WO2020048247A1

WO2020048247A1 - Settlement data processing method and apparatus, and computer device and storage medium

Info

Publication number: WO2020048247A1
Application number: PCT/CN2019/096961
Authority: WO
Inventors: 夏雷
Original assignee: 平安医疗健康管理股份有限公司
Priority date: 2018-09-03
Filing date: 2019-07-22
Publication date: 2020-03-12
Also published as: CN109410069A

Abstract

Disclosed is a settlement data processing method, comprising: acquiring multiple initial settlement directories sent by a terminal, wherein the initial settlement directories comprise multiple pieces of data to be settled and multiple corresponding fields to be settled; acquiring a standard data table, matching the fields to be settled with standard fields corresponding to standard settlement data in the standard data table, and calculating the degree of matching between the fields to be settled and multiple standard fields; if there is a field to be settled with the degree of matching reaching a pre-set threshold value, adding a settlement attribute and category corresponding to the standard fields to the data to be settled; if there is a field to be settled with the degree of matching not reaching the pre-set threshold value, generating settlement data to be decided according to unmatched data to be settled; inputting the settlement data to be decided into a trained decision model for making a decision, and adding the obtained settlement attribute and category to the settlement data to be decided; and generating an item settlement directory according to the category, so that a server performs item settlement processing according to the item settlement directory.

Description

Settlement data processing method, device, computer equipment and storage medium

Cross-reference to related applications

This application claims the priority of a Chinese patent application filed on September 3, 2018 with the Chinese Patent Office under the application number 2018110201698 and the application name is "Settlement Data Processing Method, Device, Computer Equipment and Storage Medium", the entire contents of which are hereby incorporated by reference Incorporated in this application.

Technical field

The present application relates to a method, an apparatus, a computer device, and a storage medium for processing settlement data.

Background technique

With the rapid development of Internet technology, it is becoming more and more convenient to use Internet technology to settle various settlement items. For example, settlement of social insurance data and medical insurance data is very convenient. Using Internet technology to process various settlement data can effectively improve the processing efficiency of settlement data.

However, the inventors realized that traditional insurance settlement methods usually only settle according to the traditional social insurance settlement catalog. However, the scope of the traditional settlement directory is relatively limited. The settlement data in the settlement directories in various places is uneven and not uniform. As a result, when using the settlement catalog to connect and settle with the settlement items of various institutions, there may be problems such as complex calculations and the time-consuming conversion between settlement items, which leads to comparison of the efficiency and accuracy of processing according to the settlement catalog low. Therefore, how to effectively and accurately identify and match standardized settlement data in order to improve the processing efficiency of settlement data has become a technical problem that needs to be solved at present.

Summary of the Invention

According to various embodiments disclosed in the present application, a method, an apparatus, a computer device, and a storage medium for processing settlement data are provided.

A method for processing settlement data, including:

Acquiring multiple initial settlement directories sent by the terminal, where the initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled;

Obtaining a authority data table, where the authority data table includes authority settlement data and corresponding authority fields, and the authority settlement data is set with settlement attributes and categories;

Matching the field to be settled with a specification field in the specification data table, and calculating a degree of matching between the field to be settled and the plurality of specification fields;

If there is a field to be settled where the matching degree reaches a preset threshold, determine that the data to be settled corresponding to the field to be settled is standard settlement data, and add the settlement attributes and categories set by the standard settlement data corresponding to the standard field To the data to be settled;

If there is a field to be settled where the matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is non-standardized settlement data, and to-be-determined settlement data is generated based on the non-standardized to-be-settled data, and a decision model is obtained according to The decision model makes a decision on the settlement data to be decided, outputs settlement attributes and categories corresponding to the settlement data to be decided, and adds to the settlement data to be decided; and according to a plurality of added settlement attributes and categories, The settlement data generates a project settlement directory according to the category, so that the server performs project settlement processing according to the project settlement directory.

A settlement data processing device includes:

An obtaining module, configured to obtain multiple initial settlement directories sent by the terminal, the initial settlement directories including multiple pending settlement data and corresponding multiple pending settlement fields; obtaining a settlement data table, where the settlement data table includes standardized settlement data And corresponding multiple specification fields, the specification settlement data is set with settlement attributes and categories;

A matching module, configured to match the field to be settled with a specification field in the specification data table, and calculate a degree of matching between the field to be settled and the plurality of specification fields; if the degree of matching reaches a preset value A threshold to-be-settled field, determining that the to-be-settled data corresponding to the to-be-settled field is standard settlement data, and adding the settlement attribute and category set by the standard-settlement data corresponding to the specification field to the to-be-settled data;

A decision-making module, configured to output to-be-settled data corresponding to the to-be-settled fields as non-standard settlement data if there is a to-be-settled field with a matching degree that does not reach a preset threshold value, and generate to-be-determined settlement data based on the non-standard to-be-settled data, Inputting the settlement data to be decided into a trained decision model for decision making, obtaining settlement attributes and categories corresponding to the settlement data to be decided, and adding to the settlement data to be decided; and

The directory generating module is configured to generate a project settlement directory according to the category based on a plurality of settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.

A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the one or more processors execute the following: step:

If there is a field to be settled where the matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is non-standardized settlement data, and to-be-determined settlement data is generated based on the non-standardized to-be-settled data. Settlement data is input into a trained decision model for decision making, and outputs settlement attributes and categories corresponding to the settlement data to be determined, and is added to the settlement data to be determined; and settlement is added based on multiple settlement attributes and categories added The data generates a project settlement directory according to the category, so that the server performs project settlement processing according to the project settlement directory.

