CN111382279A

CN111382279A - Order examination method and device

Info

Publication number: CN111382279A
Application number: CN202010152518.2A
Authority: CN
Inventors: 卢时云; 雷鸣; 李力; 王国悦; 饶帆; 陆佳庆
Original assignee: China Construction Bank Corp; CCB Finetech Co Ltd
Current assignee: China Construction Bank Corp
Priority date: 2020-03-06
Filing date: 2020-03-06
Publication date: 2020-07-07

Abstract

The invention discloses a document examination method and a document examination device, and relates to the technical field of computers. The method comprises the following steps: after receiving an audit request, acquiring all bill element data corresponding to a service number carried by the audit request; establishing bill auditing knowledge map example data according to the bill element data; and traversing and executing each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result. Through the steps, the work efficiency of bill auditing can be improved, the time consumption of bill auditing is reduced, and the accuracy of bill auditing results is improved.

Description

Order examination method and device

Technical Field

The invention relates to the technical field of computers, in particular to a document examination method and a document examination device.

Background

In financial institutions such as banks, auditing bills is a very common business node. For example, in the documentary collection business, a collection bank needs to review paper documents submitted by customers and export a collection order according to international conventions and practice approval rules on the principle of 'document conformity'.

In the prior art, bill auditing work is basically finished by an examination expert on line. For example, for the receipt following and collection entrusting business, the order examination expert can initiate an order entering step in a system for processing international settlement related business, enter bank names, addresses, bill amounts, currencies, due dates and transportation related information, manually perform related examination on paper bills, and manually enter a small part of examination results into the system after the examination is completed.

In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art: firstly, the existing bill auditing work is completely finished manually, and the manual cost occupied by the bill auditing work is higher due to the fact that the bills are various and different in style; and secondly, the examination process is completed on line completely based on the experience of the examination experts, the time consumption and the quality of the examination are low, and the examination is also completely dependent on the proficiency of the examination experts on the business, so that the examination efficiency and the accuracy of the examination result are uncontrollable, and the customer satisfaction is low.

Disclosure of Invention

In view of this, the invention provides a document examination method and device, which can improve the work efficiency of document examination and verification, reduce the time consumption of document examination and improve the accuracy of document examination and verification results.

To achieve the above object, according to one aspect of the present invention, there is provided an order examination method.

The order examination method comprises the following steps: after receiving an audit request, acquiring all bill element data corresponding to a service number carried by the audit request; establishing bill auditing knowledge map example data according to the bill element data; and traversing and executing each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result.

Optionally, the method further comprises: before the step of constructing bill verification knowledge map instance data according to the bill element data, inquiring a map database according to a service number carried by the verification request; and when the query result shows that historical bill auditing knowledge map instance data corresponding to the service number exist in the map database, clearing the historical bill auditing knowledge map instance data.

Optionally, the constructing of the ticket audit knowledge graph instance data according to the ticket element data includes: assembling the element data under each bill and a preset bill body to obtain bill knowledge map example data; and constructing bill auditing knowledge graph instance data according to the bill knowledge graph instance data and a preset rule body.

Optionally, the constructing of the ticket audit knowledge graph instance data according to the ticket element data includes: preprocessing the bill element data according to a basic knowledge ontology; the basic ontology comprises at least one of: a company ontology, a country ontology, and a bank ontology.

Optionally, the obtaining of all the ticket element data corresponding to the service number carried by the audit request includes: and querying a relational database according to the service number carried by the audit request so as to query all the bill element data corresponding to the service number.

Optionally, the method further comprises: before querying a relational database according to the service number carried by the audit request, extracting all bill element data in a bill file by an optical image recognition technology and a natural language processing technology, and storing the bill element data in the relational database.

Optionally, the audit request is an audit request of the documentary collection service.

To achieve the above object, according to another aspect of the present invention, there is provided an examination apparatus.

The document examination device of the present invention includes: the acquisition module is used for acquiring all bill element data corresponding to the service number carried by the audit request after receiving the audit request; the construction module is used for constructing bill verification knowledge map example data according to the bill element data; and the auditing module is used for traversing and executing each auditing rule in the bill auditing knowledge graph example data to obtain an auditing result.

To achieve the above object, according to still another aspect of the present invention, there is provided an electronic apparatus.

The electronic device of the present invention includes: one or more processors; and storage means for storing one or more programs; when executed by the one or more processors, cause the one or more processors to implement the ticketing method of the present invention.

To achieve the above object, according to still another aspect of the present invention, there is provided a computer-readable medium.

The computer-readable medium of the invention has stored thereon a computer program which, when executed by a processor, implements the ticketing method of the invention.

One embodiment of the above invention has the following advantages or benefits: the method comprises the steps of obtaining all bill element data corresponding to the service number carried by the audit request after receiving the audit request, constructing bill audit knowledge map example data according to the bill element data, traversing and executing each audit rule in the bill audit knowledge map example data to obtain an audit result, and can improve the work efficiency of bill audit, reduce the time consumption of bill audit and improve the accuracy of the bill audit result.

Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.

