A method and system for processing regional development factor data

Abstract

The present invention relates to the field of data processing, and more particularly to a method and system for processing regional development factor data. The method comprises the steps of obtaining regional development factor data and constructing a knowledge graph; calculating the macroscopic degree of each value based on the number of subsequent nodes of each value; removing nodes with duplicate values from the knowledge graph to obtain a simplified knowledge graph; iteratively segmenting the simplified knowledge graph based on the macroscopic degree; and encoding the resulting subsets after each segmentation to obtain a multi-layer encoding. This method improves the decoding efficiency of data with a high macroscopic degree.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method and system for processing regional development factor data. Background Art

[0002] In today's digital age, research and decision-making on regional economic development rely heavily on vast amounts of economic data. This data encompasses multidimensional information such as regional GDP, industrial structure, employment, trade, and fiscal revenue and expenditure. Not only is the volume enormous, but its scale is also exploding over time with the advancement of monitoring methods. Therefore, in the context of this surge in regional economic data, efficiently compressing regional economic development data is crucial to overcoming development barriers and unlocking its value, addressing storage, transmission, and analysis efficiency issues, and enabling accurate economic decision-making.

[0003] Existing data processing solutions typically employ a unified compression strategy, failing to consider the varying importance of data at the macro level of decision-making. In scenarios such as regional development planning and policymaking, decision-makers often require rapid access to macro data such as GDP growth rates and industrial structure proportions to comprehensively understand regional development trends. However, this unified compression approach results in inefficient decompression of highly macro-specific data, making it impossible to meet the demands of rapid decision-making. For example, during sudden economic policy adjustments, rapid analysis of macroeconomic trends in regional industries is required. However, the time required to decompress this macro data delays decision-making. Furthermore, the lack of an effective mechanism for analyzing the macro-level nature of data makes it difficult to accurately identify and efficiently process key macro information within massive amounts of data, severely hindering the scientific and intelligent advancement of regional development. To address the urgent need for rapid acquisition and efficient analysis of macro data in regional development, a differentiated compression method based on the macro-level characteristics of data is urgently needed. This method can ensure more efficient decompression of highly macro-specific data and facilitate rapid response and precise regional development decision-making. Summary of the Invention

[0004] In order to solve the problem of how to improve the decompression efficiency of highly macroscopic data, the present invention provides a method and system for processing regional development factor data.

[0005] In a first aspect, the present invention provides a method for processing regional development factor data, which adopts the following technical solution:

[0006] A method for processing regional development factor data, comprising the steps of:

[0007] Obtain regional development factor data;

[0008] Utilize all regional development factor data to construct a knowledge graph, calculate the macroscopic degree of each value according to the number of subsequent nodes of each value data, remove the nodes with repeated values in the knowledge graph to obtain a simplified knowledge graph, calculate the first splitting times of the nodes corresponding to each value according to the macroscopic degree, split and control the simplified knowledge graph according to the first splitting times to obtain the first-level subset, use the first-level subset as the encoding object, encode the first-level subset according to the number of data of all kinds of values in the first-level subset, and use the encoding of the first-level subset as the first-level encoding of each value therein;

[0009] In response to the existence of a first-layer subset with a data quantity greater than 1, the first-layer subset with a data quantity greater than 1 is used as an analysis subset; a second division number of nodes corresponding to each value is calculated according to the macroscopic degree, the analysis subset is divided and controlled according to the second division number to obtain a second-layer subset, the second-layer subset is used as an encoding object, the second-layer subset is encoded according to the data quantity of all values in the second-layer subset, and the code of the second-layer subset is used as the second-layer code of each value therein;

[0010] In response to the fact that the number of the second layer subset data does not exist and is greater than 1, all layer codes of the regional development factor data are stored.

[0011] The present invention reduces the number of matches for each layer of coding during decoding by means of layered coding, thereby improving decoding efficiency; further, different coding layers are set for data according to the macroscopic degree of the data, so that the coding layers of data with a large macroscopic degree are smaller, thereby improving the decoding efficiency of data with a large macroscopic degree; further, in the encoding process, each layer of subset is encoded according to the number of data of all kinds of values in each layer of subset, so that the coding setting for data with a large amount of data usage is shorter, and the coding setting for data with a small amount of data usage is longer, thereby effectively improving the compression effect.

