Innovative service policy intelligent matching method and system based on data analysis

Abstract

The invention discloses an innovation service policy intelligent matching method and system based on data analysis, and relates to the technical field of data analysis, the system comprises a control center, the control center is connected with a semantic collection module, an atlas processing module, an intelligent analysis module and a key matching module; demand information data and comprehensive policy data are acquired in a semantic acquisition module, and a policy knowledge base is constructed; updating and replacing the policy knowledge base through an atlas processing module to obtain a collaborative word segmentation knowledge base and carrying out image-text construction to obtain a policy information atlas; performing flow section conversion and information combination on the policy information atlas in the intelligent analysis module to obtain an atlas extraction set; performing form unification on the demand information data in a key matching module, obtaining a demand data code segment, performing code segment matching and matching verification, and obtaining a push policy set; the policy matching efficiency is improved, personalized and accurate policy matching is realized, the accuracy of policy applicability evaluation is enhanced, and the policy implementation effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to an innovative service policy intelligent matching method and system based on data analysis. Background Art

[0002] Data analysis is the process of using statistics, algorithms, and computer technology to analyze data and extract valuable information from it. It is an important part of data science.

[0003] Innovative service policies refer to a series of policies, measures, and plans formulated by the government or relevant departments to promote innovative development in the service industry, optimize the service industry structure, and improve service quality and efficiency. With the rapid development of my country's economy, innovative service policies have played a significant role in promoting industrial restructuring and enhancing corporate competitiveness.

[0004] However, there are some problems in the existing innovation service policy matching process: policies are scattered across various government departments and public service platforms, making it difficult for enterprises to obtain policy information; manual screening and matching of policies requires a lot of time and energy and is inefficient; in the policy matching process, there is a lack of in-depth analysis of the characteristics and needs of service recipients, resulting in insufficient assessment of policy applicability; there is a lack of effective means to track and evaluate the effectiveness of policy implementation, which affects policy adjustment and optimization; therefore, according to the specific characteristics and needs of service recipients, innovative service policies suitable for their development should be automatically identified and recommended to ensure the effective use of policy resources. To this end, an intelligent matching method and system for innovation service policies based on data analysis are now provided. Summary of the Invention

[0005] The purpose of the present invention is to provide an innovative service policy intelligent matching method and system based on data analysis.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An innovative service policy intelligent matching system based on data analysis includes a control center connected to a semantic acquisition module, a graph processing module, an intelligent analysis module, and a key matching module;

[0008] The process of collecting demand information data and comprehensive policy data by the semantic collection module includes:

[0009] Set up a multilateral capture terminal and a user collection terminal, collect target user demand information data through the user collection terminal, and collect data from policy producers through the multilateral capture terminal to obtain comprehensive policy data;

[0010] Perform data cleaning on comprehensive policy data to obtain net value policy data, combine and match the net value policy data to obtain policy information pairs, and build a policy knowledge base based on the policy information pairs.

[0011] The process of obtaining the policy information map includes:

[0012] Conduct standard collaboration on the obtained policy knowledge base to obtain net value policy segmentation words, update and replace the policy knowledge base based on the net value policy segmentation words, and obtain a collaborative segmentation knowledge base;

[0013] Perform entity recognition on the obtained collaborative word segmentation knowledge base to obtain collaborative knowledge entities, perform relationship matching on the collaborative knowledge entities based on the collaborative word segmentation knowledge base to obtain entity docking relationships;

[0014] Based on the collaborative word segmentation knowledge base, attributes of collaborative knowledge entities are extracted to obtain entity attribute words.

[0015] Based on collaborative knowledge entities, entity docking relationships and entity attribute words, graph and text construction is carried out to obtain a policy information graph.

[0016] The process of constructing graphs and texts based on collaborative knowledge entities, entity docking relationships, and entity attribute words includes:

[0017] Set up a construction space, upload the collaborative knowledge entity to the construction space, perform graphic conversion on the collaborative knowledge entity through the construction space, and obtain the collaborative entity node;

[0018] The collaborative entity nodes are linked according to the entity docking relationship to obtain the node docking edges, and the node docking edges are marked and paired based on the entity attribute words to obtain the policy information map.