One or more non-volatile computer-readable storage media storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the following steps: obtaining multiple initial A settlement directory, where the initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled;

Obtaining a authority data table sent by the terminal, where the authority data table includes authority settlement data and corresponding authority fields, and the authority settlement data is set with a settlement attribute and category;

If there is a field to be settled where the matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is irregular settlement data, and to-be-determined settlement data is generated based on the non-standard to-be-settled data, and Settlement data is input into a trained decision model for decision making, outputting settlement attributes and categories corresponding to the pending decision settlement data, and added to the pending decision settlement data; and

Generate a project settlement directory according to the category based on multiple settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.

Details of one or more embodiments of the present application are set forth in the accompanying drawings and description below. Other features and advantages of the application will become apparent from the description, the drawings, and the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to explain the technical solutions in the embodiments of the present application more clearly, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. Those of ordinary skill in the art can obtain other drawings according to the drawings without paying creative labor.

FIG. 1 is an application scenario diagram of a settlement data processing method according to one or more embodiments.

FIG. 2 is a schematic flowchart of a settlement data processing method according to one or more embodiments.

FIG. 3 is a schematic flowchart of a decision-making process for decision-making settlement data according to one or more embodiments.

FIG. 4 is a schematic flowchart of steps of constructing a decision model according to one or more embodiments.

FIG. 5 is a structural block diagram of a settlement data processing apparatus according to one or more embodiments.

FIG. 6 is an internal structural diagram of a computer device according to one or more embodiments.

detailed description

In order to make the technical solution and advantages of the present application more clear and clear, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

The settlement data processing method provided in this application can be applied to the application environment shown in FIG. 1. The terminal 102 communicates with the server 104 through the network through the network. The terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices. The server 104 may be implemented by an independent server or a server cluster composed of multiple servers. The server 104 may obtain multiple initial settlement directories sent by multiple terminals 102. The initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled. The server 104 further obtains a authority data table, and the authority data table includes authority settlement data and corresponding authority fields. The authority settlement data is set with a settlement attribute and a category. Match the fields to be settled with the canonical settlement fields. If there is a field to be settled whose matching degree reaches a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is standard settlement data, and the settlement attributes and categories set by the standard settlement data corresponding to the standard field are added to the data to be settled. If there is a field to be settled whose matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is irregular settlement data, and the settlement data to be decided is generated according to the non-standard settled data. The server 104 inputs the settlement data to be decided into the trained decision model for decision making, outputs the settlement attributes and categories corresponding to the settlement data to be decided, and adds it to the settlement data to be decided. The server 104 generates a project settlement directory according to a corresponding category based on a plurality of settlement data to which settlement attributes and categories are added, and performs project settlement processing according to the project settlement directory.

In some embodiments, as shown in FIG. 2, a method for processing settlement data is provided. The method is applied to the server in FIG. 1 as an example, and includes the following steps:

Step 202: Obtain multiple initial settlement directories sent by the terminal. The initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled.

The initial settlement directory may be a settlement directory released in various places, such as a social insurance settlement directory, a medical insurance settlement directory, and a drug settlement directory. The data to be settled refers to the settlement information of the items to be settled.

The server can obtain multiple initial settlement directories sent by multiple terminals. The initial settlement directory includes multiple data to be settled, and each piece of data to be settled includes corresponding multiple fields to be settled. For example, the drug settlement directory may include multiple drug settlement data, and each drug settlement data may include multiple field information such as "general name", "dosage form", "specification", and "manufacturer".

Step 204: Obtain a normative data table. The normative data table includes normative settlement data and corresponding normative fields. The normative settlement data is set with settlement attributes and categories.

After the server obtains multiple initial settlement directories, it further obtains the specification data table. The authority data table may be authority settlement data defined in advance according to a preset rule. It may also be that after the server obtains a large amount of settlement data, it performs a big data analysis on the large amount of settlement data, and according to the analysis result and a plurality of standardized settlement data defined by preset rules. The authority data table includes a plurality of authority fields corresponding to authority settlement data. The authority settlement data also includes corresponding settlement attributes and categories.

Step 206: Match the fields to be settled with the specification fields in the specification data table, and calculate the degree of matching between the field to be settled and the plurality of specification fields.

In step 208, if there is a field to be settled whose matching degree reaches a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is standardized settlement data, and the settlement attributes and categories set by the standard settlement data corresponding to the specification field are added to the data to be settled. in.

After the server obtains the specification data table, it matches the to-be-settled fields in the to-be-settled data with the specification fields in the specification-settlement data, and calculates the degree of matching between the to-be-settled fields and multiple specification fields. When the degree of matching between the to-be-settled field in the to-be-settled data and the canonical field in the to-be-settled data reaches a preset threshold, it indicates that the to-be-settled data is normative settlement data, and it is determined that the to-be-settled data corresponding to the to-be-settled field is normative. Settle the data, and add the settlement attributes and categories set by the standard settlement data corresponding to the specification field to the data to be settled. In this way, the standardized settlement data in the data to be settled can be directly identified, and corresponding settlement attributes and categories are added.

In step 210, if there is a field to be settled whose matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is non-standardized settlement data, and to-be-determined settlement data is generated based on the non-standardized to-be-settled data. Make a decision in the trained decision model, output settlement attributes and categories corresponding to the settlement data to be decided, and add it to the settlement data to be decided.

If there is a field to be settled whose matching degree does not reach a preset threshold, it indicates that the data to be settled is non-standardized settlement data, and the non-standardized settlement data needs to be further processed. The server determines that the data to be settled corresponding to the field to be settled is irregular settlement data, and generates the settlement data to be decided according to the non-standardized settlement data. Specifically, when the server matches the field to be settled in the data to be settled with the specification field in the standard settlement data, the server extracts data to be settled that does not reach the preset matching degree, and the data to be settled that does not reach the preset matching degree is Data to be settled that are inconsistent with the standard settlement data.

