CN109325102B

CN109325102B - Method and device for identifying illegal document

Info

Publication number: CN109325102B
Application number: CN201811094734.5A
Authority: CN
Inventors: 张国锐
Original assignee: Kingdee Software China Co Ltd
Current assignee: Kingdee Software China Co Ltd
Priority date: 2018-09-19
Filing date: 2018-09-19
Publication date: 2021-02-19
Anticipated expiration: 2038-09-19
Also published as: CN109325102A

Abstract

The embodiment of the application discloses a method for identifying illegal documents, which comprises the following steps: determining an abnormal value corresponding to the target field according to the historical document data; acquiring a target document, wherein the target document comprises a target field; judging whether the corresponding value of the target field in the target document is matched with the abnormal value; and if so, prompting that the target document is an illegal document. The embodiment of the application also provides a corresponding identification device. According to the technical scheme, the illegal documents can be effectively identified in the process of checking a large number of documents, the problem caused by checking a large number of documents by means of the credit score is solved, and the checking efficiency and accuracy are improved.

Description

Method and device for identifying illegal document

Technical Field

The application relates to the technical field of identification, in particular to a method and an identification device for identifying illegal documents.

Background

Documents are well known as written proofs obtained or filled in when economic business occurs, which specify the actual conditions of transactions and matters, such as receipts, warehousing documents, and the like. In the financial management process of enterprises, the financial management system is the original data and important basis for accounting, so financial personnel of each enterprise can spend a great deal of energy on the checking of documents, such as the personnel, the amount, the reasons and the like on the documents. Today, financial management software is widely used by enterprises, and the accuracy and the normalization of documents checked by the software determine the financial management efficiency of the enterprises.

In the prior art, the document is usually checked by credit, which is mainly based on credit score. Most of the time, the bill of people with low credit score is paid more attention as a potential illegal document, and the bill of people with high credit score is not paid much attention.

Therefore, when a large number of documents need to be checked, the credit is only used for observation from the credit score, the angle is single, and illegal documents are difficult to find in the checking process.

Disclosure of Invention

The embodiment of the application provides a method and a device for identifying illegal documents, which can effectively identify the illegal documents in the process of checking a large number of documents.

In view of this, the embodiments of the present application provide the following solutions:

the first aspect of the present application provides a method for identifying an illegal document, which can be applied to the financial field, and the method for identifying an illegal document may include: determining an abnormal value corresponding to the target field according to the historical document data; acquiring a target document, wherein the target document comprises a target field; judging whether the corresponding value of the target field in the target document is matched with the abnormal value; and if so, prompting that the target document is an illegal document. The historical document data may have been stored in a database, and the target fields may include various types of fields, such as: numerical type, literal type, etc. According to the first aspect, the value corresponding to the target field in the target document is matched with the abnormal value determined according to the historical document data, if the value is matched, the target document can be considered to have higher illegal potential risk, and the document can be prompted to be an illegal document; the method solves the problem caused by only depending on the single angle of credit score to identify the illegal documents in the process of checking a large number of documents, and effectively identifies the illegal documents.

Optionally, with reference to the first aspect, in a first possible implementation manner, determining an abnormal value corresponding to a target field according to historical document data may include: dividing values corresponding to target fields in the historical document data into N data groups, wherein N is a positive integer; calculating the abnormal proportion corresponding to each data group and the average proportion corresponding to N data groups, wherein the abnormal proportion is obtained by calculating the number of abnormal documents and the total number of the documents, the total number of the documents is the number of all documents of which the value of the target field belongs to the data group, and the number of the abnormal documents is the number of illegal documents in all the documents; and determining the data group with the difference value between the abnormal proportion and the average proportion larger than a preset threshold value as a target data group, and determining the value in the target data group as the abnormal value of the target field. Because different fields belong to different types, the target fields in the historical document data can be divided into different types, and then the abnormal proportion and the average proportion of each field in each type are calculated respectively.

Optionally, with reference to the first possible implementation manner of the first aspect, in a second possible implementation manner, the target field includes a name field, and dividing a value corresponding to the target field in the historical document data into N data groups may include: dividing each name corresponding to the name field into a data group, wherein the sum of the names is N; correspondingly, after determining the value in the target data group as the abnormal value of the target field, judging whether the corresponding value of the target field in the target document is matched with the abnormal value includes: acquiring a corresponding name of a name field in a target document; judging whether the acquired name is the same as the abnormal value of the name field; and if so, prompting that the target document is an illegal document. The target document can comprise a plurality of types of target fields, but different statistical modes are required for different target fields, so that when the target fields are name fields, abnormal values can be determined according to the statistical mode corresponding to the name fields; and the name corresponding to the name field in the target document is matched with the abnormal value, so that the identification accuracy is effectively improved.