[0019] The process of obtaining the atlas extraction set includes:

[0020] The collaborative entity nodes, node docking edges and entity attribute words are marked as pre-processed data, and the pre-processed data is converted into flow segments to obtain the policy data flow;

[0021] Upload the policy data stream to the construction space, intercept the code element of the policy data stream, and obtain the policy data code segment;

[0022] According to the policy information map, the policy data code segments are grouped within the map to obtain a net value data group, and the net value data group is clustered and changed to obtain a changed net value group;

[0023] The net value change groups are combined according to the policy information graph to obtain a graph extraction group set.

[0024] The process of combining information of the net value change group according to the policy information map includes:

[0025] Perform association selection on the obtained net change value group to obtain the node edge attribute group;

[0026] The direction group distance is set, and the obtained node edge attribute groups are extracted and arranged according to the direction group distance to obtain a graph extraction group set.

[0027] The process of obtaining a push policy set includes:

[0028] Unify the form of demand information data to obtain demand data code segments;

[0029] Upload the acquired demand data code segments to the construction space, perform code segment matching on the demand data code segments through the graph extraction group set, and obtain the matching policy code group;

[0030] Perform matching verification on the obtained matching policy code group to obtain a push policy set.

[0031] The process of matching the required data segments by extracting the graph group includes:

[0032] Coordinate and match the obtained demand data code segment with the change net value group in the atlas extraction group set until all the change net value groups in the atlas extraction group set are coordinated and matched with the demand data code segment to obtain an intra-group matching code;

[0033] Perform state decomposition on the matching code within the group to obtain balanced code and unbalanced code;

[0034] Update the code state of the matching code within the group according to the balance code and the imbalance code to obtain the updated result code;

[0035] The update result codes are divided and combined based on the graph extraction group set to obtain the matching policy code group.

[0036] Based on the above-mentioned innovative service policy intelligent matching system based on data analysis, the present invention also provides an innovative service policy intelligent matching method based on data analysis, comprising the following steps:

[0037] Step 1: Collect demand information data and comprehensive policy data, and build a policy knowledge base based on the comprehensive policy data;

[0038] Step 2: Update and replace the policy knowledge base to obtain a collaborative word segmentation knowledge base, perform entity recognition on the collaborative word segmentation knowledge base to obtain collaborative knowledge entities, perform relationship matching and attribute extraction on the collaborative knowledge entities, and perform graph-text construction with the collaborative knowledge entities to obtain a policy information graph;

[0039] Step 3: Perform stream segment conversion on the policy information graph to obtain the policy data stream, intercept the code element of the policy data stream to obtain the policy data code segment, group the policy data code segment within the graph and cluster the changes to obtain the net value group, combine the information of the net value group to obtain the graph extraction group set;

[0040] Step 4: Unify the form of the demand information data to obtain the demand data code segment, perform code segment matching and matching verification on the demand data code segment through the graph extraction group set, and obtain the push policy set.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] Update and replace the constructed policy knowledge base, obtain a collaborative word segmentation knowledge base and perform graphic and text construction to obtain a policy information map; convert the collected innovation service policy information into a related map form based on their mutual relationships, greatly improving the speed of intelligent retrieval and intuitively displaying the overall picture and details of the policy system;

[0043] Perform flow segment conversion and information combination on the policy information graph to obtain a graph extraction group set, unify the form of the demand information data, perform code segment matching and matching verification on the demand data code segments through the graph extraction group set, and obtain the push policy set; convert the obtained graph extraction group set into a data stream form, perform automatic matching of innovative service policies, achieve personalized and accurate policy matching, enhance the accuracy of policy applicability assessment, and improve the effectiveness of policy implementation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 This is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0046] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0047] like Figure 1As shown, the innovative service policy intelligent matching system based on data analysis includes a control center connected to a semantic acquisition module, a graph processing module, an intelligent analysis module and a key matching module;

