to complete the classification operation of all government information data sequences of the target category includes: All historical government information sequences of each target category are divided into two categories using one of all classification features, the category segmentation degree is calculated, and the classification feature with the largest category segmentation degree is selected from all classification features to divide all historical government information sequences of the target category into two categories.

4. According to the method for identifying smart government information based on big data in claim 1, it is characterized in that: The step of stopping the classification operation until the category segmentation degree is less than a preset threshold to obtain a fitted classification model includes: Any bottom-level category is recorded as the category to be analyzed, and a classification feature is selected from all classification features to divide the category to be analyzed into two categories. The category segmentation degree is calculated, and the category segmentation degree of each classification feature is obtained. If the category segmentation degree of all classification features is less than the preset threshold, the classification operation of the category to be analyzed is stopped; the classification cutoff judgment is performed on other bottom-level categories until each bottom-level category meets the classification cutoff judgment condition, and a fitted classification model is obtained.

5. According to the method for identifying smart government information based on big data in claim 1, it is characterized in that: The method of using the fitted classification model to classify and identify the newly collected government information sequence includes: Obtain the subject information of each bottom-level category in the fitted classification model; input the newly collected government information sequence into the fitted classification model to obtain the category to which the newly collected government information sequence belongs, and use the subject information of the category as the subject information of the newly collected government information sequence.

6. According to the big data-based smart government information identification method of claim 1, it is characterized in that: The principal component analysis of all historical government information sequences of each category is performed to obtain a number of eigenvalues ​​and the principal component direction corresponding to each eigenvalue, including: The principal component analysis algorithm is used to perform principal component analysis on all historical government information sequences of each category to obtain several eigenvalues ​​and the principal component direction corresponding to each eigenvalue.

7. According to the big data-based smart government information identification method of claim 1, it is characterized in that: The acquisition of historical government affairs information sequences and corresponding classification features includes: Acquire some historical government information, convert non-data information in the historical government information into data information to obtain converted historical government information, and record a sequence of the converted historical government information as a historical government information sequence; Obtain several classification features of historical government information.

8. A smart government information identification system based on big data, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for identifying smart government information based on big data according to any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Kmeans log classification method based on adaptive radius center selection

    CN116644328A

  • Text classification method based on self-attention mechanism

    CN115577111A

  • Information processing method, device and equipment and computer readable storage medium

    CN115878750A