Drawings

The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:

fig. 1 is a main flow diagram of an examination method according to a first embodiment of the present invention;

FIG. 2 is a schematic main flow diagram of a document examination method according to a second embodiment of the present invention;

FIG. 3 is a schematic diagram of the structure of example data of a document knowledge graph constructed according to an embodiment of the invention;

FIG. 4 is a block diagram of example document audit knowledge graph data constructed in accordance with an embodiment of the invention;

FIG. 5 is a schematic diagram of the main blocks of a ticket reviewing apparatus according to a third embodiment of the present invention;

FIG. 6 is a schematic diagram of the main blocks of a ticket reviewing apparatus according to a fourth embodiment of the present invention;

FIG. 7 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;

FIG. 8 is a schematic block diagram of a computer system suitable for use with the electronic device to implement an embodiment of the invention.

Detailed Description

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

It should be noted that the individual features of the embodiments of the invention can be combined with each other without affecting the implementation of the invention.

Before describing embodiments of the present invention in detail, some technical terms related to the embodiments of the present invention will be described.

Knowledge graph: in narrow terms, the knowledge graph was first proposed by Google corporation and used by internet companies to semantically organize network data to provide a large knowledge base of intelligent search services. Formally, it is a knowledge carrier represented by a graph data structure, describing objects of the objective world and their relationships, wherein nodes represent objects of the objective world and edges represent relationships between objects. In the concrete implementation, the knowledge graph spectrum uses a resource description framework in a semantic network to uniformly represent the contents of two levels of a knowledge system and example data, and a complete knowledge system is formed together. By extension, knowledge descriptions, instance data and related matching standards, technologies, tools and application systems form a generalized knowledge graph.

A body: the knowledge graph stores structured knowledge, and data such as entities, attributes, relations and the like in the knowledge graph have clear semantics. In fact, a knowledge graph not only contains specific instance knowledge data, but also contains descriptions and definitions of knowledge data, and the "meta" data describing and defining the data is called a knowledge hierarchy or ontology. The knowledge graph represents ontology and instance data in a unified triple format. The ontology is typically stored in the schema layer of the knowledge graph, and the specific instance knowledge data is stored in the data layer.

Fig. 1 is a main flow diagram of an examination method according to a first embodiment of the present invention. As shown in fig. 1, the order examination method according to the embodiment of the present invention includes:

step S101, after receiving an audit request, acquiring all bill element data corresponding to a service number carried by the audit request.

Illustratively, in the documentary collection service scenario, the audit request is a documentary collection service audit request. The receipt collection is one of the most common settlement modes in international settlement services. In the documentary collection business, after the exporter delivers the stock according to the contract, the related documentary is signed to the collection bank, and the collection bank transfers the documentary to the collection bank to collect the payment of the goods by the collection bank. Although the collection bank and the collection bank have no obligation to check the document according to the contract, the collection bank or the collection bank receives the corresponding examination business under the entrepreneur and the exporter. It should be noted that the method of the embodiment of the present invention is not only suitable for a documentary collection service scenario, but also suitable for other service scenarios requiring bill auditing.

After receiving the audit request, the audit device can analyze the audit request to obtain the service number carried by the audit request. Next, the document examination apparatus may acquire all the document element data corresponding to the service number.

In an alternative embodiment, the ticket element data is stored in a relational database. In this optional embodiment, the relational database may be queried according to the service number carried in the audit request, so as to query all the ticket element data under the service. In the implementation, before a business (such as a documentary collection business) enters an auditing link, all bill element data in a bill file can be extracted through an OCR (optical image recognition) technology and an NLP (natural language processing) technology, and the bill element data is stored in a relational database in a structured form. In addition, preferably, the granularity of the extraction of the bill elements is consistent with the granularity of the attributes defined by the bill body.

And S102, establishing bill auditing knowledge graph example data according to the bill element data.

In an optional example, step S102 may specifically include: assembling the element data under each bill and a preset bill body to obtain bill knowledge map example data; and constructing bill auditing knowledge graph instance data according to the bill knowledge graph instance data and a preset rule body. The bill body refers to a bill body related to the examination bill. For example, in a documentary collection service audit scene, the bill body includes the body of bills such as draft, invoice, facial letter, bill of lading, and the like. The rule ontology refers to an ontology of an audit rule related to the examination order. For example, in a documentary collection business audit scenario, the rule ontology includes an ontology of an indelible item audit rule and an element consistency audit rule. The ontology of the element consistency audit rule can be divided into an ontology of an amount consistency audit rule and an applicant consistency audit rule.

For example, suppose that the business number of a particular documentary collection business is "TS 000122", the business has only two bills, namely, an Invoice (Invoint) and a Draft (Draft). After acquiring the element data of all the bills under the service number "TS 000122", the element data of the bills and the bill body can be assembled, the element data of the draft and the bill body can be assembled, and the bill assembly result are associated through the same service vertex (which may be called as "transaction vertex", "service entity vertex" or "transaction entity vertex"), so that the bill knowledge map example data corresponding to the service number "TS 000122" can be obtained. And then, establishing bill auditing knowledge graph instance data according to the bill knowledge graph instance data and a preset rule body. For example, when the amount of a money order and the amount of an invoice appear in the bill knowledge map example data, instantiation of an amount consistency auditing rule is triggered, namely, the bill knowledge map example data and an amount consistency auditing rule body are assembled; when the invoice exporter data and the bill drawer data appear in the bill knowledge map example data, instantiation of the beneficiary consistency audit rule is triggered, namely the bill knowledge map example data and the body of the beneficiary consistency audit rule are assembled.