[0048] The semantic collection module is used to collect demand information data and comprehensive policy data to build a policy knowledge base. The specific process includes:

[0049] A multilateral capture terminal and a user collection terminal are set up. The user collection terminal is used to collect demand information data of target users. The target users are users who need to obtain innovation service policy information. The demand information is keyword information about relevant policies uploaded by target users in order to obtain the required innovation service policies, including basic user information, enterprise characteristic information, innovation activity information, and policy demand information. Among them, the enterprise characteristic information represents the scale, type, and qualification certification of the target user's enterprise, and the innovation activity information represents the R&D status and the transformation of R&D results;

[0050] Collect data from policy producers through multilateral capture terminals to obtain comprehensive policy data and production time;

[0051] Furthermore, the policy production sources include policy documents, news reports and official announcements, and the production time indicates the release time of the innovation service policy;

[0052] The comprehensive policy data includes policy name, policy text, policy background, policy objectives, support measures, applicable conditions, application process and policy effects; in particular, the collected comprehensive policy data is in the form of Chinese text;

[0053] Perform data cleaning on the obtained comprehensive policy data to obtain net value policy data;

[0054] Furthermore, the data cleaning means identifying the innovative service policies in the collected comprehensive policy data and deleting irrelevant information, such as HTMI tags, irrelevant advertising content, and duplicate innovative service policies;

[0055] Combine and match the obtained net worth policy data to obtain policy information pairs;

[0056] Furthermore, the combination pairing means grouping the net worth policy data after data cleaning according to policy names, classifying them into groups with one policy name, and assigning corresponding policy text, policy background, policy objectives, support measures, applicable conditions, application procedures, and policy effects to each policy name to form a policy information pair;

[0057] Building a policy knowledge base based on the obtained policy information pairs, and uploading the obtained policy information pairs to the policy knowledge base based on production time;

[0058] Furthermore, uploading to the policy knowledge base is performed in chronological order of production time, that is, the order of policy information pairs in the policy knowledge base is based on the chronological order of production time.

[0059] The graph processing module is used to update and replace the policy knowledge base, obtain the collaborative word segmentation knowledge base and perform graph-text construction to obtain the policy information graph. The specific process includes:

[0060] Conduct standard collaboration on the obtained policy knowledge base to obtain net value policy segmentation;

[0061] Furthermore, the standard collaboration means standardizing and unifying the policy information pairs in the policy knowledge base to obtain standard net value policy pairs, and then performing text segmentation processing on the standard net value policy pairs to obtain net value policy segmentations;

[0062] The standardized unified representation is to represent similar categories of information with the same category, for example, "province" and "autonomous region" are standardized into the "region" category, and "policy time", "policy start time" and "policy end time" are standardized into "policy start and end time";

[0063] The text word segmentation processing means performing word segmentation processing on the text of the obtained standard net value policy pair, that is, extracting nouns, verbs and attribute keywords;

[0064] The policy knowledge base is updated and replaced based on the net value policy segmentation, that is, the policy information pairs in the policy knowledge base are replaced with the corresponding net value policy segmentation to obtain a collaborative segmentation knowledge base;

[0065] Perform entity recognition on the obtained collaborative word segmentation knowledge base to obtain collaborative knowledge entities;

[0066] Furthermore, the entity recognition refers to identifying text data in the collaborative word segmentation knowledge base through natural language processing technology, that is, identifying key entities in the net value policy word segmentation through natural language processing technology, that is, collaborative knowledge entities, for example, key entities are policy names, government departments, service objects and policy categories;

[0067] Perform relationship matching on collaborative knowledge entities based on the collaborative word segmentation knowledge base to obtain entity docking relationships;

[0068] It should be further explained that, in the specific implementation process, the relationship matching means analyzing the relationship between collaborative knowledge entities to obtain the relationship between each policy name and other collaborative knowledge entities. The relationship includes release, applicability, support, and qualification conditions. That is, through the collaborative word segmentation knowledge base, the relationship between two adjacent collaborative knowledge entities is obtained, which is the entity docking relationship. For example, the existing innovation service policy A corresponds to the issuing department a1 and the service objects b1, b2 and b3. Then the entity docking relationship of policy A is that it is issued by the issuing department a1 and is applicable to the service objects b1, b2 and b3.

