CN111159285A

CN111159285A - Enterprise cross-system retrieval method based on distributed index service deployment

Info

Publication number: CN111159285A
Application number: CN201911235726.2A
Authority: CN
Inventors: 董秋艳; 孟祥鑫; 许瑞; 韩皓睿; 范杰; 权志旺
Original assignee: Beijing Research Institute of Mechanical and Electrical Technology
Current assignee: Beijing Research Institute of Mechanical and Electrical Technology
Priority date: 2019-12-05
Filing date: 2019-12-05
Publication date: 2020-05-15
Anticipated expiration: 2039-12-05
Also published as: CN111159285B

Abstract

The invention discloses an enterprise cross-system retrieval method based on distributed index service deployment, belongs to the technical field of information retrieval, and can support retrieval requirements of enterprise cross-service systems, support a data processing and merging mechanism after new addition, update and deletion of data information, and support independent search and combined search aiming at each service system. Setting a data source to be indexed in each service system according to requirements, wherein the data source comprises a document and a database, taking a server where the service system is located as an index server, creating an index base by adopting the sequence of index server deployment, data source registration, index configuration and index combination aiming at any service system, then creating a retrieval condition, retrieving an index file in the index base of the service system by adopting the retrieval condition, returning a retrieval result and displaying.

Description

Enterprise cross-system retrieval method based on distributed index service deployment

Technical Field

The invention relates to the technical field of information retrieval, in particular to an enterprise cross-system retrieval method based on distributed index service deployment.

Background

Each business activity in the enterprise product development process runs through the processing, application and innovation of information such as design experience, principle, specification and the like, and a large number of information resources such as development cases, successful experience, design methods and the like are deposited and accumulated in the development process. With the increasing information level of modern enterprises, the information resources are widely existed in various business systems in the enterprises in a non-structured document data or structured data mode, but most of the business systems are mutually independent, a large amount of information is in a discrete and isolated state, the information is mainly manually entered into each business system to be collected and analyzed to form available information resources which are applied to the development process, the information acquisition efficiency is low, the sharing is difficult, the flowing is difficult, and the working efficiency is greatly restricted. How to rapidly, accurately and cross-system acquire information resources in an enterprise, exert the advantages and capabilities of the enterprise and strengthen the accumulation and precipitation of knowledge assets of the enterprise becomes a problem which needs to be solved urgently. Therefore, it is necessary to form an enterprise cross-system retrieval method, so as to implement unified retrieval of information resources in each service system inside an enterprise, improve the utilization rate of resources inside the enterprise, and further improve the product development efficiency and comprehensive competitiveness of the enterprise.

Chinese patent: (patent No. CN201710899236.7) discloses a cross-system multi-dimensional data retrieval processing method and device, the method comprising: acquiring a plurality of types of data in a database of each system; sorting the plurality of types of data according to a preset dimension to obtain summarized data of the preset dimension; and retrieving the summarized data according to preset screening parameters to obtain target retrieval data. The device comprises: the device comprises a data acquisition module, a data sorting module and a data retrieval module. According to the embodiment of the invention, the plurality of types of data in each system database are sorted to obtain the summarized data with the preset dimensionality, and the summarized data are directly retrieved, so that the times of data retrieval and query are reduced, excessive intermediate temporary results are avoided, the system resource occupation is less, the interface response is fast, and the system expansion is facilitated.

Chinese patent: (patent No. CN201910622980.1) discloses an information retrieval method and system. The information retrieval method and the system provided by the invention firstly calculate the relevance of each webpage document in the keyword set to be searched and the webpage document set of the data source to be searched in the field of national defense science and technology information; and then outputting the webpage documents with the relevance larger than or equal to the similarity threshold, and outputting the webpage documents with the relevance smaller than the similarity threshold in the order from high to low according to the time sequence. The retrieval method and the retrieval system provided by the invention output the webpage documents with larger relevance as the retrieval result, can ensure the coverage rate of the retrieval result, and simultaneously output the webpage documents with smaller relevance to the user according to the sequence of the webpage documents from high to low, and can meet the requirements of high timeliness and high coverage rate of information retrieval.

The existing method does not support distributed deployment of index services of all service systems, does not consider a data processing and merging mechanism after data information of all service systems is newly added, updated and deleted, cannot carry out independent search and combined search of all service systems through scheduling strategy configuration, cannot flexibly carry out registration, index and retrieval on the newly added service systems, and cannot conveniently and quickly complete cross-system retrieval related configuration. Thus, the above-described prior art methods fail to meet the need for cross-system search that can be dynamically configured by an enterprise.

