Big data-based charity information management system and method

Abstract

The invention relates to the technical field of charity information management, in particular to a charity information management system and method based on big data. The system has the advantages that the demand end and the release end are classified, labeled and stored in the database through the big data server to obtain a first data result, and the big data algorithm processing module performs semantic analysis on the first data result stored in the database through a semantic analysis algorithm; the big data algorithm processing module carries out mutual matching on the second data results after analysis and assignment of the demand end and the release end according to semantic values; and the big data server sends the third data result to the charity payment publishing end and the charity demand end, the semantics of the published information is analyzed and assigned through an algorithm, and then publishers of the assigned and matched information are matched with each other through a semantic classification algorithm.

Description

Technical Field

[0001] The present invention relates to the technical field of charity information management, and in particular to a charity information management system and method based on big data. Background Art

[0002] The charity cause is an act of people voluntarily dedicating love and assistance and a social cause engaged in helping the weak and the poor. The objects, scope, standards, and projects of the charity cause are determined by the donors.

[0003] Currently, with the progress of big data technology, more and more charity management begins to use digital systems to manage charity business information. However, the current charity information management system still has the following problems:

[0004] The current charity information management system has a simple system structure. It mostly only lists the information of charity demanders and charity initiators, and mostly cannot perform corresponding matching according to actual charity needs and charity giving conditions. Therefore, it often causes the charity contributions given by the initiators to not match the demanders, resulting in a waste of charity resources. Moreover, it has no ability to screen and distinguish many charity needs, and situations such as false donation fraud often occur. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a charity information management system and method based on big data, effectively solving the deficiencies of the prior art.

[0006] To achieve the above purpose, an embodiment of one aspect of the present invention provides a charity information management system based on big data, including a database for storing charity data information, a big data server for processing charity data, a big data algorithm processing module, a user terminal, and an identity authentication module; the user terminal includes a charity giving release end and a charity demand end;

[0007] The big data server has each data node, the user terminal is interconnected with the big data server, and the big data server is used to process the release information of the user terminal and classify and label it according to the demand end and the release end and store it in the database to obtain a first data result;

[0008] The big data algorithm processing module has a semantic parsing algorithm. The big data algorithm processing module performs semantic parsing on the first data result stored in the database through the semantic parsing algorithm and assigns the parsed information based on semantics to obtain a second data result;

[0009] The big data algorithm processing module also has a semantic classification algorithm. The big data algorithm processing module matches the second data results after parsing and assignment at the demand side and the release side according to semantic values through the semantic classification algorithm, and uploads the matching information to the big data server to obtain a third data result;

[0010] The big data server sends the third data result to the charity giving release side and the charity demand side respectively; the identity verification module is used to verify the identity information of the charity demand side.

[0011] Preferably, according to any of the above solutions, the big data server includes a login module for the charity giving release side and the charity demand side, a data release module for the respective information of the charity giving release side and the charity demand side, a data automatic annotation module for classifying and labeling the charity information data categories of the charity giving release side and the charity demand side, and a two-end interactive communication module.

[0012] Preferably, according to any of the above solutions, the two-end interactive communication module is used for the mutual communication after matching between the charity giving release side and the charity demand side. The two-end interactive communication module conducts data interaction with the identity verification module, and the identity verification module has several identity authentication units.

[0013] Preferably, according to any of the above solutions, the big data algorithm processing module is a storage module embedded inside the big data server, and both the semantic parsing algorithm and the semantic classification algorithm are stored in the storage module for the big data server to call.

[0014] Preferably, according to any of the above solutions, the semantic parsing algorithm includes a word semantic parsing method, and the word semantic parsing method is based on the degree to which two words can be used interchangeably in different contexts without changing the syntactic and semantic structure of the text;

[0015] The greater the possibility that the semantic parsing algorithm can be replaced with each other in different contexts without changing the syntactic and semantic structure of the text, the higher the similarity between the two, otherwise the similarity is lower.