And S103, traversing and executing each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result.

Wherein the data audit knowledge graph instance data is stored in the form of a graph structure. In this step, each rule entity vertex in the ticket audit knowledge-graph instance data may be traversed to execute each audit rule. For example, when traversing to an amount checking vertex, all edges pointing to the vertex and attributes of the edges are obtained, and by parsing the contents of the attributes, all attributes participating in the amount checking can be obtained, and assuming that the attribute values are invoke. Figure 67776.6 for invoice amount, USD for figure currency for invoice amount, 67776.6 for draft figure money for draft currency, and USD for draft figure currency for draft currency. Next, the values of invice, figure and draft, figure are compared, and the values of invice, currency and draft, currency are compared. And if the comparison result shows that the money amount and the currency are consistent, the verification result aiming at the verification rule is consistent. And if the comparison result is that the amount or currency is inconsistent, the verification result aiming at the verification rule is inconsistent. And after all the audit rules are executed, outputting a final audit result.

In the embodiment of the invention, after the audit request is received, all the bill element data corresponding to the service number carried by the audit request are obtained, the bill audit knowledge map example data is constructed according to the bill element data, and each audit rule in the bill audit knowledge map example data is traversed and executed to obtain the audit result.

Fig. 2 is a main flow diagram of an order examination method according to a second embodiment of the present invention. As shown in fig. 2, the order examination method according to the embodiment of the present invention includes:

step S201, receiving an audit request from the user terminal.

Illustratively, in the documentary collection service scenario, the audit request is a documentary collection service audit request. In specific implementation, in response to a trigger operation of a user (for example, the user clicks an "enter order" button on the user terminal), the user terminal sends an examination request of the documentary receipt acceptance service to the examination device. And then, the examination device can receive the examination request of the order following accepting service of the user terminal. It should be noted that the method of the embodiment of the present invention is not only suitable for a documentary collection service scenario, but also suitable for other service scenarios requiring bill auditing.

And step S202, analyzing the audit request to obtain the service number carried by the audit request.

Illustratively, the documentary collection service audit request may include a service number of the documentary collection service and other service information. In specific implementation, the audit request may be transmitted in the form of an HTTP request message or other request messages. After receiving the request message, the service number carried by the audit request can be obtained by analyzing the request message.

And step S203, inquiring a database according to the service number.

Illustratively, the graph database may employ Neo4j, OrientDB, tiger graph, Arangodb, or other graph databases. In a specific example, the knowledge graph of the documentary collection service is stored using Arangodb. Arangodb supports flexible data models such as Documents (Documents), graphs (Graph), and Key-Value pairs (Key-Value). The Arango db is a high-performance database, which carries SQL-like AQL query and also supports access query of programming languages such as Java and P ython in a Rest mode. In particular, the upper data storage query application does not directly interact with Arangodb, but can connect the two through Gremlin Ser ver. Gremlin is a graph traversal language under the Apache TinkerPop framework. When a data system is enabled, its user can model its domain using Gremlin. Gremlin follows the design philosophy of 'write once, run everywhere', and all tinkerPop-supporting graphics systems can perform Gremlin traversal, which means that even if we want to replace a graph database under another Apache2 framework later, the application code at the upper layer is not affected much.

When the query result of the step S203 shows that corresponding historical bill knowledge map instance data exists in the map database, the step S204 is executed next; when the query result of step S203 indicates that no corresponding historical ticket knowledge map instance data exists in the map database, step S205 is executed next.

And S204, clearing the corresponding historical bill auditing knowledge map example data.

When the historical bill auditing knowledge map example data corresponding to the service number already exists in the map database, the fact that the current auditing is not the first auditing of the service is shown. Considering that business personnel can modify the bill element values on the system page at any time, in order to ensure that the output is accurate, historical bill auditing knowledge map example data corresponding to the business number needs to be cleared first. After the historical ticket audit knowledge graph instance data is cleared, new ticket audit knowledge graph instance data can be constructed according to step S205 and step S206.

And step S205, inquiring a relational database according to the service number so as to inquire all bill element data corresponding to the service number.

In the embodiment of the invention, all the bill element data under the business is stored in the relational database. In this step, the relational database can be queried according to the service number carried by the audit request to query all the bill element data under the service. In the implementation, before a business (such as a documentary collection business) enters an auditing link, all bill element data in a bill file can be extracted through an OCR (optical image recognition) technology and an NLP (natural language processing) technology, and the bill element data is stored in a relational database in a structured form. In addition, preferably, the granularity of the extraction of the bill elements is consistent with the granularity of the attributes defined by the bill body.

And S206, establishing bill verification knowledge map example data according to the bill element data.