Disclosure of Invention

In view of this, the invention provides an enterprise cross-system retrieval method based on distributed index service deployment, which can support retrieval requirements of enterprise cross-service systems, support a data processing and merging mechanism after new addition, update and deletion of data information, and support independent search and combined search for each service system.

In order to achieve the purpose, the technical scheme of the invention is as follows: the method is used for performing cross-system retrieval aiming at enterprise business systems, and the method performs cross-system retrieval aiming at more than two business systems, and comprises the following steps:

step one, setting a data source to be indexed in each service system according to requirements, wherein the data source comprises a document and a database, taking a server where the service system is located as an index server, and aiming at any service system, adopting the following S101-S104 to create an index database:

s101, deploying on the index server as follows:

and setting a deployment directory, wherein the deployment directory is a storage path of the index service of the business system.

And configuring an index path, wherein the index path is a storage position of an index file provided by the index service of the business system.

And defining an index configuration file, wherein the index configuration file is in a script form of a document to be indexed and a database in the service system.

S102, constructing a database link pool and a file cabinet; registering link resources of a database to be indexed in a service system in a database link pool; and registering the document to be indexed in the service system in the file cabinet.

S103, configuring index fields, wherein the index fields comprise database index fields and document index fields.

And the data index field is used for indexing the registration database in the database connection pool.

And the document index field is used for indexing the registered documents in the file cabinet.

The index field and the corresponding index configuration file form an index file and are stored in the index path.

And S104, creating a full index file for all the registered databases and all the registered documents in the database link pool and the file cabinet through full index.

And creating an incremental index file for the newly added database or the newly added document, the database update or the document update.

And merging the full index file and the incremental index file, and eliminating repeated items to form an index base of the service system.

And step two, creating a retrieval condition, retrieving the index file in an index library of the service system by adopting the retrieval condition, returning a retrieval result, and displaying.

Further, the service system comprises a product data management system PDM, a test data management TDM, a quality management system, an intelligence bus system and a collaborative design system.

Further, in S101, the index configuration file includes the following attributes: the index server unique identification and the index server type;

further, the data index field includes database type, query method, lookup table, query view, query function, and indexed database field.

The document index field includes the operating system of the document store, the file server directory, the file type, and the document class or name of the index.

Further, the retrieval condition is a data keyword, the retrieval condition is adopted to retrieve the index file in the index database of the service system, and the retrieval result is returned, specifically:

setting the retrieval priority as PDM system index library, collaborative design system index library, TDM system index library, quality management system index library and information bus system index library.

And comparing the data keywords with the index files in the index base of each service system in sequence according to the retrieval priority.

And taking the index file with the data keyword similarity exceeding the set threshold as a retrieval result, and returning the retrieval result.

Has the advantages that:

the invention solves the problem of difficult cross-system information retrieval caused by a heterogeneous information storage mode of multiple systems in an enterprise through an enterprise cross-system information retrieval method deployed based on distributed index service, provides convenient cross-system information retrieval service for business systems of PDM, TDM, quality management, information bus, collaborative design and the like of the enterprise, can support the retrieval requirement of the enterprise cross-business systems, supports data processing and merging base addresses after data information is newly added, updated and deleted, supports independent search and combined search aiming at each business system, can promote the flow and integration of heterogeneous information in the enterprise, improves the utilization efficiency of resource information in the enterprise, supports the accumulation and precipitation of enterprise knowledge assets, and further effectively improves the research and development quality and efficiency of enterprise products.

Drawings

FIG. 1 is a schematic diagram of a distributed index service deployment-based enterprise cross-system retrieval method.

Fig. 2 is a diagram of a data source registration implementation process in an embodiment of the present invention.

Fig. 3 is a flowchart illustrating data source registration according to an embodiment of the present invention.

FIG. 4 is a diagram illustrating an implementation process of index configuration according to an embodiment of the present invention.

FIG. 5 is a flow chart of index configuration according to an embodiment of the present invention.

FIG. 6 is a diagram illustrating an implementation process of index merging according to an embodiment of the present invention.

FIG. 7 is a flowchart illustrating index merge according to an embodiment of the present invention.

Fig. 8 is a diagram of a process for implementing unified search in the embodiment of the present invention.