The server generates the settlement data for decision-making based on the non-standard data to be settled, and then obtains a decision model that has been trained and constructed. The decision model includes multiple nodes, and the data for decision-making is input into the decision model, and the node order of the decision model is followed. Iterate until the settlement attributes and categories corresponding to the settlement data to be determined are obtained. By making decision on the settlement data to be decided according to the decision model, the settlement attributes and categories corresponding to the settlement data to be decided can be accurately determined, and the settlement attributes and categories decided on are added to the settlement data to be decided.

Step 212: Generate a project settlement directory according to the category according to a plurality of settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.

After the server determines the corresponding settlement attribute and category for the decision settlement data according to the decision model, it generates a project settlement directory according to the category based on multiple settlement data after adding settlement attributes and categories. The settlement data after adding the settlement attribute and category includes the pending settlement data with the settlement attribute and category added, and the pending decision settlement data with the settlement attribute and category added.

After the server generates the project settlement directory and forms a standardized project settlement directory, the corresponding project can be settled according to the project settlement directory.

In the method for processing settlement data, the server obtains multiple initial settlement directories sent by the terminal. The initial settlement directory includes multiple data to be settled, and the data to be settled includes corresponding multiple fields to be settled. The server obtains the authority data table, and the authority data table includes a plurality of authority fields corresponding to the authority settlement data. The authority settlement data is set with a settlement attribute and a category. The server matches the field to be settled with the specification field in the specification data table, and calculates a degree of matching between the field to be settled and the plurality of specification fields. If there is a field to be settled whose matching degree reaches a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is standard settlement data, and the settlement attributes and categories set by the standard settlement data corresponding to the standard field are added to the data to be settled. If there is a matching degree that does not reach the preset threshold, it is determined that the data to be settled corresponding to the field to be settled is non-standardized settlement data, and the decision-making settlement data is generated based on the non-standardized data to be settled. The server inputs the settlement data to be decided into a trained decision model for decision making, outputs the settlement attributes and categories corresponding to the settlement data to be decided, and adds it to the settlement data to be decided. The server generates a project settlement directory according to the category based on a plurality of settlement data after adding settlement attributes and categories, so that the project settlement processing can be performed according to the project settlement directory. By matching and classifying the settlement data in the initial settlement directory, the settlement attributes and categories corresponding to the data to be settled can be accurately obtained, and a standardized project settlement directory can be formed, which can effectively improve the processing efficiency of settlement data.

In one embodiment, matching the fields to be settled with the canonical fields in the canonical data table includes: obtaining a semantic matching model, which includes multiple canonical settlement data and corresponding multiple canonical field vectors; through semantic matching The model extracts field vectors corresponding to a plurality of fields to be settled in the data to be settled; and calculates a degree of matching between the field vectors corresponding to the data to be settled and a plurality of canonical field vectors.

The server obtains multiple initial settlement directories. The initial settlement directory includes multiple pending settlement data, and each piece of pending settlement data includes a corresponding multiple pending settlement fields. For example, the drug settlement directory may include multiple drug settlement data, and each drug settlement data may include multiple field information such as "general name", "dosage form", "specification", and "manufacturer".

After the server obtains a plurality of initial settlement directories, it further obtains a specification data table. The specification data table includes a plurality of specification fields corresponding to the specification settlement data, and the specification settlement data is set with settlement attributes and categories. The server obtains the corresponding semantic matching model according to the official data table. The semantic matching model includes a plurality of canonical field vectors corresponding to a plurality of canonical settlement data. The server extracts the field vectors corresponding to the plurality of fields to be settled in the data to be settled, and matches the field vectors corresponding to the data to be settled with the plurality of standard field vectors corresponding to the plurality of standard settlement data to calculate the corresponding The degree of match between the field vector and multiple canonical field vectors. By matching the data to be settled with the standard settlement data, it is possible to effectively match the data to be settled with the standard settlement data, thereby improving the efficiency of further processing of the data to be settled.

For example, taking the drug settlement directory as an example, the drug settlement directory includes multiple drug data, and each drug data includes the corresponding multiple to-be-cleared fields such as "general name", "dosage form", "specification", and "manufacturer". Content, and extract field vectors corresponding to the contents of multiple fields to be settled. After the server obtains the corresponding canonical data table, it further obtains the corresponding semantic matching model. The semantic matching model includes multiple canonical field vectors corresponding to multiple canonical settlement data. The server matches the multiple field vectors corresponding to the drug data with the multiple standard field vectors corresponding to multiple regulatory settlement data, that is, the corresponding "common name", "dosage form", "specification", and "manufacturer" of the pharmaceutical data. The contents of multiple to-be-settled fields are matched with multiple canonical field vectors corresponding to multiple canonical settlement data.

In one embodiment, as shown in FIG. 3, the decision model includes a plurality of nodes, and the steps of inputting settlement data to be decided into the trained decision model for decision specifically include the following:

Step 302: Use a decision model to extract multiple field vectors corresponding to the settlement data to be decided.

In step 304, the multiple field vectors corresponding to the settlement data to be decided are traversed and matched according to the node order of the decision model, and the matching degree between the field vector and the multiple nodes is calculated.

Step 306: Until a plurality of field vectors are matched to obtain the corresponding target settlement attribute and target category, the decision model outputs the settlement attribute and category corresponding to the settlement data to be decided.

After the server obtains multiple initial settlement directories, it obtains a specification data table. The initial settlement directory includes multiple data to be settled, and the data to be settled includes corresponding multiple fields to be settled. The authority data table includes a plurality of authority fields corresponding to authority settlement data, and authority data includes corresponding settlement attributes and categories. The server matches the field to be settled with the specification field, and calculates the matching degree of the field to be settled with multiple specification fields. If there is a field to be settled when the matching degree reaches a preset threshold, it means that the field to be settled is consistent with the specification field, then the data to be settled corresponding to the field to be settled is the standard settlement data, and the standard settlement data corresponding to the specification field is set Added to the pending data.