Optionally, with reference to the first possible implementation manner of the first aspect, in a third possible implementation manner, the target field includes a numeric field, and dividing a value corresponding to the target field in the historical document data into N data groups may include: arranging numerical values corresponding to the numerical value fields; the divided and arranged numerical values are N equidistant intervals, and each divided numerical value corresponds to the interval one by one; determining N equidistant intervals as N data groups; correspondingly, after determining the value in the target data group as the abnormal value of the target field, determining whether the corresponding value of the target field in the target document matches the abnormal value may include: acquiring a corresponding numerical value of the numerical field in the target document; judging whether the obtained numerical value falls into a target data group or not; and if so, prompting that the target document is an illegal document. The numerical value corresponding to the numerical value field can be divided into intervals, so that the abnormal interval can be determined according to the statistical mode corresponding to the numerical value field, and whether the numerical value corresponding to the numerical value field in the target document falls in the abnormal interval or not is judged, and the identification accuracy is effectively improved.

Optionally, with reference to the first possible implementation manner of the first aspect, in a fourth possible implementation manner, the target field includes similar fields, and dividing a value corresponding to the target field in the historical document data into N data groups may include: determining words with similar meaning in similar fields as similar words, wherein the similar words are obtained by combining statement analysis software and a word bank; counting the number sum of class ranges, wherein the number sum of the class ranges is N, and each class similar word is determined as a class range in the class ranges; determining N class ranges as N data groups; correspondingly, after determining the value in the target data group as the abnormal value of the target field, determining whether the corresponding value of the target field in the target document matches the abnormal value may include: acquiring corresponding words of the similar fields in the target document; judging whether the obtained words fall into a target data group or not; and if so, prompting that the target document is an illegal document. Because the words with similar meaning corresponding to the similar fields can be divided into similar words, the abnormal class range can be determined according to the statistical mode corresponding to the similar fields, and whether the words corresponding to the similar fields in the target document fall in the abnormal class range is judged, so that the identification accuracy and efficiency are effectively improved.

Optionally, with reference to the first aspect, in a fifth possible implementation manner, before determining an abnormal value corresponding to the target field according to the historical document data, the method may further include: obtaining a range table, wherein the range table comprises a first table, a second table and a third table; setting a first incidence relation and a second incidence relation, wherein the first incidence relation is the incidence relation between an external key in a first table and a main key in a second table, the second incidence relation is the incidence relation between the external key in the second table and a main key in a third table, the external key in the first table is the main key in the second table, the external key in the second table is the main key in the third table, and the first incidence relation and the second incidence relation are used for acquiring historical bill data. As the tables are associated with each other through the external key and the main key, the data in the table where the external key is located can be acquired, and the consistency and the integrity of the data are kept.

A second aspect of the present application provides an identification apparatus having functionality to implement the method of the first aspect or any one of the possible implementations of the first aspect. The function can be realized by hardware, and can also be realized by executing corresponding software by hardware. The hardware or software includes one or more units, modules or sub-modules corresponding to the above-described functions.

A third aspect of the present application provides a computer device comprising: a processor and a memory; the memory is configured to store program instructions that, when executed by the processor, cause the apparatus to perform a method of identifying an illegitimate document as described in the first aspect or any one of the possible implementations of the first aspect.

A fourth aspect of the present application provides a computer-readable storage medium having stored therein instructions, which, when run on a computer, enable the computer to perform the method for identifying an illegal document of the first aspect or any one of the possible implementations of the first aspect.

A fifth aspect of the present application provides a computer program product comprising instructions which, when run on a computer, enable the computer to perform the method of identifying an illegal document of the first aspect or any one of the possible implementations of the first aspect.

A sixth aspect of the present application provides a chip system, which includes a processor, configured to enable an apparatus to implement the functions recited in the first aspect or any one of the possible implementation manners of the first aspect. In one possible design, the system-on-chip further includes a memory, the memory being used to hold the necessary program instructions and data for the device. The chip system may be constituted by a chip, or may include a chip and other discrete devices.

For technical effects brought by any one implementation manner of the second aspect, the third aspect, the fourth aspect, the fifth aspect, and the sixth aspect, reference may be made to technical effects brought by different implementation manners in the first aspect, and details are not repeated here.

According to the technical scheme, the embodiment of the application has the following advantages:

dividing different types of target fields, dividing values corresponding to the target fields in the historical document data of each type into data groups after the types are divided, determining abnormal values corresponding to the target fields of each type by combining the abnormal proportion of each data group and the average proportion of all the data groups, matching the values corresponding to the target fields of each type in the target documents with the determined abnormal values, and prompting the target documents to be illegal documents if the values are matched; therefore, in the process of checking a large number of documents, illegal documents can be effectively identified, the problem caused by checking a large number of documents by relying on the credit score is solved, and the checking efficiency and the checking accuracy are improved.

Drawings

FIG. 1 is a schematic diagram of a scene of identifying an illegal document in an embodiment of the application;

FIG. 2 is a schematic diagram of an embodiment of a method for identifying an illegal document according to an embodiment of the present application;

FIG. 3 is a schematic diagram of another embodiment of a method for identifying an illegal document according to an embodiment of the present application;

FIG. 4 is a schematic diagram of another embodiment of a method for identifying an illegal document according to an embodiment of the present application;

FIG. 5 is a schematic diagram of another embodiment of a method for identifying an illegal document according to an embodiment of the present application;

FIG. 6 is a schematic diagram of an embodiment of an identification device provided in an embodiment of the present application;

fig. 7 is a schematic diagram of another embodiment of an identification device provided in an embodiment of the present application;

fig. 8 is a schematic diagram of another embodiment of an identification device provided in an embodiment of the present application.