In an alternative example, step S206 includes: preprocessing the bill element data according to a basic knowledge ontology; assembling the element data of each bill under the service with a preset bill body to obtain bill knowledge map example data; and constructing bill auditing knowledge graph instance data according to the bill knowledge graph instance data and a preset rule body.

Wherein the basic ontology comprises at least one of: a company ontology, a country ontology, and a bank ontology. For example, for a company ontology, related concepts include company name, company address, country of the company, etc., which can be used as attributes of the company entity, so that the vertex of the company entity can establish a relationship edge of "has Property" with the concepts of the name, address, country of the company. For example, for a national ontology, the attributes of this vertex of a national entity include the name of a country, the number of a country, an abbreviation of a country, and the like.

In step S206, the preprocessing of the bill element data according to the basic ontology may include: and (5) element data alignment processing. For example, for the element data of the company type in the bill, such as the Exporter element data in the invoice and the Drawee element data in the draft, since there may be a difference in the expression of the company element data in different bills, in order to facilitate the subsequent auditing and meet the requirement of the normative storage of the data, the entity alignment processing may be performed on the company element data in the bill, such as the alignment processing is performed on the company number of the Exporter in the invoice and the company number of the Drawee in the draft. The preprocessing may also include normalization of the element data, etc., without affecting the practice of the present invention.

The bill body refers to a bill body related to the examination bill. For example, in a documentary collection service audit scene, the bill body includes the body of bills such as draft, invoice, facial letter, bill of lading, and the like. Taking an invoice as an example, the concept of an invoice is an entity. In the bill of the invoice, the numeral-related concept comprises one or more of the number of pages of the invoice, the invoice number, the making date, the capital amount number, the capital amount currency, the lower-case amount number, the lower-case amount currency, the due date and the like; the role related concepts are one or more of invoice head up, head up name, head up number, vendor name, vendor address, vendor number, vendor country, beneficiary name, beneficiary address, beneficiary number, beneficiary country, applicant name, applicant address, applicant number, applicant country, etc.; the information-related concept includes one or more of commodity information, commodity name, commodity category, commodity number, commodity specification, commodity quantity, commodity amount, and the like. Concepts such as language, invoice number, capital amount, lowercase amount, applicant, beneficiary, commodity information, etc. can be used as the primary attributes of the entity of the bill, and some concepts corresponding to the concepts can be used as the sub-attributes of the above listed concepts, for example, the sub-attribute of the capital amount is the capital amount number, the capital amount currency, the sub-attribute of the applicant is the name of the applicant, the address of the applicant, the number of the applicant, the country of the applicant, etc. Thus, a relationship edge "has Property" can be established between the invoice and the properties of the applicant, beneficiary, etc., and a relationship edge "has Property" can be established between each Property and its child Property.

The rule ontology refers to an ontology of an audit rule related to the examination order. For example, in a documentary collection business audit scenario, the rule ontology includes an ontology of an indelible item audit rule and an element consistency audit rule. The ontology of the element consistency audit rule can be divided into an ontology of an amount consistency audit rule and an applicant consistency audit rule.

In step S206, after preprocessing the element data under the service number, the element data of each ticket under the service number may be assembled with a pre-configured ticket body, and the assembled result of each ticket may be associated with the same service vertex (which may also be referred to as "transaction vertex", "service entity vertex", or "transaction entity vertex"), so as to obtain the ticket knowledgegraph instance data corresponding to the service number. And then instantiating the auditing rule according to the bill knowledge graph instance data and a preset rule body, so that bill auditing knowledge graph instance data can be obtained. Different auditing rules may be triggered because different types of tickets under different service numbers may be different and the elements of the tickets may be different. When a certain auditing Rule is triggered, the Rule type is mapped to a Rule (Rule) entity vertex of the bill auditing knowledge graph instance data, and the corresponding bill entity vertex is associated, namely a relationship edge 'has Rule' is constructed between the Rule entity vertex and the corresponding bill entity vertex, and the attribute of the edge is determined to be a certain specific bill auditing attribute.

Further, before step S206, the order examination method according to the embodiment of the present invention further includes: the knowledge graph in the field of business (such as documentary collection business) is constructed by utilizing the knowledge system construction, the knowledge representation and the graph database storage technology of the knowledge graph, and the knowledge graph comprises a bill body, a basic knowledge body and a rule body, so that related business concepts are systematized and accurately stored in a computer system.

Knowledge in the knowledge graph is represented by an RDF (resource description framework) structure, the basic constitutional units of the knowledge graph are facts, each fact is a triple (S, P, O), S is a Subject, and the value of S can be any one of an entity, an event or a concept; p refers to Predicate, whose value may be a relationship or attribute, O is Object, and whose value may be an entity, an event, a concept, or a common value such as a number, a string, or the like. Taking the knowledge graph of the documentary collection business field as an example, the triplets basically represent the fact of (entities, attributes, attribute values).

In particular implementation, the storage of the knowledge graph can be divided into storage based on a table structure and storage based on a graph structure according to different storage modes. The table structure based storage stores data in the knowledge-graph using two-dimensional data tables. In addition, the knowledge is stored in a graph mode, and deep mining and reasoning of the knowledge can be facilitated by using a related algorithm of a graph theory for reference, so that the actual storage of data in the knowledge graph in the field of services (such as a documentary collection service) adopts a graph structure-based storage mode.