If there is a field to be settled that does not reach the preset threshold, it indicates that the field to be settled is inconsistent with the standard field, then the data to be settled corresponding to the field to be settled is non-standardized settlement data, and a pending decision is generated based on the non-standardized settled data Billing data. The server further extracts multiple field vectors corresponding to the settlement data to be decided, and obtains a decision model. The decision model includes a decision tree, and a plurality of nodes are set in the decision tree in advance. Multiple field vectors corresponding to the settlement data to be decided are input to the decision model, and the traversal matching is performed according to the order of the nodes in the decision model to calculate the matching degree of the field vector with multiple nodes. The target settlement attribute and the corresponding target category of each field vector are determined according to the matching degree. Until multiple field vectors are matched to obtain the corresponding target settlement attributes and target categories, the server outputs the settlement attributes and categories corresponding to the settlement data to be decided according to the decision model, and adds the corresponding settlement attributes and categories to the settlement data to be decided . By making decisions on the settlement data according to the decision model in the order of nodes, the efficiency of the decision can be improved, and the settlement attributes and categories corresponding to the settlement data to be determined can be effectively and accurately obtained.

In one embodiment, as shown in FIG. 4, before the obtaining a decision model, a step of constructing a decision model is further included. The step specifically includes the following content:

Step 402: Obtain multiple settlement data in multiple databases. The settlement data includes multiple field names and corresponding field values.

Step 404: Perform cluster analysis on the field names of the settlement data to obtain priority parameters for each field name.

Step 406: Calculate the weights of multiple field names according to the priority parameters of the field names.

Step 408: Train the association relationship between the multiple settlement data and the settlement attribute and the corresponding category according to the field names and corresponding field values of the multiple settlement data.

Step 410: Construct a decision tree according to the weights of multiple field names and the correlation between the training settlement data, settlement attributes, and categories, and generate a decision model according to the decision tree.

Before the server obtains the decision model, it needs to build a decision model in advance. Specifically, the server may obtain multiple settlement data from databases of multiple websites, and the obtained settlement data may be an initial settlement directory issued by each place or institution.

The settlement data includes multiple field names and corresponding field values. The server first performs cluster analysis on multiple field names in the settlement data. Specifically, the server analyzes multiple field names in the settlement data to analyze the probability of each field name being used in the settlement field. The more frequently used field names are more important, and according to the field names, Probability gets the priority parameter for each field name. The server further sorts the field names according to the priority probability of each field name. By prioritizing the field names and calculating the weights of multiple field names, the efficiency of decision-making on settlement data can be effectively improved.

The server further trains association relationships between the multiple settlement data and settlement attributes and categories according to the field names and corresponding field values of the multiple settlement data. Specifically, the server may take each settlement data as a sample, take the field name and corresponding field value in each settlement data as one dimension, and each settlement data has multiple dimensions. The server further performs cluster analysis on each dimension of each settlement data. The server can use a cluster analysis algorithm, such as the K-means algorithm, and iteratively calculates multiple samples using multiple dimensions of each settlement data as data objects in order to calculate the clustering result corresponding to each settlement data. For example, the server can obtain multiple dimensional variables corresponding to multiple field names and corresponding field values in the settlement data, and then perform cluster analysis on multiple settlement data samples, thereby analyzing the relationship between settlement data and settlement attributes and categories. Relationship.

Furthermore, the server can construct a decision tree based on the weights of multiple field names and the association between the training settlement data and settlement attributes and categories, and generate a decision model based on the decision tree. By constructing a decision model based on the relationship between the training data of the sorted field names and the settlement attributes and categories, it is possible to effectively improve the efficiency and accuracy of decision-making based on the decision model.

In one embodiment, the method further includes: obtaining a plurality of update settlement data, the update settlement data is set with a settlement attribute and a category; performing cluster analysis on the plurality of update settlement data to obtain the update weights and Update the association relationship parameters; adjust the parameters of the decision model according to the update weight and the update association parameter to optimize the decision model.

With the changes of various factors, the decision model also needs to be adjusted with the changes of various factors and time to improve the stability of the model. After the server constructs a decision model according to the priority probability of the field names of the plurality of settlement data and the training relationship between the settlement data and the settlement attributes and categories, it can further optimize the decision model to make the decision model accurate. higher.

The server may obtain the updated settlement data within a preset time period from the database. The preset time may be one year, or half a year, one quarter, or one month. The updated settlement data may be updated settlement data after updating relevant rules of the settlement data, or may be updated normative settlement data defined in accordance with a preset rule. Settlement attributes and categories are set in the update settlement data. The server further performs cluster analysis on multiple update settlement data to obtain update weights and update association parameters for multiple field names. The server further adjusts the relevant commitments of the decision model according to the update weight and the update relationship parameter to optimize the decision model, which can effectively ensure the stability and newness of the decision model, and can effectively improve the accuracy of classification of settlement data according to the decision model. Sex.

In one of the embodiments, the category includes multiple hierarchical categories. After generating the project settlement directory according to the category based on the multiple settlement attributes and category-added settlement data, the method further includes: according to the settlement attributes of the settlement data and multiple hierarchical category pairs. Sorting settlement data; encoding settlement data of multiple hierarchical categories according to a preset manner; storing the encoded settlement directory.

After the server generates the project settlement directory according to the category based on the multiple settlement data after adding the settlement attribute and category, the server sorts the settlement data according to the settlement attribute and category of the settlement data. The category of settlement data may also include multiple hierarchical categories, and the server sorts the settlement data in the project settlement directory according to the corresponding category or hierarchical category. After the server sorts the project settlement directory, it encodes the settlement data of each hierarchical category according to a preset method, and stores the encoded project settlement directory. Among them, the coded characters include at least two types. Coded characters can be letters and numbers, such as ABS10001, ABS100012. Sub-characters are added in sequence to the settlement data under each hierarchical category in accordance with the corresponding encoded characters. For example ABS10001001, ABS10001002. By encoding the detailed classified item settlement catalog data, the management efficiency of settlement data is effectively improved.