[0069] Based on the collaborative word segmentation knowledge base, attributes of collaborative knowledge entities are extracted to obtain entity attribute words. That is, the attributes of collaborative knowledge entities are matched in the collaborative word segmentation knowledge base to obtain the attributes corresponding to the collaborative knowledge entities, which are entity attribute words. For example, entity attribute words are the implementation time, service content, enjoyment conditions and implementation effect of the policy;

[0070] Based on the obtained collaborative knowledge entities, entity docking relationships and entity attribute words, graph and text are constructed to obtain a policy information graph;

[0071] It should be further explained that, in the specific implementation process, the process of constructing the graphic includes:

[0072] Setting a construction space, wherein the construction space is used to generate a policy information map;

[0073] Upload the obtained collaborative knowledge entity to the construction space, perform graphical conversion on the collaborative knowledge entity through the construction space, and obtain the collaborative entity node;

[0074] Furthermore, the graph conversion means arranging the uploaded collaborative knowledge entities in the construction space according to the production time sequence of the policy name, converting the collaborative knowledge entities into nodes, and marking the corresponding collaborative knowledge entities in the nodes, wherein the converted node form is converted based on the construction method of the knowledge graph and corresponds to the nodes in the knowledge graph;

[0075] According to the obtained entity docking relationship, the obtained collaborative entity nodes are linked to obtain node docking edges;

[0076] Furthermore, the relationship link is represented by finding the collaborative entity nodes involved in the entity docking relationship in the construction space with the collaborative entity node as the center according to the entity docking relationship, and linking the involved collaborative entity nodes in pairs, that is, linking the two collaborative entity nodes involved in the entity docking relationship. The obtained link channel is the node docking edge, wherein the obtained node connection edge can be bidirectional or unidirectional, depending on the relationship between the two collaborative entity nodes linked in the entity docking relationship;

[0077] Based on entity attribute words, the obtained node docking edges are labeled and paired to obtain a policy information graph;

[0078] Furthermore, the tag pairing means matching node docking edges according to entity attribute words within the construction space, and marking the corresponding entity attribute words on the node docking edges that are successfully matched, until all entity attribute words are marked and paired, thereby obtaining a policy information graph consisting of collaborative entity nodes, node docking edges, and entity attribute words; wherein a successful match means taking a collaborative entity node linked by the node docking edge as the matching party, recorded as the matching entity node, obtaining the entity attribute words related to the matching entity node, and marking them on the node docking edge. In particular, the position of the mark is related to the direction of the node connection edge;

[0079] In particular, in this embodiment, the generated policy information graph is constructed based on the construction steps of the knowledge graph.

[0080] The intelligent analysis module is used to perform flow segment conversion and information combination on the policy information graph to obtain a graph extraction group set. The specific process includes:

[0081] Mark the collaborative entity nodes, node docking edges and entity attribute words included in the policy information graph as pre-processed data;

[0082] Performing stream segment conversion on the obtained pre-processed data to obtain a policy data stream, wherein the policy data stream includes an entity node data stream, a docking edge data stream, and an attribute word data stream;

[0083] Furthermore, the flow segment conversion means converting the obtained pre-processed data into a data stream consisting of binary code elements. According to the collaborative entity nodes, node docking edges and entity attribute words included in the pre-processed data, the policy data stream after the flow segment conversion includes the entity node data stream, the docking edge data stream and the attribute word data stream.

[0084] Upload the obtained policy data stream to the construction space, and associate the obtained policy data stream with the policy information map through the construction space, that is, mark the obtained policy data stream at the corresponding position in the policy information map in the construction space;

[0085] Performing code element interception on the obtained policy data stream to obtain policy data code segments, wherein the policy data code segments include entity node code segments, docking edge code segments, and attribute word code segments;

[0086] Furthermore, the code element interception means grouping the policy data stream into groups of 7 binary code elements to obtain policy data code segments. Specifically, if the last group contains less than 7 binary code elements, a binary code element "0" is added after the last group of binary code elements until 7 binary code elements are satisfied. Furthermore, based on the entity node data stream, the docking edge data stream, and the attribute word data stream included in the policy data stream, code elements are intercepted for the entity node data stream, the docking edge data stream, and the attribute word data stream, respectively. The obtained policy data code segments include the entity node code segments, the docking edge code segments, and the attribute word code segments.