Detailed Description

The embodiment of the application provides a method for identifying illegal documents. The embodiment of the application also provides a corresponding device, so that the illegal documents can be effectively identified in the process of checking a large number of documents.

FIG. 1 is a schematic view of a scene of identifying an illegal document in an embodiment of the application.

In the field of finance, documents are often the original material and important basis for accounting, and are written proofs which are obtained or filled in when an economic transaction occurs and specify the actual conditions of transactions and matters, such as money involved when the economic transaction occurs and economic transactions of when and when people are generated are all reflected in the documents.

In the financial field, the truth of the document is generally regarded as important, and a great deal of time, energy and cost are required to check the document. And the fields and the content forms of each document are different, the authenticity needs to be checked by virtue of experience and credit score, the angle is single, and the illegal document cannot be reasonably identified.

As shown in fig. 1, the charge invoice includes a plurality of fields, such as amount, application description, payee, sponsor, verifier, etc., which all belong to different categories, such as amount is represented by numerical value, and payee, sponsor and verifier are represented by name, which is used to describe that they belong to text description. Therefore, in the scheme, the identification device judges and identifies the authenticity of the expense reimbursement document by using different statistical methods according to different types of fields.

The identification device referred to in fig. 1 may be a server or a terminal device, and is not particularly limited.

Fig. 1 is an introduction of a scene schematic diagram of the present solution, and for convenience of understanding, a method for identifying an illegal document according to an embodiment of the present application is described below.

Fig. 2 is a schematic diagram of an embodiment of a method for identifying an illegal document according to an embodiment of the present application.

As shown in fig. 2, an embodiment of a method for identifying an illegal document provided by the embodiment of the present application includes:

201. and determining abnormal values corresponding to the target fields according to the historical document data.

In the implementation, which historical documents are illegal can be known from the data in the historical documents, so that the abnormal fields corresponding to the historical illegal documents can be known on the basis, and the abnormal values can be known from the abnormal features in the historical illegal documents.

202. And acquiring a target document, wherein the target document comprises a target field.

In this embodiment, the target document is a newly acquired document, and the target document may include multiple types of target fields. Such as: the target field may include one or more fields such as a name field, a value field, and a similar field, where the similar field refers to a field formed by words with similar meanings, and values in the field are words that can be expressed by a plurality of different words, such as: terms such as sick, inappropriate and hospitalized are values in a similar field.

203. And judging whether the corresponding value of the target field in the target document is matched with the abnormal value.

In this embodiment, the value corresponding to the target field in the target document needs to be matched with the abnormal value determined according to the historical document data, and whether the value is matched with the abnormal value is judged, so that the function of improving the identification accuracy is achieved.

204. And if so, prompting that the target document is an illegal document.

In this embodiment, after the value corresponding to the target field in the target document is judged to be matched with the abnormal value determined according to the historical document data, it can be shown that the target field in the target document has a higher abnormal risk, so that the user is prompted that the target document is an illegal document according to the abnormal risk result.

It should be noted that if there is no match, the risk of abnormality existing in the target field in the target document may be considered to be low, and the risk may be ignored.

In the embodiment, whether a value corresponding to a target field in a target document is matched with an abnormal value determined according to historical document data is judged, and if yes, the target document is prompted to be an illegal document; the embodiment can effectively identify the illegal documents in the process of checking a large number of documents, and improves the checking efficiency and accuracy.

In the technical solution, the target field may include a plurality of different types of fields, and the field corresponding to each type needs to adopt different statistical methods to determine the abnormal value corresponding to each type.

The following is to divide the target field, and respectively describe in detail the scheme for identifying an illegal document provided by the embodiment of the present application.

1. The target field includes a name field;

2. the target field comprises a numerical value field;

3. the target field includes similar fields.

It should be noted that, all of the above 3 schemes are to divide the types of the target fields, and determine abnormal values corresponding to the target fields by using different statistical methods for the fields corresponding to different types, and all of the purposes are to improve the accuracy of identification.

1. The target field includes a name field

The present embodiment is described by taking an example in which the target field includes a name field. Fig. 3 is a schematic diagram of another embodiment of the method for identifying an illegal document according to the embodiment of the application.

As shown in fig. 3, another embodiment of the method for identifying an illegal document provided in the embodiment of the present application includes:

301. obtaining a range table, wherein the range table comprises a first table, a second table and a third table.

In this embodiment, since the historical documents may be stored in the database or in the cloud, and it is impossible to determine the abnormal value by using all the documents, the abnormal value is determined only by acquiring a part of the documents as a sample.

It should be noted that the first table mentioned above may be an audit record table, the second table may be a table associated with the audit record table and set as an association table, and the third table may be a table associated with the association table. The scope table is related to documents that have been approved, documents that have been approved are recorded in the audit record table, and document records that fail in the audit record table are recorded with a flag.

302. And setting the first incidence relation and the second incidence relation for obtaining the historical document data.