And step S207, traversing and executing each auditing rule in the bill auditing knowledge-graph instance data to obtain an auditing result.

Wherein the bill audit knowledge graph instance data is stored in a form of a graph structure. In this step, each rule entity vertex in the ticket audit knowledge-graph instance data may be traversed to execute each audit rule. For example, when traversing to an amount checking vertex, all edges pointing to the vertex and attributes of the edges are obtained, and by parsing the contents of the attributes, all attributes participating in the amount checking can be obtained, and assuming that the attribute values are invoke. Figure 67776.6 for invoice amount, USD for figure currency for invoice amount, 67776.6 for draft figure money for draft currency, and USD for draft figure currency for draft currency. Next, the values of invice, figure and draft, figure are compared, and the values of invice, currency and draft, currency are compared. And if the comparison result shows that the money amount and the currency are consistent, the verification result aiming at the verification rule is consistent. And if the comparison result is that the amount or currency is inconsistent, the verification result aiming at the verification rule is inconsistent. And after all the audit rules are executed, outputting a final audit result.

And step S208, returning the auditing result to the user terminal.

In this step, besides returning the audit result to the user terminal, the document risk prompt information obtained by analysis can also be returned to the user terminal.

In the embodiment of the invention, the computerization of the examination business is realized through the steps. According to the processing flow from the step S201 to the step S208, the computer system can directly output the document auditing result, and assist the document examination experts in performing document examination, so that the processing efficiency of the document examination service is improved, the time consumption of single-stroke document examination is saved, and the document examination precision is improved.

The following describes in detail a document examination method in an embodiment of the present invention with reference to a specific example.

In this specific example, assuming that the service number of a documentary collection service is TS000123, there are only two documents under the service, namely, an Invoice (invite) and a Draft (Draft), the following element data is extracted from the Invoice by using the optical image recognition technology and the natural language processing means and according to the entities and attributes related to the Invoice and the Draft body in the knowledge map: no (invoice number, Inv01), Figure (invoice amount, 67776.6), Currency (invoice Currency, US D), exporter name (exporter name, society. ltd), exporter address (exporter address, South pungroad 1888, Shanghai), exporter country (exporter country, CN); the following elements are extracted from the draft: no (bill number, DRF01), F age (bill amount, 67776.6), Currency (bill Currency, USD), DraweeNa me (drawer name, Society Company), DraweeAddress (drawer address, South pushing Road 1888, new pushing District, Shanghai), DraweeC outltry (drawer country, CN), and stores these element data in a relational database.

After the examination device receives the examination request of the documentary collection service, the examination device acquires a service number 'TS 000123' carried by the examination request, and queries a database according to the service number so as to obtain whether historical bill examination knowledge map example data corresponding to the service number exists in the database according to the query result. In order to ensure that the auditing result is obtained based on the latest bill element, if historical bill auditing map example data corresponding to the service number exists in the map database, the historical bill auditing knowledge map example data needs to be cleared first, then the relational database is inquired according to the service number to obtain all bill element data under the service number, and new bill auditing knowledge map example data is constructed according to the bill element data.

Since the expressions for the same company in different bills are different, in order to facilitate subsequent audit consistency comparison and meet the requirement of data specification storage, in this specific example, before bill audit knowledge map instance data is constructed according to all bill element data under the same service number, entity alignment processing needs to be performed on the company element data (such as company number) in the invoice and the draft according to the company knowledge ontology in the basic knowledge ontology, and the company number returned after successful alignment is "COM 01".

In this specific example, the process of constructing the bill verification knowledge graph instance data according to all the bill element data under the same service number can be roughly divided into two steps: firstly, assembling the element data of each bill under the service with a pre-configured bill body to obtain bill knowledge map example data (as shown in figure 3); and secondly, constructing bill audit knowledge graph instance data according to the bill knowledge graph instance data and a preset rule body (as shown in figure 4).

For a single-order-following collection service, complete bill auditing knowledge map example data is constructed and completed through the two steps. And traversing each regular entity vertex in the graph to execute the regular operation. For example, when the system traverses to the vertex of the amount checking (amount consistency audit rule), the document examination device obtains all edges pointing to the vertex and attributes of the edges, and by analyzing the contents of the attributes, the document examination device can obtain all the attributes participating in the amount checking, which are respectively an invoke. The invoice amount is 67776.6, USD, draft, figure 67776.6, and USD. Next, the values of inverse and draft are compared to determine whether they are consistent. And if the comparison result shows that the money amount and the currency are consistent, the verification result aiming at the verification rule is consistent. And if the comparison result is that the amount or currency is inconsistent, the verification result aiming at the verification rule is inconsistent. And after all the audit rules are executed, outputting a final audit result.

The above specific example is a simpler examination flow example. In an actual business scene, related bills can be hundreds of sheets sometimes, and each bill has at least dozens of elements, so that in the case of huge data, the constructed bills are used for auditing the example data of the knowledge map, and the auditing result can be efficiently and accurately output.