[0016] The similarity of the French semantic structure is represented by a numerical value, and generally the value range is between [0,1]. The semantic similarity of a word with itself is 1; if two words cannot be replaced in any context, then their similarity is 0. Among them, the semantic of the word has the greatest influence on the word similarity. The relationship between the word distance and the word similarity is as follows:

[0017] When the distance between two words is 0, their similarity is 1, that is, the distance of a word from itself is 0;

[0018] When the distance between two words is infinite, their similarity is 0;

[0019] The greater the distance between two words, the smaller their similarity;

[0020] If the word similarity is denoted as sim(w 1 , w 2 ), and the word distance is denoted as dis(w 1 , w 2 ), the following formula is obtained:

[0021]

[0022] where ∝ is an adjustable parameter representing the word distance value when the similarity is 0.5.

[0023] Preferably, according to any of the above solutions, the semantic parsing algorithm further includes sentence semantic parsing, and the sentence semantic parsing includes shallow semantic analysis and deep semantic analysis;

[0024] The sentence semantic parsing adopts case grammar, and the case grammar includes deep cases. The deep case is the transitive relationship between the noun and the predicate in the sentence, and the deep case is determined by the syntactic and semantic relationship between the noun and the verb in the underlying structure. Regardless of how the surface syntactic structure changes, the underlying case grammar remains unchanged; the underlying "case" and the grammatical concepts on the surface structure in any specific language.

[0025] Preferably, according to any of the above solutions, the big data algorithm processing module uses a semantic role labeling algorithm during annotation. The semantic role labeling needs to rely on the results of syntactic analysis. The semantic role labeling algorithm includes: semantic role labeling based on complete syntactic analysis, semantic role labeling based on local syntactic analysis, and semantic role labeling based on dependency syntactic analysis.

[0026] Preferably, according to any of the above solutions, the identity verification module includes an identity real-name verification unit, a family situation verification unit, a physical condition verification unit, and a question-and-answer interaction unit. The identity real-name verification unit interacts with the identity card information big data platform, the family situation verification unit interacts with the household register information big data platform, the physical condition verification unit interacts with the hospital big data platform, and the question-and-answer interaction unit is custom-set by the charity giving release end.

[0027] To achieve the above object, an embodiment of one aspect of the present invention provides a charity information management method based on big data, including the following steps:

[0028] 1), The charity giving release end and the charity demand end log in to the charity information management system based on big data;

[0029] 2), users of the charity giving release end and the charity demand end respectively conduct identity authentication through the identity real-name authentication unit, family situation authentication unit, and physical condition authentication unit of the identity authentication module;

[0030] 3), users of the charity giving release end and the charity demand end respectively release charity information and help-seeking information through the big data server based on the identity authentication information;

[0031] 4), the big data server classifies and labels the released charity information and help-seeking information and stores them in the database to obtain the first data result;

[0032] 5), the big data algorithm processing module performs semantic analysis on the first data result stored in the database through the semantic parsing algorithm, and assigns the semantic-based parsing information to obtain the second data result;

[0033] 6), the big data algorithm processing module mutually matches the second data results of the demand end and the release end after parsing and assignment according to the semantic values, and uploads the matching information to the big data server to obtain the third data result;

[0034] 7), the big data server sends the third data result to the charity giving release end and the charity demand end respectively;

[0035] 8), the charity giving release end sends a question request to the charity demand end through the question-and-answer interaction unit of the identity authentication module according to the matching information, the charity demand end gives feedback according to the question sent by the question-and-answer interaction unit, and the charity giving release end can choose whether to match according to the feedback.

[0036] The present invention has the following advantages:

[0037] 1. For the big data-based charity information management system and method, through the big data server, the demand end and the release end are classified and labeled and stored in the database to obtain the first data result, the big data algorithm processing module performs semantic analysis on the first data result stored in the database through the semantic parsing algorithm, and assigns the semantic-based parsing information to obtain the second data result, the big data algorithm processing module mutually matches the second data results of the demand end and the release end after parsing and assignment according to the semantic values, and uploads the matching information to the big data server to obtain the third data result, the big data server sends the third data result to the charity giving release end and the charity demand end respectively. By parsing and assigning the semantics of the released information through the algorithm, and then the semantic classification algorithm mutually matches the publishers of the assigned and matched information, so that the charity help needed by the demand end matches the help given by the release end, reducing the waste of charity materials and improving the resource matching utilization rate.