FIG. 9 is a flowchart illustrating a unified search procedure according to an embodiment of the present invention.

Detailed Description

The invention is described in detail below by way of example with reference to the accompanying drawings.

The invention provides an enterprise cross-system retrieval method based on distributed index service deployment, which is used for performing cross-system retrieval aiming at an enterprise business system.

The principle of the retrieval method is shown in fig. 1, namely, a distributed index library with a uniform format is formed by index server deployment, data source registration, index configuration and index merging under the coordination of security management and system control. Enterprise users acquire retrieval information from the index database according to a certain strategy and algorithm through unified retrieval, and the goal of cross-system retrieval is achieved.

The method specifically comprises the following steps:

s101, deploying on the index server as follows:

and setting a deployment directory, wherein the deployment directory is a storage path of the index service of the business system. In the embodiment of the invention, index services are established for a PDM system, a TDM system, a Quality management system, an Intelligence eighty system and a collaborative design system, and the index services of each business system are provided with deployment catalogs, namely the deployment catalogs of the PDM system index service D: \ PDM, the TDM deployment catalogs D: \ TDM, the Quality management system deployment catalogs D: \ Quality, the Intelligence bus system deployment catalogs D: \ Intelligence and the collaborative design system deployment catalogs D: \ Laboration.

Configuring an index path, wherein the index path is a storage position of an index file provided by an index service of a service system; the embodiment of the invention comprises a PDM index storage directory D of \ PDM CRS _ HOME \ data \ index, a TDM index storage directory D of \ TDM CRS _ HOME \ data \ index, a Quality management system index storage directory D of \ Quality CRS _ HOME \ data \ index, an information bus system index storage directory D of \ Intelligent CRS _ HOME \ data \ index and a collaborative design system index storage directory D of \ colletion CRS _ HOME \ data \ index.

Defining an index configuration file, wherein the index configuration file is in a script form of a document to be indexed and a database in a service system; the method comprises the steps of configuring folders and databases needing to be indexed in each business system in an xml file (namely an index configuration file) in a script mode, and finally forming data indexes of each business system by an index server according to the index configuration file, wherein key attributes of the index configuration file comprise unique identification of the index server and the type of the index server, and the type of the index server comprises a database, a folder, a database and a folder and the like.

The data source registration comprises database connection pool index management and data source file cabinet index management, and covers the data source registration of a PDM, a TDM, a quality management system, an intelligence bus system and a collaborative design system, and the registration implementation process is shown in figure 2.

The index data source link pool management is used for managing database link resources of PDM, TDM, quality management system, intelligence bus system and collaborative design system, the database link pool supports link management of Oracle, SQL Server and MySQL databases, wherein the databases registered in the link pool can be used as data sources of retrieval service, and the unregistered databases are not used as data sources of search service, thereby effectively controlling and managing the index range of the retrieval service.

The data source file cabinet index management is used for managing document data sources of PDM, TDM, SDM, quality management system, information bus system and collaborative design system, wherein file cabinets registered in a file cabinet management pool can be used as data sources of retrieval services, file cabinets not registered in the file cabinet management pool are not used as data sources of the retrieval services, and therefore the index range of the search services is effectively controlled and managed.

The data source of the cross-system information retrieval comprises a database of the service system and a file cabinet thereof, and the data source of the cross-system information retrieval is registered and managed through the data source. According to the business requirement, engineers can add and delete data sources, register database connection and file cabinet connection are respectively completed in data source registration, a system checks the uniqueness of the data sources, and if the data sources already exist, the system returns to a main interface; if the data source is the newly added item, the registration is successful, and the data source registration flow is as shown in fig. 3.

The index field and the corresponding index configuration file form an index file and are stored in the index path; each registration database or registration document corresponds to an index field and an index configuration file.

The index field is a basic element formed by the contents of the index library, the contents of the index library are organized according to the index field, different index fields are configured according to the data characteristics of each service system to support the retrieval application requirements of the user, and the index configuration implementation process is shown in fig. 4.

The database index field configures the database fields that define the database type, query method, look-up table/view/function, and index.

The document index field is configured to define the operating system, file server directory, file type, and file class or name of the index for which the document is stored.

According to business needs, index fields can be added or deleted in the index configuration, the system checks the uniqueness of the fields, if the fields exist, the addition is successful, if the fields do not exist, the main interface is returned, and the index configuration flow is shown in fig. 5.