It should be understood that although the steps in the flowcharts of FIGS. 2-4 are sequentially displayed in accordance with the directions of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless explicitly stated in this document, the execution of these steps is not strictly limited, and these steps can be performed in other orders. Moreover, at least a part of the steps in Figure 2-4 may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. These sub-steps or stages The execution order of is not necessarily performed sequentially, but may be performed in turn or alternately with at least a part of another step or a sub-step or stage of another step.

In one embodiment, as shown in FIG. 5, a settlement data processing device is provided, which includes: an acquisition module 502, a matching module 504, a decision module 506, and a directory generation module 508, where:

The obtaining module 502 is configured to obtain multiple initial settlement directories sent by the terminal. The initial settlement directory includes multiple pending settlement data and corresponding multiple pending settlement fields; and obtains a settlement data table, where the settlement data table includes standardized settlement data and corresponding Multiple specification fields. Settlement attributes and categories are set in the standard settlement data;

A matching module 504 is configured to match a field to be settled with a standard field in a specification data table, and calculate a degree of matching between the field to be settled and a plurality of standard fields; if there is a field to be settled with a matching degree that reaches a preset threshold, determine a field to be settled The data to be settled corresponding to the field is standard settlement data, and the settlement attributes and categories set by the standard settlement data corresponding to the standard field are added to the data to be settled;

A decision module 506, configured to determine that the pending data corresponding to the pending settlement field is non-standard settlement data if there is a pending field that does not reach a preset threshold, and generate pending decision settlement data based on the non-standard pending settlement data, and Settlement data is input into the trained decision model for decision-making, output settlement attributes and categories corresponding to the settlement data to be decided, and added to the settlement data to be decided;

The directory generation module 508 is configured to generate a project settlement directory according to categories based on a plurality of settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.

In one embodiment, the matching module 504 is further configured to obtain a semantic matching model. The semantic matching model includes multiple canonical settlement data and corresponding multiple canonical field vectors. A plurality of to-be-settled data in the to-be-settled data are extracted through the semantic matching model. A field vector corresponding to the settlement field; and calculating a matching degree between the field vector corresponding to the data to be settled and a plurality of canonical field vectors.

In one embodiment, the decision model includes multiple nodes, and the decision module 506 is further configured to use the decision model to extract multiple field vectors corresponding to the settlement data to be decided; Nodes are traversed and matched sequentially to calculate the matching degree between the field vector and multiple nodes; and until the multiple field vectors are matched to obtain the corresponding target settlement attributes and target categories, the decision model outputs the settlement attributes and categories corresponding to the settlement data to be decided.

In one of the embodiments, the device further includes a model building module for obtaining multiple settlement data in multiple databases, the settlement data includes multiple field names and corresponding field values; and performs cluster analysis on the field names of the settlement data To obtain the priority parameters of each field name; calculate the weight of multiple field names according to the priority parameters of the field names; train multiple settlement data and settlement attributes and corresponding categories based on the field names and corresponding field values of multiple settlement data The association relationship between them; and constructing a decision tree based on the weights of multiple field names and the association relationship between the training settlement data and settlement attributes and categories, and generating a decision model based on the decision tree.

In one of the embodiments, the device further includes a model optimization module for obtaining and acquiring multiple updated settlement data, the updated settlement data is set with a settlement attribute and category; cluster analysis is performed on the multiple updated settlement data to obtain multiple Update weights of field names and update association relationship parameters; and adjust parameters of the decision model according to the update weights and update association relationship parameters to optimize the decision model.

In one embodiment, the category includes multiple hierarchical categories, and the directory generation module 508 is further configured to sort the settlement data according to the settlement attributes and multiple hierarchical categories of the settlement data; and to settle the settlement data of the multiple hierarchical categories in a preset manner Encode; and store the encoded settlement directory.

For the specific limitation of the settlement data processing device, refer to the foregoing limitation on the settlement data processing method, which is not repeated here. Each module in the above-mentioned settlement data processing device may be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules may be embedded in the hardware form or independent of the processor in the computer device, or may be stored in the memory of the computer device in the form of software, so that the processor calls and performs the operations corresponding to the above modules.

In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. 6. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer-readable instructions, and a database. The internal memory provides an environment for operating the operating system and computer-readable instructions in a non-volatile storage medium. The computer equipment database is used to store the initial settlement data directory, the standard settlement data table, and the project settlement data directory. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer-readable instructions are executed by a processor to implement a settlement data processing method.

Those skilled in the art can understand that the structure shown in FIG. 6 is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer equipment to which the scheme of the present application is applied. The specific computer equipment may be Include more or fewer parts than shown in the figure, or combine certain parts, or have a different arrangement of parts.

A computer device includes a memory and one or more processors. Computer-readable instructions are stored in the memory. When the computer-readable instructions are executed by the processor, the one or more processors execute the following steps:

Obtaining multiple initial settlement directories sent by the terminal, the initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled;

Obtaining the authority data table, which includes authority settlement data and corresponding authority fields, and authority properties are set in the authority settlement data;

Match the field to be settled with the specification field in the specification data table, and calculate the degree of matching between the field to be settled and multiple specification fields;

If there is a field to be settled whose matching degree reaches a preset threshold, determine that the data to be settled corresponding to the field to be settled is standard settlement data, and add the settlement attributes and categories set by the standard settlement data corresponding to the standard field to the data to be settled;

If there is a field to be settled that does not reach a preset threshold, determine that the data to be settled corresponding to the field to be settled is non-standardized settlement data, generate settlement data for decision-making based on the non-standardized settlement data, and enter the settlement data for decision-making into the trained Make a decision in the decision model of the company, output the settlement attributes and categories corresponding to the settlement data to be decided, and add it to the settlement data to be decided; and

In one of the embodiments, when the processor executes the computer-readable instructions, the processor further implements the following steps: obtaining a semantic matching model, the semantic matching model including multiple canonical settlement data and corresponding multiple canonical field vectors; and extracting the data through the semantic matching model Field vectors corresponding to a plurality of fields to be settled in the data to be settled; and a degree of matching between a field vector corresponding to the data to be settled and a plurality of canonical field vectors.