In this embodiment, after obtaining the range tables, we do not know how we find the required historical document data from the range tables. For example, if the audit list record table is the primary key of the employee field in the first table and the document data related to the employee field is stored in other documents, if the relationship between the employee field and the abnormal document needs to be counted, the employee table and the audit list record table must be associated, and the second table and the third table are similar to this principle. Therefore, the association of each table needs to be set, so that the required historical document data can be conveniently acquired.

It should be noted that the first association relationship is an association relationship between an external key in the first table and a primary key in the second table, the second association relationship is an association relationship between an external key in the second table and a primary key in the third table, the external key in the first table is a primary key in the second table, the external key in the second table is a primary key in the third table, and the first association relationship and the second association relationship are used for acquiring the historical document data.

For example: assume that the first table is an audit record table, the audit record table includes the employee number as a foreign key, and the foreign key is a primary key in the employee table, because the employee number of the employee table as a primary key can uniquely identify data in the employee table. For example, employee number 1 uniquely corresponds to the name of one employee, Zhang three.

Audit recording table

Document number	Employee number
		01	1
02	2

Staff table

Employee number	Employee name
		1	Zhang three
2	Li Si

The second table is associated with a third, similar principle.

303. And dividing each name corresponding to the name field into a data group according to the historical document data.

In this embodiment, since the name field usually includes one or more names, each name is divided into two or more names, and the number of the data groups is N, where N is a positive integer; this facilitates the calculation of each name separately, i.e. the number of documents included in each data set.

It should be noted that the name field may include a company name, an employee name, or a reviewer name; in practical application, other names used for explaining the meaning of the name field can be further included, but the name field is only required to be related to the content in the document; the details are not limited herein.

304. And calculating the abnormal ratio corresponding to each data group and the average ratio corresponding to the N data groups.

In this embodiment, after dividing each name corresponding to the name field according to the historical document data into one data group, the abnormal proportion corresponding to each data group and the average proportion corresponding to N data groups can be calculated.

It should be noted that the abnormal proportion corresponding to each data group is obtained from the number of abnormal documents in the group and the total number of document data in the group, and the average proportion corresponding to N data groups is obtained from the number of abnormal documents in the N data groups and the total number of all documents in the N data groups, and is usually processed by division.

Further, assuming that the name field is the name of a person, the person may include zhang three, lie four, and wang five; when N is 3, the historical document data related to the field of the person name is assumed to be 10000, wherein the number of the abnormal documents is 3000; the total number of the bills corresponding to the data group of Zhang III is 2000, wherein the number of the abnormal bills is 250; the total number of the bills corresponding to the data group of the plum four is 4000 parts, wherein the number of the abnormal bills is 2000 parts; the total number of the bills corresponding to the Wangwu data set is 3000, wherein the number of the abnormal bills is 750;

then, the average ratio for these 3 data sets is 2000/10000 ═ 0.2; the abnormal proportion corresponding to the data group of Zhang III is 250/2000-0.125; the abnormal proportion corresponding to the data group Lifouri is 2000/4000-0.5; the abnormal proportion corresponding to the data group king five is 750/4000-0.1875.

305. And determining the data group with the difference value between the abnormal proportion and the average proportion larger than a preset threshold value as a target data group, and determining the value in the target data group as an abnormal value corresponding to the name field.

In this embodiment, after the abnormal proportion and the average proportion in each data group are calculated, a difference between the abnormal proportion and the average proportion may be calculated, and the calculated difference may be compared with a preset threshold value; if the number of the data groups is larger than the preset threshold, the data group corresponding to the number of the data groups larger than the preset threshold is determined as a target data group, and therefore the value in the target data group can be determined to be an abnormal value corresponding to the name field.

It should be noted that the preset threshold is a limit for judging whether the value in one data set is an abnormal value. Such as: assuming that the preset threshold may be 0.1, then: the difference value corresponding to the data group of zhangsan is 0.125-0.2-0.075, the difference value corresponding to the data group of lie four is 0.5-0.2-0.3, and the difference value corresponding to the data group of wangwu is 0.1875-0.2-0.0125; then it can be known that lie four is the target data set at this time, and lie four is an outlier corresponding to the name field; in practical applications, the preset threshold may also be other values, and is not limited herein.

306. And acquiring a target document, wherein the target document comprises a name field.

In this implementation, after the value in the target data group is determined as the abnormal value of the name field, the name corresponding to the name field in the target document can be obtained by obtaining the target document. For example: the obtained name field comprises a person name, and the name in the person name comprises Zhangming, Liqu, Wanglin and the like.

307. And judging whether the acquired name is the same as the value in the target data group.

In this embodiment, after obtaining the values of the name corresponding to the name field in the target document, it is possible to determine whether the values, that is, the name, are the same as the determined abnormal values. For example: from the above, it can be known that the abnormal value in the name field is lie four, and the obtained name in the target document includes zhang, lie four, wanglin and the like, and at this time, it can be clearly known that the abnormal value is matched with the lie four in the target document.

308. And if so, prompting that the target document is an illegal document.

In this embodiment, when it is determined that the name corresponding to the name field in the target document is the same as the determined abnormal value corresponding to the name field, the target document where the name is located can be prompted as an illegal document. For example: if the lie four in the target document is an abnormal value, the target document where the lie four with the abnormal value is located can be prompted to be an illegal document.