[0012] Preferably, the step of calculating the macroscopic level of each value according to the number of nodes following each value data includes:

[0013] ;

[0014] in, represents the number of nodes on the i-th path after each node, N represents the number of paths after each node, H represents the macroscopic degree of each node, and norm() represents linear normalization processing;

[0015] The average of the macroscopic degrees of all nodes with the same value is taken as the macroscopic degree of this value.

[0016] The present invention accurately reflects the content of information covering various attributes after each node through the number of nodes after each node, and further accurately reflects the macroscopic level of the node.

[0017] Preferably, removing nodes with repeated values in the knowledge graph to obtain a simplified knowledge graph includes:

[0018] For any value, obtain all nodes of this value in the knowledge graph, and among all nodes of this value, obtain the node with the smallest macroscopic difference from this value as the retained node, and remove all other nodes of this value except the retained node;

[0019] During the removal process, if there are nodes both before and after other nodes with the same value, the node with the same value will be removed and the nodes before and after it will be connected; if there are nodes only before or after other nodes with the same value, the node with the same value will be directly removed.

[0020] The present invention improves the encoding quality by removing nodes with repeated values in the knowledge graph and setting different codes for data with the same value in the subsequent encoding process.

[0021] Preferably, the step of calculating the first splitting times of the node corresponding to each value according to the macroscopic level includes:

[0022] The number of connecting edges of each node in the simplified knowledge graph is obtained, the maximum value of the number of connecting edges of all nodes in the simplified knowledge graph is used as the first benchmark value, and the product of the macro degree and the first benchmark value is normalized to obtain the first number of divisions of the node corresponding to each value.

[0023] The present invention introduces the macro degree when calculating the number of divisions, so that nodes with a large macro degree have a larger number of divisions, and thus the nodes with a large macro degree can be divided into independent data faster, providing a basis for reducing the number of coding layers of data with a large macro degree.

[0024] Preferably, the step of controlling the segmentation of the simplified knowledge graph according to the first segmentation number to obtain the first-layer subset includes:

[0025] Get the connected nodes of any node in the simplified knowledge graph, get the absolute value of the difference between the macroscopic degree of the corresponding value of the node and the corresponding value of the connected node, sort all connected nodes in descending order according to the absolute value of the difference in macroscopic degree to obtain a connected node sequence, and use the first K1 connected nodes in the connected node sequence as split nodes, where K1 represents the first splitting number;

[0026] The connection edge between the node and the split node is cut off, and the nodes that are connected to each other are classified into a subset, and the obtained subset is recorded as the first-level subset.

[0027] Preferably, the step of taking the first layer subset as the encoding object, encoding the first layer subset according to the number of data of all values in the first layer subset, and taking the encoding of the first layer subset as the first layer encoding of each value therein includes:

[0028] Obtain the number of regional development factor data for each type of value in the first-layer subset, add up the number of regional development factor data for all types of values in the first-layer subset as the frequency of occurrence of the first-layer subset, take the first-layer subset as the encoding object, use the encoding compression algorithm to encode all first-layer subsets based on the frequency of occurrence to obtain the code of each first-layer subset, and use the code of the first-layer subset as the first-layer code of the value of each node therein.

[0029] The present invention takes into account that the Huffman coding algorithm will set shorter codes for data with high frequency of occurrence, and thus uses the data accumulation sum of all regional development factor data with all values in the subset to set the frequency of occurrence, providing a basis for effectively reducing the length of codes using a large amount of data and increasing the compression amount.

[0030] Preferably, the step of calculating the second splitting times of the corresponding node for each value according to the macroscopic level includes:

[0031] The number of remaining connection edges of each node is obtained, the maximum number of remaining connection edges of all nodes in the analysis subset is used as the second benchmark value, and the product of the macro degree and the second benchmark value is regularized to obtain the second splitting number of the corresponding node for each value.

[0032] Preferably, the step of calculating the second division times of the corresponding nodes of each value according to the macroscopic level, and performing division control on the analysis subset according to the second division times to obtain the second-layer subset includes:

[0033] Get the connected nodes of any node in the analysis subset, obtain the difference in macroscopic degree between the node and the connected node, sort all connected nodes in descending order according to the difference in macroscopic degree to obtain a connected node sequence, and use the first K2 connected nodes in the connected node sequence as split nodes, where K2 represents the second splitting number;

[0034] The connection edge between the node and the split node is cut off, and the nodes in the analysis subset that are connected to each other are classified into one subset, and the newly obtained subset is recorded as the second-level subset.