In one embodiment, when the processor executes the computer-readable instructions, the processor further implements the following steps: using a decision model to extract multiple field vectors corresponding to the settlement data to be decided; and dividing the multiple field vectors corresponding to the settlement data to be decided according to the decision model. Nodes are traversed and matched sequentially to calculate the matching degree between the field vector and multiple nodes; and until the multiple field vectors are matched to obtain the corresponding target settlement attributes and target categories, the decision model outputs the settlement attributes and categories corresponding to the settlement data to be decided.

In one of the embodiments, when the processor executes the computer-readable instructions, the processor further implements the following steps: obtaining multiple settlement data in multiple databases, the settlement data including multiple field names and corresponding field values; performing field names on the settlement data Cluster analysis to obtain the priority parameters of each field name; calculate the weight of multiple field names based on the priority parameters of the field names; train multiple settlement data and settlement attributes based on the field names and corresponding field values of multiple settlement data Association with corresponding categories; and constructing a decision tree based on the weights of multiple field names and the association between training data and settlement attributes and categories, and generating a decision model based on the decision tree.

In one of the embodiments, when the processor executes the computer-readable instructions, the processor further implements the following steps: acquiring multiple updated settlement data, the updated settlement data is set with a settlement attribute and category; and performing cluster analysis on the multiple updated settlement data to obtain Update weights and update association relationship parameters of multiple field names; and adjust parameters of the decision model according to the update weights and update association relationship parameters to optimize the decision model.

In one embodiment, the category includes multiple hierarchical categories. When the processor executes the computer-readable instructions, the following steps are further implemented: sorting the settlement data according to the settlement attributes of the settlement data and multiple hierarchical categories; Encoding settlement data of each hierarchical category; and storing the encoded settlement directory.

One or more non-volatile computer-readable storage media storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the following steps:

In one of the embodiments, when the computer-readable instructions are executed by the processor, the following steps are further implemented: obtaining a semantic matching model, the semantic matching model including multiple canonical settlement data and corresponding multiple canonical field vectors; and extracting through the semantic matching model A field vector corresponding to a plurality of fields to be settled in the data to be settled is calculated; and a matching degree between a field vector corresponding to the data to be settled and a plurality of canonical field vectors is calculated.

In one embodiment, when the computer-readable instructions are executed by the processor, the following steps are further implemented: using a decision model to extract multiple field vectors corresponding to the settlement data to be decided; and using the multiple field vectors corresponding to the settlement data to be decided according to the decision model The nodes are traversed and matched sequentially to calculate the matching degree between the field vector and multiple nodes; and until the multiple field vectors are matched to obtain the corresponding target settlement attributes and target categories, the settlement attributes and categories corresponding to the settlement data to be decided are output through the decision model.

In one embodiment, when the computer-readable instructions are executed by the processor, the following steps are further implemented: obtaining multiple settlement data in multiple databases, the settlement data including multiple field names and corresponding field values; and field names of the settlement data Perform cluster analysis to obtain the priority parameters of each field name; calculate the weight of multiple field names based on the priority parameters of the field names; train multiple settlement data and settlements based on the field names and corresponding field values of multiple settlement data Association between attributes and corresponding categories; and constructing a decision tree based on the weights of multiple field names and the association between training data and settlement attributes and categories, and generating a decision model based on the decision tree.

In one of the embodiments, when the computer-readable instructions are executed by the processor, the following steps are also implemented: obtaining multiple updated settlement data, the updated settlement data is set with a settlement attribute and category; performing cluster analysis on the multiple updated settlement data, Obtain the update weights and update association relationship parameters of multiple field names; and adjust the parameters of the decision model according to the update weights and update association relationship parameters to optimize the decision model.

In one of the embodiments, the category includes multiple hierarchical categories, and when the computer-readable instructions are executed by the processor, the following steps are further implemented: sorting the settlement data according to the settlement attributes of the settlement data and multiple hierarchical categories; Encoding settlement data for multiple hierarchical categories; and storing the encoded settlement directory.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by computer-readable instructions to instruct related hardware. The computer-readable instructions can be stored in a non-volatile computer. In the readable storage medium, the computer-readable instructions, when executed, may include the processes of the embodiments of the methods described above. Wherein, any reference to the memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined. In order to make the description concise, all possible combinations of the technical features in the above embodiments have not been described. However, as long as there is no contradiction in the combination of these technical features, they should be It is considered to be the range described in this specification.

The above-mentioned embodiments only express several implementation manners of the present application, and their descriptions are more specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be noted that, for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of this application patent shall be subject to the appended claims.