In this embodiment, when the target field is divided into name fields, the name corresponding to each name in the name fields is determined to be an abnormal value by calculating an abnormal proportion and an average proportion in the name fields, and comparing the difference between the abnormal proportion and the average proportion with a preset threshold value, and then determining that the name corresponding to the name larger than the preset threshold value is an abnormal value; and prompting that the target document where the name is located is an abnormal document if the obtained name is the same as the abnormal value. The embodiment can effectively identify the illegal documents in the process of checking a large number of documents, and improves the checking efficiency and accuracy.

2. The object field includes a value field

The present embodiment is described by taking only an example that the target field includes a value field. Fig. 4 is a schematic diagram of another embodiment of the method for identifying an illegal document according to the embodiment of the application.

As shown in fig. 4, another embodiment of the method for identifying an illegal document provided in the embodiment of the present application includes:

401. obtaining a range table, wherein the range table comprises a first table, a second table and a third table.

402. And setting the first incidence relation and the second incidence relation for obtaining the historical document data.

The steps 301-302 in this embodiment are similar to the steps 201-202 in fig. 2, and are not described herein again.

403. And arranging the numerical value corresponding to the numerical field according to the historical document data.

In this embodiment, the numerical values in the numerical value field may include one or more numerical values, so that for convenience of statistics, the numerical values corresponding to the numerical value field need to be arranged and counted, and the arrangement manner may be an ascending order or a descending order, or may be another arrangement manner, which is not limited herein.

404. The divided and arranged numerical values are N equidistant intervals, and each divided numerical value corresponds to the interval one by one.

In this embodiment, after the numerical values corresponding to the numerical values in the history document are arranged, the arranged numerical values may be divided into intervals, so that each numerical value has and can only be divided into one of the intervals.

For example: assuming that the value field is an amount field, the amount field in the history document data must correspond to one or more values, assuming that the upper limit and the lower limit of the value corresponding to the amount field are 5000 yuan and 200 yuan, other sorted values may include 300, 480, 550, 600, 730, 790, 820, …, 4890, 4999, etc. At this time, if the interval is divided into 30, i.e. N is 30, the group distance e is (upper limit-lower limit)/30, i.e. e is (5000-: [200,360), [360,520), [520,680), [680,840), … [4840,5000).

405. And determining N equidistant intervals as N data groups.

In this embodiment, after dividing the value corresponding to the amount field into intervals, each interval may be regarded as one data set, and then the 30 intervals with the interval of 160 may be determined as 30 data sets. In practical applications, further intervals are certainly included, and the details are not limited herein.

406. And calculating the abnormal ratio corresponding to each data group and the average ratio corresponding to the N data groups.

In this embodiment, after each interval corresponding to the value field is determined as one data group according to the historical document data, the abnormal proportion corresponding to each data group and the average proportion corresponding to the N data groups can be calculated.

It should be noted that the abnormal proportion corresponding to each data group is obtained from the number of abnormal documents in the group and the total number of documents in the group, and the average proportion corresponding to N data groups is obtained from the number of abnormal documents in the N data groups and the total number of all documents in the N data groups, and is usually processed by division.

Further, assume that the amount field and interval of step 304 are used. Assume that the number of documents included in the 30 intervals is 20000 shares at this time, wherein the number of abnormal documents is 3000 shares; wherein the number of the documents in the interval [200,360) is 500, and the number of the abnormal documents is 60; the number of the documents in the interval [360,520) is 1000, and the number of the abnormal documents is 50; the number of the documents in the interval [520,680) is 3000, and the number of the abnormal documents is 450; the number of the bills in the interval [680,840) is 1500, and the number of the abnormal bills is 450; … …, the number of the bills in the interval [4840,5000) is 800, and the number of the abnormal bills is 180;

then, the average ratio for these 30 data sets is 3000/20000 ═ 0.15; the anomaly percentage corresponding to the data group [200,360) is 60/500 ═ 0.12; the abnormal proportion corresponding to the data group [360,520) is 50/1000 ═ 0.05; the abnormal proportion corresponding to the data group [520,680) is 450/3000 ═ 0.15; the abnormal proportion corresponding to the data group [680,840) is 450/1500 ═ 0.3; … … data set [4840,5000) has an anomaly percentage of 180/800 to 0.225.

407. And determining the data group with the difference value between the abnormal proportion and the average proportion larger than a preset threshold value as a target data group, and determining the value in the target data group as an abnormal value corresponding to the numerical value field.

In this embodiment, after the abnormal proportion and the average proportion in each data group are calculated, a difference between the abnormal proportion and the average proportion may be calculated, and the calculated difference may be compared with a preset threshold value; if the value of the abnormal value is larger than the preset threshold, the data group corresponding to the value larger than the preset threshold is determined as the target data group, and therefore the value in the target data group is determined to be the abnormal value corresponding to the numerical value field.