[0035] Preferably, encoding the second-layer subset according to the number of data of all values in the second-layer subset, and using the code of the second-layer subset as the second-layer code of each value therein, includes:

[0036] Obtain the number of regional development factor data for each value in the second-level subset, add up the number of regional development factor data for all values in the second-level subset as the occurrence frequency of the second-level subset, take the second-level subset as the encoding object, use the encoding compression algorithm to encode all the second-level subsets in the analysis subset based on the occurrence frequency to obtain the code of each second-level subset, and use the code of the second-level subset as the second-level code for each value therein.

[0037] In a second aspect, the present invention provides a regional development factor data processing system, which adopts the following technical solution:

[0038] A regional development factor data processing system 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 regional development factor data processing method is implemented.

[0039] By adopting the above technical solution, the above-mentioned method for processing regional development factor data is generated into a computer program and stored in a memory to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0040] The present invention has the following technical effects:

[0041] The present invention reduces the number of matching times for each layer of coding during decoding by means of layered coding, thereby improving decoding efficiency;

[0042] Furthermore, different coding layers are set for the data according to the macroscopic degree of the data, so that the coding layer number of the data with a high macroscopic degree is smaller, thereby improving the decoding efficiency of the data with a high macroscopic degree;

[0043] Furthermore, during the encoding process, each layer of subsets is encoded according to the number of data of all values in each layer of subsets, so that the encoding settings for those with more data usage are shorter and the encoding settings for those with less data usage are longer, effectively improving the compression effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a flowchart of a method for processing regional development factor data in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The embodiment of the present invention discloses a method for processing regional development factor data, referring to Figure 1 , including steps S1 to S5:

[0046] S1: Obtain regional development factor data.

[0047] Specifically, regional development factor data is obtained, and the types of regional development factor data include but are not limited to the following aspects: industrial data, enterprise data, and market data.

[0048] Industrial data includes, but is not limited to, the following: regional industrial structure data, industrial development scale data, industrial growth data, and industrial employment data; and local regional industrial structure data, industrial development scale data, industrial growth data, and industrial employment data. Enterprise data includes, but is not limited to, the following: regional enterprise operating status data, innovation capability data, and market share data; and local regional enterprise operating status data, innovation capability data, and market share data. Market data includes, but is not limited to, the following: regional market supply and demand data, price index data, and trade data; and local regional market supply and demand data, price index data, and trade data.

[0049] S2: Build a knowledge graph using all regional development factor data, calculate the macroscopic degree of each value according to the number of subsequent nodes of each value data, remove the nodes with repeated values in the knowledge graph to obtain a simplified knowledge graph, calculate the first splitting times of the nodes corresponding to each value according to the macroscopic degree, split and control the simplified knowledge graph according to the first splitting times to obtain the first-level subset, use the first-level subset as the encoding object, encode the first-level subset according to the number of data of all kinds of values in the first-level subset, and use the encoding of the first-level subset as the first-level encoding of each value in it.

[0050] It should be noted that the Huffman coding compression algorithm is a commonly used compression algorithm with good compression effects. Therefore, the Huffman coding compression algorithm can be used for data compression. The code obtained by the Huffman coding compression algorithm is a variable-length code, so it is impossible to identify the segmentation points of the codes of different data. Therefore, the algorithm needs to decode the data in the order of the data. Using the traditional Huffman coding compression algorithm for compression cannot achieve priority decoding of data with a high macroscopic degree. In order to improve the decoding efficiency of data with a high macroscopic degree, it is considered to add the idea of layered coding to the traditional Huffman coding algorithm, that is, to set different coding layers for different data according to the macroscopic situation of the data, set fewer coding layers for data with a high macroscopic degree, and set more coding layers for data with a low macroscopic degree, so that data with a high macroscopic degree can be decoded quickly.

[0051] S20: Build a knowledge graph using all regional development factor data.

[0052] It should be noted that in order to set the number of coding layers of data according to the macroscopic level of the data, the macroscopic level of the data must be obtained first.

[0053] It should be further explained that knowledge graphs can better reflect the associations and coverage relationships between data. Therefore, knowledge graphs can better reflect the macroscopic situation of data. First, we need to build a knowledge graph.