Claims

A method for processing settlement data, including:

Acquiring multiple initial settlement directories sent by the terminal, where the initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled;

Obtaining a authority data table, where the authority data table includes authority settlement data and corresponding authority fields, and the authority settlement data is set with settlement attributes and categories;

Matching the field to be settled with a specification field in the specification data table, and calculating a degree of matching between the field to be settled and the plurality of specification fields;

If there is a field to be settled where the matching degree reaches a preset threshold, determine that the data to be settled corresponding to the field to be settled is standard settlement data, and add the settlement attributes and categories set by the standard settlement data corresponding to the standard field to In the data to be settled;

If there is a field to be settled where the matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is irregular settlement data, and to-be-determined settlement data is generated based on the non-standard to-be-settled data, and Settlement data is input into a trained decision model for decision making, outputting settlement attributes and categories corresponding to the pending decision settlement data, and added to the pending decision settlement data; and

Generate a project settlement directory according to the category based on multiple settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.
The method according to claim 1, wherein the matching the field to be settled with a specification field in the specification data table comprises:

Obtaining a semantic matching model, where the semantic matching model includes multiple canonical settlement data and corresponding multiple canonical field vectors;

Extracting a field vector corresponding to a plurality of fields to be settled in the data to be settled through the semantic matching model; and

Calculate a matching degree between a field vector corresponding to the data to be settled and a plurality of canonical field vectors.
The method according to claim 1, wherein the decision model includes a plurality of nodes, and the step of inputting the settlement data to be decided into a trained decision model for decision making comprises:

Using the decision model to extract a plurality of field vectors corresponding to the settlement data to be decided;

Performing traversal matching on the plurality of field vectors corresponding to the settlement data to be decided according to the node order of the decision model, and calculating the degree of matching between the field vector and the plurality of nodes; and

Until the plurality of field vectors are matched to obtain the corresponding target settlement attribute and target category, the decision model outputs the settlement attribute and category corresponding to the settlement data to be decided.
The method according to claim 1, before the obtaining a decision model, further comprising:

Obtaining multiple settlement data in multiple databases, where the settlement data includes multiple field names and corresponding field values;

Performing cluster analysis on the field names of the settlement data to obtain priority parameters for each field name;

Calculating weights of multiple field names according to the priority parameters of the field names;

Training association relationships between multiple settlement data and settlement attributes and corresponding categories according to field names and corresponding field values of the multiple settlement data; and

A decision tree is constructed according to the weights of multiple field names and the association relationship between the training settlement data and settlement attributes and categories, and a decision model is generated according to the decision tree.
The method according to claim 1, further comprising:

Obtaining a plurality of updated settlement data, wherein the updated settlement data is set with a settlement attribute and a category;

Perform cluster analysis on the multiple update settlement data to obtain update weights and update association parameters of multiple field names; and

The parameters of the decision model are adjusted according to the update weight and the update association parameter to optimize the decision model.
The method according to claim 1, wherein the category comprises a plurality of hierarchical categories, and after the generating a project settlement directory according to the category based on a plurality of settlement attributes after adding settlement attributes and categories, further comprising:

Sorting the settlement data according to the settlement attributes and multiple hierarchical categories of the settlement data;

Encoding settlement data for multiple tier categories in a preset manner; and

Store the encoded settlement directory.
A settlement data processing device includes:

An obtaining module, configured to obtain multiple initial settlement directories sent by the terminal, where the initial settlement directories include multiple data to be settled, the data to be settled includes corresponding multiple fields to be settled; and a settlement data table, the settlement data The table includes a plurality of specification fields corresponding to the specification settlement data, and the specification settlement data is set with settlement attributes and categories;

A matching module, configured to compare the field to be settled with a specification field in the specification data table, and calculate a degree of matching between the field to be settled and the plurality of specification fields; Setting a threshold to-be-settled field, determining that the to-be-settled data corresponding to the to-be-settled field is normative settlement data, and adding the settlement attribute and category set by the normative settlement data corresponding to the specification field to the to-be-settled data;

A decision module, configured to determine that the pending data corresponding to the pending settlement field is non-standard settlement data if there is a pending field to be settled where the matching degree does not reach a preset threshold, and generate the pending decision settlement data according to the non-standard pending settlement data; Inputting the settlement data to be decided into a trained decision model for decision making, outputting settlement attributes and categories corresponding to the settlement data to be decided, and adding to the settlement data to be decided; and

The directory generating module is configured to generate a project settlement directory according to the category based on a plurality of settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.
The device according to claim 7, wherein the matching module is further configured to obtain a semantic matching model, wherein the semantic matching model includes a plurality of canonical settlement data and a corresponding plurality of canonical field vectors; The matching model extracts field vectors corresponding to a plurality of fields to be settled in the data to be settled; and calculates a degree of matching between a field vector corresponding to the data to be settled and a plurality of canonical field vectors.
The device according to claim 7, wherein the decision model includes a plurality of nodes, and the decision module is further configured to use the decision model to extract multiple field vectors corresponding to the settlement data to be decided; The multiple field vectors corresponding to the settlement data to be determined are traversed and matched according to the node order of the decision model, and the degree of matching between the field vector and the multiple nodes is calculated; and until the multiple field vectors are matched to obtain the corresponding target settlement attributes and The target category outputs the settlement attribute and category corresponding to the settlement data to be decided through the decision model.
The device according to claim 7, wherein the device further comprises a model building module, configured to obtain multiple settlement data in multiple databases, where the settlement data includes multiple field names and corresponding field values; Perform cluster analysis on the field names of the settlement data to obtain priority parameters for each field name; calculate the weight of multiple field names according to the priority parameters of the field names; and according to the field names and corresponding fields of the multiple settlement data Value training out the correlation between multiple settlement data and settlement attributes and corresponding categories; and constructing a decision tree based on the weights of multiple field names and the correlation between the training settlement data and settlement attributes and categories, and according to the Describe the decision tree generation decision model.
The device according to claim 7, characterized in that the device further comprises a model optimization module for obtaining a plurality of updated settlement data, the updated settlement data is set with a settlement attribute and a category; and the plurality of updates The settlement data is subjected to cluster analysis to obtain update weights and update association relationship parameters of multiple field names; and adjust parameters of the decision model according to the update weights and update association relationship parameters to optimize the decision model.
The device according to claim 7, wherein the category includes multiple hierarchical categories, and the directory generation module is further configured to sort the settlement data according to the settlement attributes of the settlement data and multiple hierarchical categories. ; Encoding settlement data of multiple hierarchical categories in a preset manner; and storing the encoded settlement directory.
A computer device includes a memory and one or more processors. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the one or more processors are caused. Each processor performs the following steps:

Acquiring multiple initial settlement directories sent by the terminal, where the initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled;

Obtaining a authority data table, where the authority data table includes authority settlement data and corresponding authority fields, and the authority settlement data is set with settlement attributes and categories;

Matching the field to be settled with a specification field in the specification data table, and calculating a degree of matching between the field to be settled and the plurality of specification fields;

If there is a field to be settled where the matching degree reaches a preset threshold, determine that the data to be settled corresponding to the field to be settled is standard settlement data, and add the settlement attributes and categories set by the standard settlement data corresponding to the standard field to In the data to be settled;

If there is a field to be settled where the matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is irregular settlement data, and to-be-determined settlement data is generated based on the non-standard to-be-settled data, and Settlement data is input into a trained decision model for decision making, outputting settlement attributes and categories corresponding to the pending decision settlement data, and added to the pending decision settlement data; and

Generate a project settlement directory according to the category based on multiple settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.
The computer device according to claim 13, wherein when the processor executes the computer-readable instructions, the processor further executes the following steps: acquiring a semantic matching model, the semantic matching model comprising a plurality of normative settlement data and correspondences A plurality of canonical field vectors; extracting a field vector corresponding to a plurality of to-be-cleared fields in the data to be settled through the semantic matching model; and calculating a field vector corresponding to the data to be settled and a plurality of canonical field vectors Match.
The computer device according to claim 13, wherein the decision model includes a plurality of nodes, and the processor further executes the following steps when executing the computer-readable instructions: using the decision model to extract the pending decision Multiple field vectors corresponding to the settlement data; performing traversal matching on the multiple field vectors corresponding to the settlement data to be determined according to the node order of the decision model, and calculating the degree of matching between the field vector and the multiple nodes; and up to the multiple The field vectors are matched to obtain the corresponding target settlement attribute and target category, and the settlement attribute and category corresponding to the settlement data to be decided are output through the decision model.
The computer device according to claim 13, wherein when the processor executes the computer-readable instructions, the processor further performs the following steps: obtaining multiple settlement data in multiple databases, the settlement data including multiple fields Cluster name and corresponding field value; perform cluster analysis on the field names of the settlement data to obtain priority parameters for each field name; calculate the weights of multiple field names according to the priority parameters of the field names; and based on multiple settlements The field names of the data and the corresponding field values train the association relationships between multiple settlement data and settlement attributes and corresponding categories; and the association relationships between the training field settlement data and settlement attributes and categories according to the weights of multiple field names A decision tree is constructed, and a decision model is generated based on the decision tree.
One or more non-transitory computer-readable storage media storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the following steps:

Acquiring multiple initial settlement directories sent by the terminal, where the initial settlement directory includes multiple data to be settled and corresponding multiple fields to be settled;

Obtaining a authority data table, where the authority data table includes authority settlement data and corresponding authority fields, and the authority settlement data is set with settlement attributes and categories;

Matching the field to be settled with a specification field in the specification data table, and calculating a degree of matching between the field to be settled and the plurality of specification fields;

If there is a field to be settled where the matching degree reaches a preset threshold, determine that the data to be settled corresponding to the field to be settled is standard settlement data, and add the settlement attributes and categories set by the standard settlement data corresponding to the standard field to In the data to be settled;

If there is a field to be settled where the matching degree does not reach a preset threshold, it is determined that the data to be settled corresponding to the field to be settled is irregular settlement data, and to-be-determined settlement data is generated based on the non-standard to-be-settled data, and Settlement data is input into a trained decision model for decision making, outputting settlement attributes and categories corresponding to the pending decision settlement data, and added to the pending decision settlement data; and

Generate a project settlement directory according to the category based on multiple settlement data after adding settlement attributes and categories, so that the server performs project settlement processing according to the project settlement directory.
The storage medium according to claim 17, wherein when the computer-readable instructions are executed by the processor, the following steps are further performed: obtaining a semantic matching model, the semantic matching model comprising a plurality of normative settlement data and Corresponding multiple canonical field vectors; extracting field vectors corresponding to multiple to-be-cleared fields in the to-be-settled data through the semantic matching model; and calculating a field vector corresponding to the to-be-settled data and a plurality of canonical field vectors Match.
The storage medium according to claim 17, wherein the decision model includes a plurality of nodes, and when the computer-readable instructions are executed by the processor, the following step is further performed: using the decision model to extract the pending Multiple field vectors corresponding to the decision settlement data; performing traversal matching on the multiple field vectors corresponding to the settlement data to be decided according to the node order of the decision model, and calculating the degree of matching between the field vector and the multiple nodes; and until the A plurality of field vectors are matched to obtain a corresponding target settlement attribute and target category, and the settlement attribute and category corresponding to the settlement data to be decided are output through the decision model.
The storage medium according to claim 17, wherein when the computer-readable instructions are executed by the processor, the following steps are further performed: acquiring multiple settlement data in multiple databases, and the settlement data includes multiple Field names and corresponding field values; performing cluster analysis on the field names of the settlement data to obtain priority parameters for each field name; calculating weights of multiple field names according to the priority parameters of the field names; and The field names and corresponding field values of the settlement data train the association relationships between multiple settlement data and settlement attributes and corresponding categories; and the association between the training field settlement data and settlement attributes and categories according to the weights of multiple field names The relationship builds a decision tree, and generates a decision model based on the decision tree.