For example: adopting the amount field and the corresponding interval of the step 306, assuming that the preset threshold is 0.1, at this time: data set [200,360) corresponds to a difference of 0.12-0.15 ═ 0.03; data set [360,520) corresponds to a difference of 0.05-0.15 ═ 0.1; data set [520,680) corresponds to a difference of 0.15-0; data set [680,840) corresponds to a difference of 0.3-0.15 to 0.15; … … data set [4840,5000) corresponds to a difference of 0.225-0.15 to 0.075; then it can be known that the data set [680,840) is the target data set, and the value in the data set [680,840) is the outlier to which the amount field corresponds.

408. And acquiring a target document, wherein the target document comprises a numerical value field.

In this implementation, after the value in the target data group is determined as the abnormal value of the value field, the value corresponding to the value field in the target document can be obtained by obtaining the target document.

For example: the value field obtained includes an amount field, and the values in the amount field include 280, 720, etc.

409. And judging whether the acquired numerical value falls into the target data group.

In this embodiment, after obtaining the amount values corresponding to the amount fields in the target document, it is possible to determine whether the amount values fall into the determined target data group.

For example: from the above, it can be seen that the data group [680,840) in the amount field contains abnormal values, and the amount value in the obtained target document includes 280, 720, etc., at this time, it is clear that the value of 280 in the target document does not fall within the data group [680,840). And the value 720 falls within the data set [680,840), the value 720 is an outlier.

410. And if so, prompting that the target document is an illegal document.

In this embodiment, when it is determined that the value corresponding to the value field in the target document falls into the target data group corresponding to the determined value field, the target document in which the value is located can be prompted as an illegal document.

For example: if the value 720 in the target document is an abnormal value, it can indicate that the target document in which the abnormal value 720 is located is an illegal document.

In this embodiment, whether the value corresponding to the obtained value field falls into the target data group containing the abnormal value is judged, and if yes, the target document where the value is located is prompted to be an illegal document. The embodiment can effectively identify the illegal documents in the process of checking a large number of documents, and improves the checking efficiency and accuracy.

3. The object field includes similar fields

The present embodiment is described by taking only an example that the target field includes similar fields. Fig. 5 is a schematic diagram of another embodiment of the method for identifying an illegal document according to the embodiment of the application.

As shown in fig. 5, another embodiment of the method for identifying an illegal document provided in the embodiment of the present application includes:

501. obtaining a range table, wherein the range table comprises a first table, a second table and a third table.

502. And setting the first incidence relation and the second incidence relation for obtaining the historical document data.

Steps 501-502 in this embodiment are similar to steps 301-302 in fig. 2, and are not described herein again.

503. And determining that the words with similar meaning in the similar fields are similar words according to the historical document data, wherein the similar words are obtained by combining statement analysis software and a word bank.

In this embodiment, since the words with similar meanings in the similar fields cannot be directly recognized by the device as the words with similar meanings, the words are recognized by combining the sentence analysis software and the word bank, and it is determined which words can be used as the same kind of words.

It should be noted that the similar fields mentioned refer to fields formed by words with similar meanings, such as reasons, comments, remarks, and the like; the words in these fields may be expressed in a variety of words, but all have the same meaning. The field values of the similar fields are not limited herein.

For example: the reason field in the reimbursement document can be expressed by a plurality of similar words. The words of illness, discomfort or hospitalization are the values in the field, but all the words are the meaning of expressing illness, so the words of illness, discomfort or hospitalization can be used as the words of the same kind.

504. Counting the number sum of class ranges, wherein the number sum of the class ranges is N, and the class range is determined to be a class range by each class homonym.

In this embodiment, after similar words in similar fields are obtained by combining the sentence analysis software and the word library, each similar word can be determined as a class range, and at this time, the number of all class ranges can be calculated.

For example: the reason field in step 503 may further include: the term "overdue", overtime or overdue "can be used as a category. So the sum of the number of such ranges is N ═ 2; in practical applications, the reason field further includes more words of the same kind that express the same meaning, and is not limited herein.

505. Determining N class ranges as N data sets

In this embodiment, after determining the similar words corresponding to the reason field as class ranges, each class range may be regarded as one data set, and then the 2 class ranges may be determined as 2 data sets. In practical applications, it is understood that the scope of the present invention includes more categories, and is not limited herein.

506. And calculating the abnormal ratio corresponding to each data group and the average ratio corresponding to the N data groups.

In this embodiment, after determining that each class range corresponding to the reason field is a data group according to the historical document data, the abnormal proportion corresponding to each data group and the average proportion corresponding to N data groups can be calculated.

Assume that the class ranges of

steps

504 and 505 described above are used. Suppose that the number of documents contained in the 2 data sets is 3000 at this time, wherein the number of abnormal documents is 600; the number of the bills in the data group of the word of illness is 800, and the number of the abnormal bills is 380; the number of documents in the data group where the term is overdue is 2200, and the number of abnormal documents is 220;

then, the average ratio for these 2 data sets is 600/3000 ═ 0.2; the abnormal proportion corresponding to the data group in which the word of illness is located is 380/800-0.475; the abnormal proportion corresponding to the data group in which the term is overdue is 220/2200-0.1.

507. And determining the data group with the difference value between the abnormal proportion and the average proportion larger than a preset threshold value as a target data group, and determining the value in the target data group as an abnormal value corresponding to the similar field.