[0038] 2. The charity information management system and method based on big data can effectively verify the identity information, family information, and physical condition information of the personnel on the user terminal through the identity verification module to prevent fraud and deception in donations. At the same time, through the Q&A interaction unit, users on the charity information publishing side can customize questions to confirm whether the requesters meet their funding expectations, so that the charity providers can make the help they provide meet their expectations. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic structural diagram of the system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The following further describes the present invention with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the following. As Figure 1 shown, a charity information management system and method based on big data includes a database for storing charity data information, a big data server for processing charity data, a big data algorithm processing module, a user terminal, and an identity verification module; the user terminal includes a charity provider publishing side and a charity requester side;

[0042] The big data server has each data node, the user terminal is interconnected with the big data server, and the big data server is used to process the publishing information of the user terminal and store it in the database classified and labeled according to the requester side and the publishing side to obtain a first data result;

[0043] The big data algorithm processing module has a semantic parsing algorithm. The big data algorithm processing module performs semantic parsing on the first data result stored in the database through the semantic parsing algorithm and assigns the parsed information based on semantics to obtain a second data result;

[0044] The big data algorithm processing module also has a semantic classification algorithm. The big data algorithm processing module matches the second data results of the requester side and the publishing side after parsing and assignment according to the semantic values through the semantic classification algorithm, and uploads the matching information to the big data server to obtain a third data result;

[0045] The big data server sends the third data result to the charity provider publishing side and the charity requester side respectively; the identity verification module is used to verify the identity information of the charity requester side.

[0046] As an alternative technical solution of the present invention, the big data server includes a login module for the charity giving release end and the charity demand end, a data release module for the respective information of the charity giving release end and the charity demand end, a data automatic annotation module for classifying and labeling the charity information data categories of the charity giving release end and the charity demand end, and a two-end interactive communication module.

[0047] As an alternative technical solution of the present invention, the two-end interactive communication module is used for the mutual communication after the matching of the charity giving release end and the charity demand end. The two-end interactive communication module performs data interaction with the identity verification module, and the identity verification module includes several identity authentication units.

[0048] As an alternative technical solution of the present invention, the big data algorithm processing module is an embedded storage module provided inside the big data server. The semantic parsing algorithm and the semantic classification algorithm are both stored in the storage module for the big data server to call.

[0049] As an alternative technical solution of the present invention, the semantic parsing algorithm includes the word semantic parsing method. The word semantic parsing method is based on the degree to which two words can be used interchangeably in different contexts without changing the syntactic and semantic structure of the text.

[0050] The greater the possibility that the semantic parsing algorithm can be replaced with each other in different contexts without changing the syntactic and semantic structure of the text, the higher the similarity between the two, otherwise the similarity is lower.

[0051] The similarity of the French semantic structure is represented by a numerical value, generally in the range of [0, 1]. The semantic similarity of a word with itself is 1; if two words cannot be replaced in any context, then their similarity is 0. Among them, the semantic of the word has the greatest influence on the word similarity. The relationship between the word distance and the word similarity is as follows:

[0052] When the distance between two words is 0, their similarity is 1, that is, the distance of a word from itself is 0.

[0053] When the distance between two words is infinite, their similarity is 0.

[0054] The greater the distance between two words, the smaller their similarity.

[0055] If the word similarity is denoted as sim(w 1 , w 2 ), and the word distance is denoted as dis(w 1 , w 2 ), the following formula is obtained:

[0056]

[0057] Among them, ∝ is an adjustable parameter representing the word distance value when the similarity is 0.5.

[0058] As an alternative technical solution of the present invention, the semantic parsing algorithm further includes sentence semantic parsing, and the sentence semantic parsing includes shallow semantic analysis and deep semantic analysis;

[0059] The sentence semantic parsing adopts case grammar, and the case grammar includes deep cases. The deep cases are the transitive relationships between the nominal words and the predicates in the sentence, and the deep cases are determined by the syntactic and semantic relationships between nouns and verbs in the underlying structure. Regardless of how the surface syntactic structure changes, the underlying case grammar remains unchanged; the underlying "cases" are grammatical concepts on the surface structure in any specific language.

[0060] As an alternative technical solution of the present invention, the big data algorithm processing module uses a semantic role labeling algorithm during annotation. Semantic role labeling needs to rely on the results of syntactic analysis. The semantic role labeling algorithm includes: semantic role labeling based on complete syntactic analysis, semantic role labeling based on local syntactic analysis, and semantic role labeling based on dependency syntactic analysis, so as to analyze and identify sentences.