Fig. 5 is a main block diagram of an examination apparatus according to a third embodiment of the present invention. As shown in fig. 5, the document examination apparatus 500 according to the embodiment of the present invention includes: the system comprises an acquisition module 501, a construction module 502 and an auditing module 503.

The obtaining module 501 is configured to obtain all ticket element data corresponding to the service number carried in the audit request after receiving the audit request.

After receiving the audit request, the audit device can analyze the audit request to obtain the service number carried by the audit request. Next, the document examination apparatus may acquire all the document element data corresponding to the service number through the acquisition module 501.

In an alternative embodiment, the ticket element data is stored in a relational database. In this optional embodiment, the obtaining module 501 may query the relational database according to the service number carried in the audit request, so as to query all the ticket element data in the service. In the implementation, before a business (such as a documentary collection business) enters an auditing link, all bill element data in a bill file can be extracted through an OCR (optical image recognition) technology and an NLP (natural language processing) technology, and the bill element data is stored in a relational database in a structured form. In addition, preferably, the granularity of the extraction of the bill elements is consistent with the granularity of the attributes defined by the bill body.

And the constructing module 502 is used for constructing bill auditing knowledge graph example data according to the bill element data.

In an optional example, the constructing module 502 may specifically construct the ticket audit knowledge graph instance data according to the ticket element data, including: the construction module 502 assembles the element data under each bill and a pre-configured bill body to obtain bill knowledge map instance data; the construction module 502 constructs bill audit knowledge graph instance data according to the bill knowledge graph instance data and a preconfigured rule ontology. The bill body refers to a bill body related to the examination bill. For example, in a documentary collection service audit scene, the bill body includes the body of bills such as draft, invoice, facial letter, bill of lading, and the like. The rule ontology refers to an ontology of an audit rule related to the examination order. For example, in a documentary collection business audit scenario, the rule ontology includes an ontology of an indelible item audit rule and an element consistency audit rule. The ontology of the element consistency audit rule can be divided into an ontology of an amount consistency audit rule and an applicant consistency audit rule.

For example, suppose that the business number of a particular documentary collection business is "TS 000122", the business has only two bills, namely, an Invoice (Invoint) and a Draft (Draft). After acquiring the element data of all the bills under the service number "TS 000122", the building module 502 may assemble the element data of the bills and the bill body, assemble the element data of the draft and the bill body, and associate the bill assembly result and the bill assembly result through the same service vertex (which may also be referred to as "transaction vertex", "service entity vertex", or "transaction entity vertex"), thereby obtaining bill knowledge graph instance data corresponding to the service number "TS 000122". Then, the construction module 502 constructs bill audit knowledge graph instance data according to the bill knowledge graph instance data and a preset rule ontology. For example, when the amount of a money order and the amount of an invoice appear in the bill knowledge map example data, instantiation of an amount consistency auditing rule is triggered, namely, the bill knowledge map example data and an amount consistency auditing rule body are assembled; when the invoice exporter data and the bill drawer data appear in the bill knowledge map example data, instantiation of the beneficiary consistency audit rule is triggered, namely the bill knowledge map example data and the body of the beneficiary consistency audit rule are assembled.

The auditing module 503 is configured to traverse and execute each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result.

Wherein the data audit knowledge graph instance data is stored in the form of a graph structure. In particular, audit module 503 may traverse each rule entity vertex in the ticket audit knowledge-graph instance data in order to execute each audit rule.

For example, when traversing to an amount checking vertex, all edges pointing to the vertex and attributes of the edges are obtained, and by parsing the contents of the attributes, all attributes participating in the amount checking can be obtained, and assuming that the attribute values are invoke. The invoice amount is 67776.6, USD, draft, figure 67776.6, and USD. Next, the values of inverse and draft are compared to determine whether they are consistent. And if the comparison result shows that the money amount and the currency are consistent, the verification result aiming at the verification rule is consistent. And if the comparison result is that the amount or currency is inconsistent, the verification result aiming at the verification rule is inconsistent. And after all the audit rules are executed, outputting a final audit result.

In the embodiment of the invention, after the acquisition module receives the audit request, all bill element data corresponding to the service number carried by the audit request are acquired, the construction module constructs bill audit knowledge map example data according to the bill element data, and each audit rule in the bill audit knowledge map example data is traversed and executed by the audit module to obtain the audit result, so that the work efficiency of bill audit can be improved, the time consumption of bill audit is reduced, and the accuracy of the bill audit result is improved.

Fig. 6 is a schematic view of main blocks of an examination apparatus according to a fourth embodiment of the present invention. As shown in fig. 6, the order examination apparatus 600 according to the embodiment of the present invention includes: a determination module 601, a query module 602, a clearing module 603, an acquisition module 604, a construction module 605, and an audit module 606.

The determining module 601 is configured to, after receiving the audit request, analyze the audit request to obtain a service number carried by the audit request.

Illustratively, in the documentary collection service scenario, the audit request is a documentary collection service audit request. In specific implementation, in response to a trigger operation of a user (for example, the user clicks an "enter order" button on the user terminal), the user terminal sends an examination request of the documentary receipt acceptance service to the examination device. And then, the examination device can receive the examination request of the order following accepting service of the user terminal. It should be noted that the document examination device in the embodiment of the present invention is not only suitable for document following and collection service scenarios, but also suitable for other service scenarios requiring document examination.