[0087] According to the obtained policy information graph, the policy data code segments are grouped within the graph to obtain a net value data group, and the obtained net value data group is marked as J;

[0088] It should be further explained that, in a specific implementation process, the process of grouping within the graph includes:

[0089] The obtained policy data code segments are grouped within the segments to obtain net value code elements and docking code elements;

[0090] Furthermore, the intra-segment grouping means recording the first four binary code elements in the policy data code segment as net value code elements, and recording the last three binary code elements in the policy data code segment as docking code elements;

[0091] The policy data code segments are grouped based on the policy data stream, with three policy data code segments forming a group, forming a matrix of three rows and seven columns, which is the net value data group. In particular, if there are any remaining policy data code segments after the policy data stream is divided into groups of three, the remaining policy data code segments are combined with the previous policy data code segments until three policy data code segments form a group, which is recorded as the net value data group.

[0092] According to the entity node code segment, docking edge code segment and attribute word code segment included in the policy data code segment, the net value data group correspondingly includes a node net value group, a docking net value group and an attribute net value group;

[0093] The net value data group is partitioned into modules according to the obtained net value code element and docking code element to obtain a net value module group and a docking module group, wherein the module partitioning means that according to the corresponding binary code element of the net value code element and the docking code element, the first four columns in the net value data group are recorded as the net value module group, that is, a matrix of three rows and four columns, and the last three columns in the net value data group are recorded as the docking module group, that is, a matrix of three rows and three columns, and the obtained net value module group is marked as k, and the obtained docking module group is marked as i, then the net value data group can be recorded as

[0094] The obtained net value data group is clustered and changed to obtain a changed net value group, which includes a changed node group, a changed docking group and a changed attribute group. The obtained changed net value group is marked as J 变 ;

[0095] Furthermore, the cluster change indicates that a row modulo 2 addition is performed on the net value data group until the docking module group in the net value data group becomes a unit matrix, and the net value data group after the modulo 2 addition is recorded as a changed net value group, wherein the modulo 2 addition is a modulo 2 addition operation performed on two adjacent rows in the net value data group;

[0096] In particular, the changed net value group is obtained by changing the net value data group. According to the node net value group, the docking net value group and the attribute net value group included in the net value data group, the changed net value group includes the changed node group, the changed docking group and the changed attribute group.

[0097] Combining the obtained net value change groups according to the policy information graph to obtain a graph extraction group set;

[0098] It should be further explained that, in a specific implementation process, the information combination process includes:

[0099] In the policy information graph, obtain the change net value group corresponding to the collaborative entity node, that is, the change node group, obtain the change net value group corresponding to the node docking edge, that is, the change docking group, and obtain the change net value group corresponding to the entity attribute word, that is, the change attribute group;

[0100] Performing association selection on the obtained change node group, change docking group, and change attribute group to obtain a node edge attribute group, wherein the association selection means associating the node docking edge with the adjacent collaborative entity node and entity attribute word according to the relationship link of the node docking edge, combining them into a "collaborative entity node-node docking edge-entity attribute word" format, obtaining a change net value group corresponding to the "collaborative entity node-node docking edge-entity attribute word", converting the "collaborative entity node-node docking edge-entity attribute word" into a change net value group format, that is, "change node group-change docking group-change attribute group", and recording it as a node edge attribute group;

[0101] Set the direction grouping distance, which indicates setting a selection direction in the policy data stream and selecting the corresponding number of change net value groups based on the selection direction. For example, if the direction grouping distance is to select e change net value groups from the starting end of the policy data stream, then in the entity node data stream, e change net value groups are selected from the starting end, in the docking edge data stream, e change docking groups are selected from the starting end, and in the attribute word data stream, e change attribute groups are selected from the starting end.