And creating an incremental index file for the newly added database or the newly added document, the database update or the document update. And merging the full index file and the incremental index file, and eliminating repeated items to form an index base of the service system.

The index merging is to form a cross-system index library through full index, incremental index and index merging, and the implementation process is shown in fig. 6. When the index is initialized, the index service creates full index files of a database and a file cabinet through full indexes; when a data source has a new item or an updated item, the index service creates an incremental index file of a database and a file cabinet through an incremental index, the full index file and the incremental index file are respectively stored in different directories for management, when the server is idle, the incremental index file is compared with the full index file according to a relevant strategy set by time, the existing index of the full index file is deleted, the incremental index is merged into the full index file, and the index merging process is shown in fig. 7.

In the embodiment of the present invention, the retrieval condition is a data keyword, the retrieval condition is adopted to retrieve the index file in the index database of the service system, and the retrieval result is returned, specifically:

Through a unified cross-system information retrieval entry, independent search and combined search of each service system are supported, information sources are identified in retrieval results, meanwhile, a personalized retrieval setting method is provided, the retrieval results are sorted and output according to setting conditions, and the process of realizing unified retrieval is shown in an attached figure 8.

When the index server is deployed, defining deployment directories, index file storage directories and index configuration files of the index server of each service system; in the data source registration, all database link strings and file cabinet link strings in a database link pool are acquired; in index configuration, acquiring index field configuration items related to database links and index field configuration items related to file cabinets; forming an index file library of each service system through index merging; in a dialog box of unified retrieval, a user inputs key information of retrieval, a Chinese word segmentation tool is used for realizing the decomposition of retrieval demand information, irrelevant information is removed, key information is reserved to form a retrieval expression, a sequential scheduling strategy is matched in database retrieval scheduling, the retrieval expression is sent to a corresponding index base for retrieval according to the scheduling strategy, a default preferential calling entry is a PDM system index base, the collaborative design system, the TDM system, the quality management system and the information bus system index base are retrieved in sequence, retrieval results are sequenced and then returned to the user, and the application process of the unified retrieval is shown in figure 9.

In summary, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for searching across systems of enterprise based on distributed index service deployment is used for searching across systems of enterprise business systems, and is characterized in that the method carries out the search across systems of more than two business systems, and the method comprises the following steps:

s101, deploying on the index server as follows:

setting a deployment directory, wherein the deployment directory is a storage path of the index service of the business system;

configuring an index path, wherein the index path is a storage position of an index file provided by an index service of the business system;

defining an index configuration file, wherein the index configuration file is in a script form of a document to be indexed and a database in the service system;

s102, constructing a database link pool and a file cabinet; registering the link resources of the database to be indexed in the service system in the database link pool; registering the document to be indexed in the service system in the file cabinet;

s103, configuring index fields, wherein the index fields comprise database index fields and document index fields;

the data index field is used for indexing a registration database in the database connection pool;

the document index field is used for indexing the registered documents in the file cabinet;

the index field and the corresponding index configuration file form an index file and are stored in the index path;

s104, creating a full index file for all registered databases and all registered documents in the database link pool and the file cabinet through full indexes;

creating an incremental index file for a newly added database or a newly added document, database update or document update;

merging the full index file and the incremental index file, and eliminating repeated items to form an index base of the service system;

and step two, creating a retrieval condition, retrieving the index file in the index database of the service system by adopting the retrieval condition, returning a retrieval result, and displaying.

2. The method of claim 1, wherein the business systems include a product data management system (PDM), an experimental data management (TDM), a quality management system, an intelligence bus system, and a co-design system.

3. The method according to claim 1 or 2, wherein in S101, the index profile comprises the following attributes: the index server uniquely identifies and index server type.

4. The method of claim 1 or 2, wherein the data index field comprises database fields of a database type, a query method, a lookup table, a query view, a query function, and an index;

the document index field includes the operating system of the document storage, the file server directory, the file type, and the document classification or name of the index.

5. The method according to claim 2, wherein the search condition is a data keyword, and the searching of the index file in the index repository of the business system by using the search condition and returning the search result specifically comprises:

setting retrieval priority as a PDM system index library, a collaborative design system index library, a TDM system index library, a quality management system index library and an information bus system index library;

comparing the data keywords with the index files in the index base of each service system in sequence according to the retrieval priority;

and taking the index file with the data keyword similarity exceeding a set threshold as a retrieval result, and returning the retrieval result.