[0054] Preferably, as an example, all regional development factor data are used to construct a knowledge graph, including:

[0055] Based on all regional development factor data, a knowledge graph is constructed using an existing knowledge graph construction method. The existing knowledge graph construction method can be a Neo4j method or an Apache Jena method, or other knowledge graph construction methods, which are not specifically limited in this embodiment.

[0056] S21: Calculate the macroscopic level of each value according to the number of subsequent nodes of each value data.

[0057] Preferably, as an example, the macroscopic degree of each node is calculated according to the position of each node in the knowledge graph, including:

[0058]

[0059] in, represents the number of nodes on the i-th path after each node, N represents the number of paths after each node, H represents the macroscopic degree of each node, norm() represents linear normalization processing, this embodiment adopts the maximum and minimum value normalization method, other embodiments may adopt other normalization methods, and this embodiment does not specifically limit it;

[0060] The average of the macroscopic degrees of all nodes with the same value is taken as the macroscopic degree of this value.

[0061] It can be understood that the more nodes there are in the paths following the node, the more sub-attribute information the node contains, and thus the greater the macroscopic degree of the node. The larger the value, the more sub-attribute information there is after the node, and thus the greater the macro level of the node.

[0062] S22: Remove nodes with repeated values in the knowledge graph to obtain a simplified knowledge graph.

[0063] It should be noted that in order to allow data of different macroscopic levels to have different numbers of coding layers, we consider using a subset-based encoding method. This means that a large subset is iteratively divided until a single data point is contained in the subset. Each division is encoded once, so the more divisions are performed, the more coding layers there are.

[0064] It's important to note that when using Huffman coding, each value corresponds to a unique code. When using subset segmentation, it's easy to split data with the same value into different subsets, resulting in different codes for each subset. To prevent this, only one node for each value is retained in the knowledge graph.

[0065] Preferably, as an example, removing nodes with repeated values in the knowledge graph to obtain a simplified knowledge graph includes:

[0066] For any value, obtain all nodes of this value in the knowledge graph, and among all nodes of this value, obtain the node with the smallest macroscopic difference from this value as the retained node, and remove all other nodes of this value except the retained node;

[0067] During the removal process, if there are nodes both before and after other nodes with the same value, the node with the same value will be removed and the nodes before and after it will be connected; if there are nodes only before or after other nodes with the same value, the node with the same value will be directly removed.

[0068] It can be understood that through the above processing, only one node can be retained for one value in the knowledge graph, providing a basis for subsequent hierarchical coding.

[0069] S23: Calculate the first split times of the node corresponding to each value according to the macroscopic level.

[0070] It should be noted that when performing subset splitting, consider implementing subset splitting by segmenting the simplified knowledge graph. In order to reduce the number of splits for data with a high macro level, the number of splits for the nodes corresponding to the data at each split should be set according to the macro level of the data. This will allow the nodes corresponding to the data with a high macro level to have more splits at each split, thereby allowing the nodes corresponding to the data with a high macro level to be separated from other nodes more quickly, reducing the number of splits for the data, and thus providing a basis for subsequently reducing the number of coding layers.

[0071] Preferably, as an example, calculating the first split times of the node corresponding to each value according to the macroscopic level includes:

[0072] The number of connected edges of each node is obtained in the simplified knowledge graph, and the maximum value of the number of connected edges of all nodes in the simplified knowledge graph is used as the first benchmark value. If the product of the macro degree of the node corresponding to any value in the simplified knowledge graph and the first benchmark value is less than the number of connected edges, the rounded-down value of the product of the macro degree and the first benchmark value is used as the first splitting number of the node corresponding to this value; if the product of the macro degree of the node corresponding to this value and the first benchmark value is not less than the number of connected edges, the number of connected edges of the node is used as the first splitting number.

[0073] It can be understood that the number of segmentation times obtained in this way can ensure that the data with a large macroscopic degree is segmented more times, so that the data with a large macroscopic degree can be separated from other nodes more quickly, reducing the number of times the data participates in segmentation, and thus providing a basis for the subsequent reduction of the number of coding layers.

[0074] S24: Segment the simplified knowledge graph according to the first number of segmentations to obtain a first-layer subset.