[0102] Extract and arrange the obtained node edge attribute groups according to the set direction group distance to obtain a graph extraction group set;

[0103] The extraction arrangement means extracting the corresponding change net value group at the corresponding position in the node edge attribute group according to the set direction group distance, which is recorded as the selected net value group. The selected net value group includes the selected node group, the selected docking group and the selected attribute group, and is combined into "selected node group-selected docking group-selected attribute group" in the order of "change node group-change docking group-change attribute group", and recorded as the graph extraction group set;

[0104] Furthermore, the selected net value group is a changing net value group extracted according to the directional group interval. For example, according to the directional group interval of "selecting e changing net value groups from the starting end of the policy data stream", e changing node groups are selected starting from the starting end of the policy data stream of the collaborative entity node in the "collaborative entity node-node docking edge-entity attribute word", e changing docking groups are selected starting from the starting end of the policy data stream of the node docking edge, and e changing attribute groups are selected starting from the starting end of the policy data stream of the entity attribute word, which together constitute the selected net value group.

[0105] The key matching module is used to unify the form of demand information data, obtain demand data code segments, perform code segment matching and match verification, and obtain a push policy set. The specific process includes:

[0106] Unify the form of the acquired demand information data to obtain the demand data code segment;

[0107] It should be further explained that, in the specific implementation process, the process of morphological unification includes:

[0108] Conduct standard collaboration on demand information data to obtain net value demand segmentation;

[0109] Perform flow segment conversion on the obtained net value demand segmentation to obtain the demand data stream, perform code element interception on the obtained demand data stream to obtain the demand data code segment, and mark the obtained demand data code segment as X m , where m represents the number of the required data code segment, m=1, 2, 3, ..., v1, v1 is a positive integer;

[0110] Upload the acquired demand data code segments to the construction space, perform code segment matching on the demand data code segments through the graph extraction group set, and obtain the matching policy code group;

[0111] Furthermore, the code segment matching means coordinating and matching the obtained demand data code segment with the change net value group in the spectrum extraction group set until all the change net value groups in the spectrum extraction group set are coordinated and matched with the demand data code segment, that is, performing matrix multiplication of the obtained demand data code segment and the change net value group to obtain the intra-group matching code, and marking the obtained intra-group matching code as ZN g ,in, Indicates J 变-g The matrix transpose of J 变-g represents the net value change group in the atlas extraction group set in the order of number, g represents the number of the net value change group in the atlas extraction group set, g = 1, 2, 3, ..., v2, v2 is a positive integer;

[0112] Perform state decomposition on the matching code within the group to obtain balanced code and unbalanced code;

[0113] Furthermore, the state decomposition indicates that an intra-group matching code is obtained according to the result of the coordinated matching, and the matrix multiplication result of the intra-group matching code is a zero matrix and a non-zero matrix. The intra-group matching code with a zero matrix after the coordinated matching is recorded as a balanced code, and the intra-group matching code with a non-zero matrix after the coordinated matching is recorded as an unbalanced code.

[0114] The code state of the intra-group matching code is updated according to the obtained balanced code and unbalanced code to obtain an updated result code. The code state update indicates analyzing the intra-group matching code. If the intra-group matching code is a balanced code, the intra-group matching code is marked as 0; if the intra-group matching code is an unbalanced code, the intra-group matching code is marked as 1;

[0115] Based on the graph extraction group set, the update result code is divided and combined to obtain the matching policy code group;

[0116] Furthermore, the code combination indicates that the update result codes are combined according to the coding order of the net change value group in the graph extraction group set to obtain a matching policy code group. For example, if the update result codes ZN1=0, ZN2=0, ZN3=0, ZN4=0, and ZN5=0 are present, the matching policy code group after code combination=

[00000] ;

[0117] Perform matching verification on the obtained matching policy code group to obtain a push policy set;

[0118] It should be further explained that, in the specific implementation process, the matching verification process includes:

[0119] The obtained matching policy code group result is monitored. When all binary code elements of the matching policy code group are 0, the "collaborative entity node-node docking edge-entity attribute word" corresponding to the matching policy code group is obtained, and the policy name corresponding to the "collaborative entity node-node docking edge-entity attribute word" is obtained. The comprehensive policy data corresponding to the policy name is obtained and recorded as a push policy set. The obtained push policy set is uploaded to the control center and sent to the target user through control communication. Among them, "obtaining the comprehensive policy data corresponding to the policy name" means obtaining all information data corresponding to the policy name;

[0120] When all binary code elements of the matching policy code group are not all 0, the next graph extraction group set is replaced for code segment matching, and the process of obtaining the matching policy code group is repeated until all binary code elements of the obtained matching policy code group are 0, and the innovative service policy required by the target user is matched.

[0121] Based on the above-mentioned innovative service policy intelligent matching system based on data analysis, the present invention also provides an innovative service policy intelligent matching method based on data analysis, comprising the following steps:

[0122] Step 1: Collect demand information data and comprehensive policy data, and build a policy knowledge base based on the comprehensive policy data;

[0123] Step 2: Update and replace the policy knowledge base to obtain a collaborative word segmentation knowledge base, perform entity recognition on the collaborative word segmentation knowledge base to obtain collaborative knowledge entities, perform relationship matching and attribute extraction on the collaborative knowledge entities, and perform graph-text construction with the collaborative knowledge entities to obtain a policy information graph;

[0124] Step 3: Perform stream segment conversion on the policy information graph to obtain the policy data stream, intercept the code element of the policy data stream to obtain the policy data code segment, group the policy data code segment within the graph and cluster the changes to obtain the net value group, combine the information of the net value group to obtain the graph extraction group set;

[0125] Step 4: Unify the form of the demand information data to obtain the demand data code segment, perform code segment matching and matching verification on the demand data code segment through the graph extraction group set, and obtain the push policy set.

[0126] The preferred embodiments of the present invention disclosed above are intended only to help illustrate the present invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the present invention to specific embodiments. Obviously, many modifications and variations are possible based on the contents of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better understand and utilize the present invention. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An innovative service policy intelligent matching system based on data analysis, including a control center, characterized in that: The control center is connected to a semantic acquisition module, a graph processing module, an intelligent analysis module and a key matching module; The semantic collection module is used to collect demand information data and comprehensive policy data, and build a policy knowledge base based on the comprehensive policy data; The graph processing module is used to update and replace the policy knowledge base, obtain the collaborative word segmentation knowledge base, perform entity recognition on the collaborative word segmentation knowledge base, obtain collaborative knowledge entities, perform relationship matching and attribute extraction on the collaborative knowledge entities, and perform graph-text construction with the collaborative knowledge entities to obtain a policy information graph; The intelligent analysis module is used to perform flow segment conversion on the policy information graph to obtain the policy data stream, intercept the code element of the policy data stream to obtain the policy data code segment, obtain the change net value group by performing intra-graph grouping and clustering changes on the policy data code segment, and combine the information of the change net value group to obtain the graph extraction group set; The key matching module is used to unify the form of the demand information data, obtain the demand data code segment, perform code segment matching and matching verification on the demand data code segment through the graph extraction group set, and obtain the push policy set.

2. The innovative service policy intelligent matching system based on data analysis according to claim 1 is characterized in that: The process of the semantic collection module collecting demand information data and comprehensive policy data includes: Set up a multilateral capture terminal and a user collection terminal, collect the target user's demand information data through the user collection terminal, and collect data from the policy producer through the multilateral capture terminal to obtain comprehensive policy data; Perform data cleansing on the comprehensive policy data to obtain the net value policy data, combine and match the net value policy data to obtain policy information pairs, and build a policy knowledge base based on the policy information pairs.