In this embodiment, after the abnormal proportion and the average proportion in each data group are calculated, a difference between the abnormal proportion and the average proportion may be calculated, and the calculated difference may be compared with a preset threshold; if the data set is larger than the preset threshold, the data set corresponding to the data set larger than the preset threshold is determined as the target data set, and therefore the value in the target data set can be determined to be the abnormal value corresponding to the similar field.

For example: with the reason field and the corresponding data set in step 506 above, assuming that the preset threshold is 0.1, at this time: the difference value corresponding to the data group where the word is ill is 0.475-0.2-0.275; the difference value corresponding to the data group where the term is overdue is 0.1-0.2-0.1; then it can be known that the data set in which the word with illness is located is the target data set, and the word with illness, discomfort or hospitalization in the data set is the abnormal value corresponding to the reason field.

508. And acquiring a target document, wherein the target document comprises similar fields.

In this implementation, after the value in the target data group is determined as the abnormal value of the similar field, the corresponding word of the similar field in the target document can be obtained by obtaining the target document.

For example: the obtained similar fields comprise reason fields, and words in the reason fields comprise discomfort and the like; the details are not limited herein.

509. And judging whether the acquired words fall into the target data group.

In this embodiment, after the words corresponding to the reason fields in the target document are obtained, it is possible to determine whether the words fall into the determined target data group.

For example: from the above, it can be known that the cause field contains an abnormal value, and the word in the obtained target document is uncomfortable, at this time, it can be clearly known that the word of the discomfort in the target document falls in the data group, which indicates that the discomfort is an abnormal value.

510. And if so, prompting that the target document is an illegal document.

In this embodiment, when it is determined that the term corresponding to the similar field in the target document falls into the target data group corresponding to the determined similar field, the target document in which the term is located can be prompted as an illegal document.

For example: if the word of the discomfort in the target document is an abnormal value, the target document with the improper value can be prompted to be an illegal document.

In this embodiment, whether the obtained word corresponding to the similar field falls into the target data group containing the abnormal value is judged, and if yes, the target document where the word is located is prompted to be an illegal document. The embodiment can effectively identify the illegal documents in the process of checking a large number of documents, and improves the checking efficiency and accuracy.

The scheme provided by the embodiment of the application is mainly introduced from a single-side perspective. It is to be understood that the hardware structure and/or software modules for performing the respective functions are included to realize the above functions. Those of skill in the art will readily appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

In the embodiment of the present application, functional modules of the apparatus may be divided according to the above method example, for example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. It should be noted that, in the embodiment of the present application, the division of the module is schematic, and is only one logic function division, and there may be another division manner in actual implementation.

Fig. 6 is a schematic diagram of an embodiment of an identification device provided in an embodiment of the present application.

As shown in fig. 6, the identification apparatus provided in the embodiment of the present application includes a determining unit 601, an obtaining unit 602, a determining unit 603, and a prompting unit 604;

the determining unit 601 is configured to determine an abnormal value corresponding to the target field according to the historical document data;

the obtaining unit 602 is configured to obtain a target document, where the target document includes a target field corresponding to the abnormal value determined by the determining unit 601;

the judging unit 603 is configured to judge whether a value corresponding to the target field in the target document obtained by the obtaining unit 602 matches the abnormal value determined by the determining unit;

a prompting unit 604, configured to prompt the target document to be an illegal document when the value of the target field determined by the determining unit 603 is matched with the abnormal value.

In this embodiment, whether the value corresponding to the acquired target field matches the determined abnormal value is judged, and if yes, the target document where the value is located is prompted to be an illegal document. The embodiment can effectively identify the illegal documents in the process of checking a large number of documents, and improves the checking efficiency and accuracy.

For easy understanding, please refer to fig. 7 to understand the identification device in the embodiment of the present application in detail, fig. 7 is a schematic view of another embodiment of the identification device in the embodiment of the present application, and includes a determining unit 701, an obtaining unit 702, a determining unit 703 and a prompting unit 704, which have functions similar to those of the above 601-604;

the determining unit 701 in this embodiment may include:

the dividing module 7011 is configured to divide a value corresponding to the target field in the historical document data into N data groups;

a calculating module 7012, configured to calculate an abnormal ratio corresponding to each data group divided by the dividing module 7011 and an average ratio corresponding to N data groups divided by the dividing module 7011;

a determining module 7013, configured to determine, as a target data group, a data group in which a difference between the abnormal proportion calculated by the calculating module 7012 and the average proportion is greater than a preset threshold, and determine a value in the target data group as an abnormal value of the target field.

The identification apparatus in the embodiment of the present application is described above from the perspective of the modular functional entity, and the identification apparatus module in the embodiment of the present application is described below from the perspective of the hardware processing, please refer to fig. 8, where another embodiment of the identification apparatus in the embodiment of the present application includes:

an input interface 801, an output interface 802, a processor 803 and a memory 804, (wherein the number of processors 801 in the identification device may be one or more, one processor 801 is taken as an example in fig. 8). In some embodiments of the present application, the input interface 801, the output interface 802, the processor 803 and the memory 804 may be connected by a bus or other means, wherein fig. 8 illustrates the connection by a bus.

The processor 803 is configured to implement the above-provided embodiment of identifying illegal documents by calling the operating instructions stored in the memory 804.