[0061] As an alternative technical solution of the present invention, the identity verification module includes an identity real-name verification unit, a family situation verification unit, a physical condition verification unit, and a question-and-answer interaction unit. The identity real-name verification unit interacts with the ID card information big data platform, the family situation verification unit interacts with the household register information big data platform, the physical condition verification unit interacts with the hospital big data platform, and the question-and-answer interaction unit is custom-set by the charity giving release end. Through the question-and-answer interaction unit, users of the charity information release end can customize questions to confirm whether the requester meets their funding expectations, so that the charity givers can make the help they provide meet their expectations.

[0062] A charity information management method based on big data includes the following steps:

[0063] 1), The charity giving release end and the charity demand end log in to the charity information management system based on big data;

[0064] 2), Users of the charity giving release end and the charity demand end respectively conduct identity verification through the identity real-name verification unit, the family situation verification unit, and the physical condition verification unit of the identity verification module;

[0065] 3), Users of the charity giving release end and the charity demand end respectively publish charity information and help-seeking information through the big data server based on the identity verification information;

[0066] 4) The big data server classifies and labels the published charity information and help requests and stores them in the database to obtain a first data result;

[0067] 5) The big data algorithm processing module performs semantic analysis on the first data result stored in the database through a semantic parsing algorithm, and assigns semantic-based parsing information to obtain a second data result;

[0068] 6) The big data algorithm processing module mutually matches the second data results after parsing and assignment at the demand side and the publishing side according to the semantic values, and uploads the matching information to the big data server to obtain a third data result;

[0069] 7) The big data server sends the third data result to the charity giving publishing side and the charity demand side respectively;

[0070] 8) The charity giving publishing side sends a question request to the charity demand side through the Q&A interaction unit of the identity verification module according to the matching information. The charity demand side gives feedback according to the question sent by the Q&A interaction unit, and the charity giving publishing side can choose whether to match according to the feedback.

[0071] In summary, through the big data server, the demand side and the publishing side are classified and labeled and stored in the database to obtain a first data result. The big data algorithm processing module performs semantic analysis on the first data result stored in the database through a semantic parsing algorithm, and assigns semantic-based parsing information to obtain a second data result. The big data algorithm processing module mutually matches the second data results after parsing and assignment at the demand side and the publishing side according to the semantic values, and uploads the matching information to the big data server to obtain a third data result. The big data server sends the third data result to the charity giving publishing side and the charity demand side respectively. By parsing and assigning the semantics of the published information through an algorithm, and then the semantic classification algorithm matches the publishers of the assigned and matched information with each other, so that the charity help needed by the demand side matches the help given by the publishing side, reducing the waste of charity materials and improving the resource matching utilization rate. Through the identity verification module, the identity information, family information and physical condition information of the user terminal personnel can be effectively verified to prevent people from fraudulently donating or scamming. At the same time, through the Q&A interaction unit, the charity information publisher can customize questions to confirm whether the applicant meets their funding expectations, so that the charity giver can make the help provided meet their expectations.

[0072] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A charity information management system based on big data, characterized by: It includes a database for storing charity data information, a big data server for processing charity data, a big data algorithm processing module, a user terminal and an identity authentication module; the user terminal includes a charity payment issuing end and a charity demand end; The big data server has various data nodes, the user terminal is interconnected with the big data server, and the big data server is used to process the publishing information of the user terminal, and store it in the database according to the classification and labeling of the demand end and the publishing end to obtain the first data result; The big data algorithm processing module has a semantic parsing algorithm, and the big data algorithm processing module performs semantic parsing on the first data result stored in the database through the semantic parsing algorithm, and assigns parsing information based on the semantics to obtain the second data result; The big data algorithm processing module also has a semantic classification algorithm, which matches the second data results after the demand side and the publishing side parse and assign values ​​according to the semantic value through the semantic classification algorithm, and uploads the matching information to the big data server to obtain the third data result; The big data server sends the third data result to the charity payment issuing end and the charity demand end respectively; the identity authentication module is used to verify the identity information of the charity demand end.

2. The charity information management system based on big data according to claim 1 is characterized by: The big data server includes login modules for the charity payment publishing end and the charity demand end, a data publishing module for the respective information of the charity payment publishing end and the charity demand end, an automatic data labeling module for classifying and labeling charity information data categories of the charity payment publishing end and the charity demand end, and a two-end interactive communication module.