3. The innovative service policy intelligent matching system based on data analysis according to claim 2 is characterized in that: The process of obtaining the policy information map includes: Perform standard coordination on the obtained policy knowledge base to obtain net value policy segmentation words, update and replace the policy knowledge base based on the net value policy segmentation words, and obtain a coordinated segmentation knowledge base; Perform entity recognition on the obtained collaborative word segmentation knowledge base to obtain collaborative knowledge entities, perform relationship matching on the collaborative knowledge entities based on the collaborative word segmentation knowledge base to obtain entity docking relationships; Based on the collaborative word segmentation knowledge base, attributes of collaborative knowledge entities are extracted to obtain entity attribute words. Based on collaborative knowledge entities, entity docking relationships and entity attribute words, graphs and texts are constructed to obtain a policy information graph.

4. The innovative service policy intelligent matching system based on data analysis according to claim 3 is characterized in that: The process of constructing graphics and texts based on collaborative knowledge entities, entity docking relationships, and entity attribute words includes: Set up a construction space, upload the collaborative knowledge entity to the construction space, perform graphical conversion on the collaborative knowledge entity through the construction space, and obtain collaborative entity nodes; The collaborative entity nodes are linked according to the entity docking relationship to obtain the node docking edges, and the node docking edges are marked and paired based on the entity attribute words to obtain the policy information map.

5. The innovative service policy intelligent matching system based on data analysis according to claim 4 is characterized in that: The process of obtaining the atlas extraction set includes: The collaborative entity nodes, node docking edges and entity attribute words are marked as pre-processed data, and the pre-processed data is converted into flow segments to obtain the policy data flow; Upload the policy data stream to the construction space, intercept the code element of the policy data stream, and obtain the policy data code segment; According to the policy information graph, the policy data code segments are grouped in the graph to obtain a net value data group, and the net value data group is clustered and changed to obtain a changed net value group; The information of the net value change group is combined according to the policy information graph to obtain the graph extraction group set.

6. The innovative service policy intelligent matching system based on data analysis according to claim 5 is characterized in that: The process of combining information of the net value change group according to the policy information map includes: Perform association selection on the obtained net value change group to obtain the node edge attribute group; The direction group distance is set, and the obtained node edge attribute groups are extracted and arranged according to the direction group distance to obtain a graph extraction group set.

7. The innovative service policy intelligent matching system based on data analysis according to claim 6 is characterized in that: The process of obtaining a push policy set includes: Unify the form of demand information data to obtain demand data code segments; The acquired demand data code segments are uploaded to the construction space, and the demand data code segments are matched by the graph extraction group set to obtain the matching policy code group; A matching check is performed on the obtained matching policy code group to obtain a push policy set.

8. The innovative service policy intelligent matching system based on data analysis according to claim 7 is characterized in that: The process of matching the required data code segments by extracting the set of graphs includes: Coordinate and match the obtained demand data code segment with the change net value group in the spectrum extraction group set until all the change net value groups in the spectrum extraction group set are coordinated and matched with the demand data code segment to obtain the intra-group matching code; Perform state decomposition on the matching code within the group to obtain balanced code and unbalanced code; Update the code state of the matching code within the group according to the balance code and the imbalance code to obtain an update result code; The update result codes are divided and combined based on the graph extraction group set to obtain the matching policy code group.

9. The intelligent matching method of the innovative service policy intelligent matching system based on data analysis according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Collect demand information data and comprehensive policy data, and build a policy knowledge base based on the comprehensive policy data; Step 2: Update and replace the policy knowledge base to obtain the collaborative word segmentation knowledge base, perform entity recognition on the collaborative word segmentation knowledge base to obtain collaborative knowledge entities, perform relationship matching and attribute extraction on the collaborative knowledge entities, and perform graph-text construction with the collaborative knowledge entities to obtain the policy information graph; Step 3: Perform stream segment conversion on the policy information graph to obtain the policy data stream, perform code element interception on the policy data stream to obtain the policy data code segment, obtain the change net value group by performing intra-graph grouping and clustering changes on the policy data code segment, perform information combination on the change net value group, and obtain the graph extraction group set; Step 4: Unify the form of the demand information data to obtain the demand data code segment, perform code segment matching and matching verification on the demand data code segment through the graph extraction group set, and obtain the push policy set.

Images