Specifically, the functions/implementation processes of the determining unit 601, the obtaining unit 602, the judging unit 603, and the prompting unit 604 in fig. 6, and the determining unit 701, the obtaining unit 702, the judging unit 703, the prompting unit 704, the dividing module 7011, the calculating module 7012, and the determining module 7013 in fig. 7 may be implemented by the processor 803 in fig. 8 calling a computer stored in the memory 804 to execute instructions.

It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims

1. A method of identifying an illegal document, comprising:

determining an abnormal value corresponding to the target field according to the historical document data;

acquiring a target receipt, wherein the target receipt comprises the target field;

judging whether the value of the target field in the target receipt is matched with the abnormal value;

if the target document is matched with the illegal document, prompting that the target document is the illegal document;

determining an abnormal value corresponding to the target field according to the historical document data, wherein the abnormal value comprises the following steps:

dividing values corresponding to the target fields in the historical receipt data into N data groups, wherein N is a positive integer;

calculating an abnormal proportion corresponding to each data group and an average proportion corresponding to the N data groups, wherein the abnormal proportion is obtained by calculating the number of abnormal bills and the total number of the bills, the total number of the bills is the number of all the bills belonging to the data groups, and the number of the abnormal bills is the number of illegal bills in all the bills;

and determining a data group with the difference value between the abnormal proportion and the average proportion larger than a preset threshold value as a target data group, and determining the value in the target data group as the abnormal value of the target field.

2. The method of claim 1, wherein the target field comprises a name field, and wherein dividing values corresponding to the target field in the historical document data into N data groups comprises:

dividing each name corresponding to the name field into a data group, wherein the sum of the names is N;

correspondingly, after determining the value in the target data group as the abnormal value of the target field, determining whether the corresponding value of the target field in the target document matches the abnormal value includes:

acquiring a name corresponding to the name field in the target document;

judging whether the obtained name is the same as an abnormal value of the name field;

and if so, prompting that the target document is an illegal document.

3. The method of claim 1, wherein the target field comprises a numeric field, and wherein dividing values corresponding to the target field in the historical document data into N data groups comprises:

arranging the numerical values corresponding to the numerical value fields;

dividing the arranged numerical values into N equidistant intervals, wherein each divided numerical value corresponds to one interval;

determining the N equidistant intervals as the N data groups;

acquiring a value corresponding to the value field in the target document;

judging whether the obtained numerical value falls into the target data group;

and if so, prompting that the target document is an illegal document.

4. The method of claim 1, wherein the target field comprises a similar field, and dividing values corresponding to the target field in the historical document data into N data groups comprises:

determining words with similar meaning in the similar fields as similar words, wherein the similar words are obtained by combining statement analysis software and a word bank;

counting the number sum of class ranges, wherein the number sum of the class ranges is N, and each class similar word is determined as a class range in the class ranges;

determining the N class ranges as the N data groups;

acquiring corresponding words of the similar fields in the target document;

judging whether the obtained words fall into the target data group or not;

and if so, prompting that the target document is an illegal document.

5. The method of claim 1, further comprising, prior to determining the outlier corresponding to the target field from the historical document data:

obtaining a range table, wherein the range table comprises a first table, a second table and a third table;

setting a first incidence relation and a second incidence relation, wherein the first incidence relation is an incidence relation between a foreign key in the first table and a primary key in the second table, the second incidence relation is an incidence relation between a foreign key in the second table and a primary key in the third table, the foreign key in the first table is the primary key in the second table, the foreign key in the second table is the primary key in the third table, and the first incidence relation and the second incidence relation are used for acquiring the historical document data.

6. An identification device, comprising:

the determining unit is used for determining an abnormal value corresponding to the target field according to the historical document data;

the acquisition unit is used for acquiring a target document, and the target document comprises the target field corresponding to the abnormal value determined by the determination unit;

the judging unit is used for judging whether the value corresponding to the target field in the target document acquired by the acquiring unit is matched with the abnormal value determined by the determining unit;

the prompting unit is used for prompting that the target document is an illegal document when the judging unit judges that the value of the target field corresponding to the target document is matched with the abnormal value;

the determination unit includes:

the dividing module is used for dividing the value corresponding to the target field in the historical receipt data into N data groups;

the calculation module is used for calculating the abnormal proportion corresponding to each data group divided by the division module and the average proportion corresponding to the N data groups divided by the division module, the abnormal proportion is obtained by calculating the number of abnormal bills and the total number of bills, the total number of the bills is the number of all bills of the data group of the value of a target field, and the number of the abnormal bills is the number of illegal bills in all the bills;

and the determining module is used for determining a data group of which the difference value between the abnormal proportion and the average proportion calculated by the calculating module is larger than a preset threshold value as a target data group, and determining a value in the target data group as an abnormal value of the target field.

7. A computer device, characterized in that the computer device comprises: a processor and a memory, wherein the processor is capable of processing a plurality of data,

the memory has stored therein program instructions;

the processor is configured to execute program instructions stored in the memory to perform the method of any of claims 1-5.

8. A computer-readable storage medium comprising instructions that, when executed on a computer device, cause the computer device to perform the method of any of claims 1-5.