The documentary collection service audit request may include a service number of the documentary collection service and other service information. In specific implementation, the audit request may be transmitted in the form of an HTTP request message or other request messages. After receiving the request message, the determining module 601 analyzes the request message, so as to obtain the service number carried by the audit request.

The query module 602 is configured to query a map database according to the service number to determine whether historical ticket audit knowledge map instance data corresponding to the service number already exists in the map database.

Illustratively, the graph database may employ Neo4j, OrientDB, tiger graph, Arangodb, or other graph databases. In a specific example, the knowledge graph of the documentary collection service is stored using Arangodb. Arangodb supports flexible data models such as Documents (Documents), graphs (Graph), and Key-Value pairs (Key-Value). Arangpodb is a high-performance database with SQL-like AQL queries and also supports access queries in Java, Python, and other programming languages in the Rest mode. In particular, the data storage query application at the upper layer does not directly interact with Arangodb, but can connect the two through a Gremlin Server. Gremlin is a graph traversal language under the Apache TinkerPop framework. When a data system is enabled, its user can model its domain using Gremlin. Gremlin follows the design philosophy of 'write once, run everywhere', and all tinkerPop-supporting graphics systems can perform Gremlin traversal, which means that even if we want to replace a graph database under another Apache2 framework later, the application code at the upper layer is not affected much.

When the query result of the query module 602 indicates that corresponding historical ticket knowledge map instance data exists in the map database, the removal module 603 can be called next; when the query result of the query module 602 indicates that there is no corresponding historical ticket knowledge graph instance data in the graph database, the obtaining module 604 and the constructing module 605 may be directly invoked.

And a clearing module 603, configured to clear the corresponding historical ticket audit knowledge graph instance data.

When the historical bill auditing knowledge map example data corresponding to the service number already exists in the map database, the fact that the current auditing is not the first auditing of the service is shown. Considering that service personnel can modify the bill element values on the system page at any time, in order to ensure that the output is accurate, the historical bill auditing knowledge map example data corresponding to the service number needs to be cleared by the clearing module 603. After the historical ticket audit knowledge graph instance data is cleared, new ticket audit knowledge graph instance data can be constructed via the call acquisition module 604 and the construction module 605.

The obtaining module 604 is configured to query the relational database according to the service number to query all the ticket element data corresponding to the service number.

In the embodiment of the invention, all the bill element data under the business is stored in the relational database. The obtaining module 604 may query the relational database according to the service number carried in the audit request to query all the ticket element data under the service. In the implementation, before a business (such as a documentary collection business) enters an auditing link, all bill element data in a bill file can be extracted through an OCR (optical image recognition) technology and an NLP (natural language processing) technology, and the bill element data is stored in a relational database in a structured form. In addition, preferably, the granularity of the extraction of the bill elements is consistent with the granularity of the attributes defined by the bill body.

And the constructing module 605 is configured to construct the bill verification knowledge graph instance data according to the bill element data.

In an optional example, the constructing module 605 constructing the ticket review knowledge-graph instance data according to the ticket element data may specifically include: the construction module 605 assembles the element data under each bill and the pre-configured bill body to obtain bill knowledge map instance data; the construction module 605 constructs bill audit knowledge graph instance data according to the bill knowledge graph instance data and a preconfigured rule ontology.

The bill body refers to a bill body related to the examination bill. For example, in a documentary collection service audit scene, the bill body includes the body of bills such as draft, invoice, facial letter, bill of lading, and the like. The rule ontology refers to an ontology of an audit rule related to the examination order. For example, in a documentary collection business audit scenario, the rule ontology includes an ontology of an indelible item audit rule and an element consistency audit rule. The ontology of the element consistency audit rule can be divided into an ontology of an amount consistency audit rule and an applicant consistency audit rule.

And the auditing module 606 is used for traversing and executing each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result.

Wherein the bill audit knowledge graph instance data is stored in a form of a graph structure. The audit module 606 can traverse each rule entity vertex in the ticket audit knowledge-graph instance data to execute each audit rule. For example, when traversing to an amount checking vertex, all edges pointing to the vertex and attributes of the edges are obtained, and by parsing the contents of the attributes, all attributes participating in the amount checking can be obtained, and assuming that the attribute values are invoke. Figure 67776.6 for invoice amount, USD for figure currency for invoice amount, 67776.6 for draft figure money for draft currency, and USD for draft figure currency for draft currency. Next, the values of invice, figure and draft, figure are compared, and the values of invice, currency and draft, currency are compared. And if the comparison result shows that the money amount and the currency are consistent, the verification result aiming at the verification rule is consistent. And if the comparison result is that the amount or currency is inconsistent, the verification result aiming at the verification rule is inconsistent. And after all the audit rules are executed, outputting a final audit result.

In the embodiment of the invention, the automatic execution of the order examination service is realized through the model. According to the processing flow realized by the document examination device, the document examination result can be directly output, the document examination expert is assisted to examine the document, the processing efficiency of the document examination service is improved, the time consumed by a single-pen document examination is saved, and the document examination precision is improved.