3. The charity information management system based on big data according to claim 2 is characterized by: The two-end interactive communication module is used for mutual communication between the charity payment issuing end and the charity demand end after matching. The two-end interactive communication module exchanges data with the identity authentication module, and the identity authentication module has several identity authentication units.

4. The charity information management system and method based on big data according to claim 3 is characterized by: The big data algorithm processing module is a storage module embedded in the big data server. The semantic parsing algorithm and the semantic classification algorithm are both stored in the storage module for the big data server to call.

5. The charity information management system based on big data according to claim 4 is characterized by: The semantic parsing algorithm includes a word semantic parsing method, which is based on the degree to which two words can be used interchangeably in different contexts without changing the syntactic and semantic structure of the text; the greater the possibility that the semantic parsing algorithm can replace each other in different contexts without changing the syntactic and semantic structure of the text, the higher the similarity between the two, otherwise the similarity is lower. The semantic structure similarity is expressed as a numerical value, generally in the range of [0,1]. The semantic similarity between a word and itself is 1; if two words cannot be replaced in any context, then their similarity is 0. The semantics of the word should have the greatest impact on the similarity of the words. The relationship between the distance between the words and the similarity of the words is as follows: When the distance between two words is 0, their similarity is 1, that is, the distance between a word and itself is 0; When the distance between two words is infinite, their similarity is 0; When the distance between two words is greater, their similarity is smaller; If the word similarity is recorded as sim(w1,w2) and the word distance is recorded as dis(w1,w2), the following formula is obtained: Among them, ∝ is an adjustable parameter, which represents the word distance value when the similarity is 0.

5.

6. The charity information management system based on big data according to claim 5 is characterized by: The semantic parsing algorithm also includes sentence semantic parsing, and the sentence semantic parsing includes shallow semantic analysis and deep semantic analysis; The sentence semantic analysis adopts case grammar, which includes deep case. The deep case is the transitivity relationship between the subject and the predicate in the sentence. The deep case is determined by the syntactic semantic relationship between nouns and verbs in the underlying structure. No matter how the surface syntactic structure changes, the underlying case grammar remains unchanged; the underlying "case" is the grammatical concept on the surface structure in any specific language.

7. The charity information management system based on big data according to claim 6 is characterized by: The big data algorithm processing module adopts a semantic role labeling algorithm when labeling. The semantic role labeling needs to rely on the results of syntactic analysis. The semantic role labeling algorithm includes: semantic role labeling based on complete syntactic analysis, semantic role labeling based on local syntactic analysis, and semantic role labeling based on dependency syntactic analysis.

8. The charity information management system based on big data according to claim 7 is characterized by: The identity authentication module includes a real-name identity authentication unit, a family situation authentication unit, a physical condition authentication unit and a question-and-answer interactive unit. The real-name identity authentication unit interacts with the identity card information big data platform, the family situation authentication unit interacts with the household registration information big data platform, the physical condition authentication unit interacts with the hospital big data platform, and the question-and-answer interactive unit is customized by the charity payment publishing end.

9. A charity information management method based on big data, used in the charity information management system based on big data according to any one of claims 1 to 8, characterized in that The following steps are involved: 1) The charity payment issuing end and the charity demand end log in to the charity information management system based on big data as described in any one of claims 1 to 8; 2) Users at the charity payment issuing end and charity demand end respectively authenticate their identities through the identity real-name authentication unit, family status authentication unit and physical status authentication form of the identity authentication module; 3) Users at the charity payment publishing end and charity demand end publish charity information and help-seeking information through the big data server based on identity authentication information; 4) The big data server classifies and labels the charity information and help-seeking information released and stores them in the database to obtain the first data result; 5) The big data algorithm processing module performs semantic analysis on the first data result stored in the database through a semantic analysis algorithm, and assigns the semantic-based analysis information to obtain a second data result; 6) The big data algorithm processing module matches the second data results after the demand side and the publishing side parse and assign values ​​according to the semantic value, and uploads the matching information to the big data server to obtain the third data result; 7) The big data server sends the third data result to the charity payment issuing end and the charity demand end respectively; 8) The charity payment issuing end sends a question request to the charity demand end through the question and answer interactive unit of the identity authentication module according to the matching information. The charity demand end gives feedback according to the questions sent by the question and answer interactive unit, and the charity payment issuing end can choose whether to match according to the feedback.

Images