[0075] Preferably, as an example, the simplified knowledge graph is segmented and controlled according to the first segmentation number to obtain the first layer subset, including:

[0076] Get the connected nodes of any node in the simplified knowledge graph, get the absolute value of the difference between the macroscopic degree of the corresponding value of the node and the corresponding value of the connected node, sort all connected nodes in descending order according to the absolute value of the difference in macroscopic degree to obtain a connected node sequence, and use the first K1 connected nodes in the connected node sequence as split nodes, where K1 represents the first splitting number;

[0077] The connection edge between the node and the split node is cut off, and the nodes that are connected to each other are classified into a subset, and the obtained subset is recorded as the first-level subset.

[0078] It can be understood that by controlling the segmentation through the first number of segmentation times, macroscopically large data can be separated from other nodes more quickly, reducing the number of times the data participates in segmentation, thereby reducing the number of coding layers in the subsequent step.

[0079] S25: Taking the first layer subset as a coding object, encoding the first layer subset according to the number of data of all kinds of values in the first layer subset, and using the code of the first layer subset as the first layer code of each kind of value in the first layer subset.

[0080] It should be noted that in order to increase the compression amount, the length of the encoding of data with a higher number of uses needs to be set shorter, and the length of the encoding of data with a lower number of uses needs to be set longer, so encoding processing needs to be performed based on this.

[0081] Preferably, as an example, the first layer subset is taken as the encoding object, the first layer subset is encoded according to the number of data of all values in the first layer subset, and the encoding of the first layer subset is used as the first layer encoding of each value therein, including:

[0082] Obtain the number of regional development factor data for each type of value in the first-layer subset, add up the number of regional development factor data for all types of values in the first-layer subset as the frequency of occurrence of the first-layer subset, take the first-layer subset as the encoding object, use the encoding compression algorithm to encode all first-layer subsets based on the frequency of occurrence to obtain the code of each first-layer subset, and use the code of the first-layer subset as the first-layer code of the value of each node therein.

[0083] It's understandable that every value in the first-level subset uses the code for that subset. Therefore, the more data of all possible values in the first-level subset, the more data using the code for that subset, resulting in a shorter code for the first-level subset, thus saving compression. The Huffman coding algorithm controls the code length based on frequency of occurrence. Therefore, the frequency of occurrence is set by the number of values in the first-level subset, adjusting the code length for frequently occurring data.

[0084] S3: In response to the existence of a data quantity greater than 1 in the first-layer subset, the first-layer subset with the data quantity greater than 1 is used as the analysis subset; the second division times of the corresponding nodes of each value are calculated according to the macroscopic degree, the analysis subset is divided and controlled according to the second division times to obtain the second-layer subset, the second-layer subset is used as the encoding object, the second-layer subset is encoded according to the data quantity of all values in the second-layer subset, and the encoding of the second-layer subset is used as the second-layer encoding of each value therein.

[0085] It should be noted that if the number of data in the first-layer subset is greater than 1, it means that it has not been segmented into independent data, and therefore needs to be further segmented and encoded.

[0086] S30: Calculate the second splitting times of the corresponding node for each value according to the macroscopic level.

[0087] Preferably, as an example, calculating the second splitting times of the corresponding node of each value according to the macroscopic level includes:

[0088] Obtain the number of remaining connection edges of each node, and take the maximum value of the number of remaining connection edges of all nodes in the analysis subset as the second benchmark value. If the product of the macro degree of the node corresponding to any value in the analysis subset and the second benchmark value is less than the number of remaining connection edges, then the rounded-down value of the product of the macro degree and the second benchmark value is taken as the second splitting number of the node corresponding to this value; if the product of the macro degree of the node corresponding to this value and the second benchmark value is less than the number of remaining connection edges, then the number of remaining connection edges of the node is taken as the second splitting number.

[0089] S31: performing segmentation control on the analysis subset according to a second segmentation number to obtain a second-layer subset.

[0090] Preferably, as an example, the second layer subsets are obtained by segmenting the analysis subset according to the second segmentation number, including:

[0091] Get the connected nodes of any node in the analysis subset, obtain the difference in macroscopic degree between the node and the connected node, sort all connected nodes in descending order according to the difference in macroscopic degree to obtain a connected node sequence, and use the first K2 connected nodes in the connected node sequence as split nodes, where K2 represents the second splitting number;

[0092] The connection edge between the node and the split node is cut off, and the nodes in the analysis subset that are connected to each other are classified into one subset, and the newly obtained subset is recorded as the second-level subset.

[0093] S32: Take the second layer subset as the encoding object, encode the second layer subset according to the number of data of all values in the second layer subset, and use the encoding of the second layer subset as the second layer encoding of each value in the second layer subset.