Fig. 7 illustrates an exemplary system architecture 700 to which the document reviewing method or apparatus of the present invention may be applied.

As shown in fig. 7, the system architecture 700 may include

terminal devices

701, 702, 703, a network 704, and a server 705. The network 704 serves to provide a medium for communication links between the

terminal devices

701, 702, 703 and the server 705. Network 704 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

A user may use the

terminal devices

701, 702, 703 to interact with a server 705 over a network 704, to receive or send messages or the like. Various communication client applications, such as a business audit management application, a shopping application, a web browser application, a search application, an instant messaging tool, a mailbox client, social platform software, and the like, may be installed on the

terminal devices

701, 702, and 703.

The

terminal devices

701, 702, 703 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 705 may be a server providing various services, such as a background management server providing support for a business audit management application or a website browsed by a user using the

terminal devices

701, 702, and 703. The background management server may analyze and perform other processing on the received data such as the audit request, and feed back a processing result (e.g., an audit result) to the terminal device.

It should be noted that the order examination method provided by the embodiment of the present invention is generally executed by the server 705, and accordingly, the order examination apparatus is generally disposed in the server 705.

It should be understood that the number of terminal devices, networks, and servers in fig. 7 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

Referring now to FIG. 8, shown is a block diagram of a computer system 800 suitable for use in implementing an electronic device of an embodiment of the present invention. The computer system illustrated in FIG. 8 is only one example and should not impose any limitations on the scope of use or functionality of embodiments of the invention.

As shown in fig. 8, the computer system 800 includes a Central Processing Unit (CPU)801 that can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)802 or a program loaded from a storage section 808 into a Random Access Memory (RAM) 803. In the RAM 803, various programs and data necessary for the operation of the system 800 are also stored. The CPU 801, ROM 802, and RAM 803 are connected to each other via a bus 804. An input/output (I/O) interface 805 is also connected to bus 804.

The following components are connected to the I/O interface 805: an input portion 806 including a keyboard, a mouse, and the like; an output section 807 including a signal such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage portion 808 including a hard disk and the like; and a communication section 809 including a network interface card such as a LAN card, a modem, or the like. The communication section 809 performs communication processing via a network such as the internet. A drive 810 is also connected to the I/O interface 805 as necessary. A removable medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 810 as necessary, so that a computer program read out therefrom is mounted on the storage section 808 as necessary.

In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication section 809 and/or installed from the removable medium 811. The computer program executes the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 801.

It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes an acquisition module, a construction module, and an audit module. The names of these modules do not in some cases constitute a limitation to the module itself, and for example, the acquiring module may also be described as a "module that acquires ticket element data".

As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: after receiving an audit request, acquiring all bill element data corresponding to a service number carried by the audit request; establishing bill auditing knowledge map example data according to the bill element data; and traversing and executing each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result.

According to the technical scheme of the embodiment of the invention, the work efficiency of bill auditing can be improved, the time consumption of bill auditing is reduced, and the accuracy of the bill auditing result is improved.

The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method of conducting an audit, the method comprising:

after receiving an audit request, acquiring all bill element data corresponding to a service number carried by the audit request;

establishing bill auditing knowledge map example data according to the bill element data;

and traversing and executing each auditing rule in the bill auditing knowledge graph instance data to obtain an auditing result.

2. The method of claim 1, further comprising:

prior to the step of constructing document audit knowledge map instance data from the document element data,

inquiring a database according to the service number carried by the audit request; and when the query result shows that historical bill auditing knowledge map instance data corresponding to the service number exist in the map database, clearing the historical bill auditing knowledge map instance data.

3. The method of claim 2, wherein constructing document audit knowledge graph instance data from the document element data comprises:

assembling the element data under each bill and a preset bill body to obtain bill knowledge map example data; and constructing bill auditing knowledge graph instance data according to the bill knowledge graph instance data and a preset rule body.

4. The method of claim 3, wherein constructing document audit knowledge graph instance data from the document element data comprises:

preprocessing the bill element data according to a basic knowledge ontology; the basic ontology comprises at least one of: a company ontology, a country ontology, and a bank ontology.

5. The method according to claim 4, wherein the acquiring all ticket element data corresponding to the service number carried by the audit request comprises:

and querying a relational database according to the service number carried by the audit request so as to query all the bill element data corresponding to the service number.

6. The method of claim 5, further comprising:

before querying a relational database according to the service number carried by the audit request, extracting all bill element data in a bill file by an optical image recognition technology and a natural language processing technology, and storing the bill element data in the relational database.

7. The method of claim 6, wherein the audit request is a documentary collection service audit request.

8. An apparatus for conducting an audit, the apparatus comprising:

the acquisition module is used for acquiring all bill element data corresponding to the service number carried by the audit request after receiving the audit request;

the construction module is used for constructing bill verification knowledge map example data according to the bill element data;

and the auditing module is used for traversing and executing each auditing rule in the bill auditing knowledge graph example data to obtain an auditing result.

9. An electronic device, comprising:

one or more processors;

a storage device for storing one or more programs,

when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-7.

10. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-7.