[0094] Preferably, as an example, the second layer subset is taken as the encoding object, the second layer subset is encoded according to the number of data of all values in the second layer subset, and the encoding of the second layer subset is used as the second layer encoding of each value therein, including:

[0095] Obtain the number of regional development factor data for each value in the second-level subset, add up the number of regional development factor data for all values in the second-level subset as the occurrence frequency of the second-level subset, take the second-level subset as the encoding object, use the encoding compression algorithm to encode all the second-level subsets in the analysis subset based on the occurrence frequency to obtain the code of each second-level subset, and use the code of the second-level subset as the second-level code for each value therein.

[0096] S4: In response to the fact that the number of the second layer subset data is not greater than 1, all layer codes of the regional development factor data are stored.

[0097] It should be noted that when the number of data in the second-layer subsets is less than or equal to 1, it means that there is only a single data in each second-layer subset. In this case, there is no need to split it and the layered coding is completed.

[0098] Preferably, as an example, all layers of regional development factor data are coded and stored, including:

[0099] For any layer, if any of the regional development factor data does not have a code at that layer, the empty code is used as the code for that data at that layer, and the codes of all the regional development factor data at that layer are concatenated and stored. Similarly, the codes of all the regional development factor data at each layer are stored. The correspondence between all the values of all the regional development factor data and the subsets at each layer is recorded, and the Huffman tree obtained for each subset is stored.

[0100] In particular, if there is only one value in the subset, then this value is used as the encoding object, and the number of times this value appears in all regional development factor data is used as the frequency of occurrence, and encoding is performed based on the frequency of occurrence.

[0101] It should be noted that for ease of explanation, the following example illustrates how to set a null code. Assume that this layer has four codes: 00, 11, 100, and 101. By observing, we can see that code 01 does not exist in these four codes, so we can set the null code to 01.

[0102] S5: Perform decoding processing.

[0103] Preferably, as an example, the decoding process includes:

[0104] Subsets are divided according to the corresponding relationship between all values and subsets of each layer to obtain subsets of each layer.

[0105] Starting from the first layer, the encoding of regional development factor data at each layer is obtained. Based on the mapping relationship between the encoding and the dataset in the Huffman tree of each subset, the subset to which each data belongs is decoded. This process continues until the corresponding encoding is empty. The decoding is completed by obtaining the data corresponding to the encoding in the previous layer of the empty encoding.

[0106] It's understandable that compared to all regional development factor data, each subset contains less data and fewer encoding types, resulting in fewer matching objects during decoding. Consequently, decoding each subset is more efficient. Therefore, data with fewer encoding layers can be decoded quickly.

[0107] An embodiment of the present invention also discloses a regional development factor data processing system, 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 regional development factor data processing method according to the present invention is implemented.

[0108] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0109] 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 resistive random access memory, dynamic random access memory, static random access memory, enhanced dynamic random access memory, high bandwidth memory, hybrid memory cube, etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium may be part of, accessible to, or connectable to the device.

Claims

1. A method for processing regional development factor data, characterized in that: Including steps: Obtain regional development factor data; Utilize all regional development factor data to construct a knowledge graph. Calculate the macroscopic degree of each value based on the number of subsequent nodes of each value. Remove nodes with repeated values from the knowledge graph to obtain a simplified knowledge graph. Calculate the first split times of the corresponding nodes for each value based on the macroscopic degree. The simplified knowledge graph is segmented and controlled according to the first number of segmentations to obtain a first-level subset, including: obtaining the connected nodes of any node in the simplified knowledge graph, obtaining the absolute value of the difference in macroscopic degree between the corresponding value of the node and the corresponding value of the connected node, arranging all connected nodes in descending order according to the absolute value of the difference in macroscopic degree to obtain a connected node sequence, and taking the first K1 connected nodes in the connected node sequence as segmentation nodes, where K1 represents the first number of segmentations; cutting off the connecting edge between the node and the segmentation node, and classifying the nodes that have a connection relationship with each other into a subset, and recording the obtained subset as the first-level subset; Take the first layer subset as the encoding object, encode the first layer subset according to the number of data of all values in the first layer subset, and use the code of the first layer subset as the first layer code of each value in it; In response to the existence of a first-layer subset with a data quantity greater than 1, the first-layer subset with a data quantity greater than 1 is used as an analysis subset; a second division number of nodes corresponding to each value is calculated according to the macroscopic degree, the analysis subset is divided and controlled according to the second division number to obtain a second-layer subset, the second-layer subset is used as an encoding object, the second-layer subset is encoded according to the data quantity of all values in the second-layer subset, and the code of the second-layer subset is used as the second-layer code of each value therein; In response to the fact that the number of the second layer subset data does not exist and is greater than 1, all layer codes of the regional development factor data are stored.

2. A method for processing regional development factor data according to claim 1, characterized in that: The macroscopic level of each value is calculated based on the number of nodes after each value data, including: ； in, represents the number of nodes on the i-th path after each node, N represents the number of paths after each node, H represents the macroscopic degree of each node, and norm() represents linear normalization processing; The average of the macroscopic degrees of all nodes with the same value is taken as the macroscopic degree of this value.

3. A method for processing regional development factor data according to claim 1, characterized in that: The simplified knowledge graph obtained by removing nodes with repeated values in the knowledge graph includes: For any value, obtain all nodes of this value in the knowledge graph, and among all nodes of this value, obtain the node with the smallest macroscopic difference from this value as the retained node, and remove all other nodes of this value except the retained node; During the removal process, if there are nodes both before and after other nodes with the same value, the node with the same value will be removed and the nodes before and after it will be connected; if there are nodes only before or after other nodes with the same value, the node with the same value will be directly removed.

4. A method for processing regional development factor data according to claim 1, characterized in that: The first splitting times of the node corresponding to each value is calculated according to the macroscopic level, including: The number of connecting edges of each node in the simplified knowledge graph is obtained, the maximum value of the number of connecting edges of all nodes in the simplified knowledge graph is used as the first benchmark value, and the product of the macro degree and the first benchmark value is normalized to obtain the first number of divisions of the node corresponding to each value.

5. A method for processing regional development factor data according to claim 1, characterized in that: The step of taking the first layer subset as the encoding object, encoding the first layer subset according to the number of data of all values in the first layer subset, and taking the encoding of the first layer subset as the first layer encoding of each value in the first layer subset includes: Obtain the number of regional development factor data for each type of value in the first-layer subset, add up the number of regional development factor data for all types of values in the first-layer subset as the frequency of occurrence of the first-layer subset, take the first-layer subset as the encoding object, use the encoding compression algorithm to encode all first-layer subsets based on the frequency of occurrence to obtain the code of each first-layer subset, and use the code of the first-layer subset as the first-layer code of the value of each node therein.

6. A method for processing regional development factor data according to claim 1, characterized in that: The calculating, according to the macroscopic level, the second splitting times of the corresponding node for each value includes: The number of remaining connection edges of each node is obtained, the maximum number of remaining connection edges of all nodes in the analysis subset is used as the second benchmark value, and the product of the macro degree and the second benchmark value is regularized to obtain the second splitting number of the corresponding node for each value.

7. A method for processing regional development factor data according to claim 1, characterized in that: The second division times of the corresponding nodes of each value are calculated according to the macroscopic level, and the second-layer subsets are obtained by performing division control on the analysis subset according to the second division times, including: Get the connected nodes of any node in the analysis subset, obtain the difference in macroscopic degree between the node and the connected node, sort all connected nodes in descending order according to the difference in macroscopic degree to obtain a connected node sequence, and use the first K2 connected nodes in the connected node sequence as split nodes, where K2 represents the second splitting number; The connection edge between the node and the split node is cut off, and the nodes in the analysis subset that are connected to each other are classified into one subset, and the newly obtained subset is recorded as the second-level subset.

8. A method for processing regional development factor data according to claim 1, characterized in that: The encoding of the second-layer subset according to the number of data of all values in the second-layer subset, and using the encoding of the second-layer subset as the second-layer encoding of each value therein, includes: Obtain the number of regional development factor data for each value in the second-level subset, add up the number of regional development factor data for all values in the second-level subset as the occurrence frequency of the second-level subset, take the second-level subset as the encoding object, use the encoding compression algorithm to encode all the second-level subsets in the analysis subset based on the occurrence frequency to obtain the code of each second-level subset, and use the code of the second-level subset as the second-level code for each value therein.

9. A regional development factor data processing system, 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 processing regional development factor data according to any one of claims 1 to 8 is implemented.

Images