TW201947492A - System and method for operational data convergence - Google Patents

Abstract

The disclosure is related to a system for operational data convergence. The system adopts a three-tier data storage structure. A high-performance relational database is used to implement an original data zone for providing original data. A big data platform with high price-performance ratio storing and distributed processing capability is used to implement a data value-added zone for synchronizing data with the relational database in real time, and data collection. A low-cost file server is used to implement a data storage zone for data storage and auditing. Further, an information system is provided to be a data transmission tool between the relational database and the big data platform. A software method is introduced to monitor the flow change of the relational database and inform the change to the information system. The changed data stored in the information system is then written into the big data platform in sequence for data synchronization.

Description

   Operational data converging system and method   

The specification discloses an information system, in particular an operational data convergence system implemented by a three-tiered data storage architecture, and an operating method therein.

Financial institutions such as banks and securities companies often generate a large amount of information. Information processing requires a powerful information system. However, if there is information that needs to be processed in real time, it cannot be processed in real time because of the low efficiency of the information system. Give users instant and useful information.

In financial transactions, it is necessary to exchange data across systems. Such system data exchange is mostly provided through batch files. If the timeliness of information is low, it will not be able to effectively provide business value-added applications. For example, if a consumer spends up to a credit card in A, the bank will give a discount on local consumption. However, because of a system problem, the transaction cannot be processed immediately and the preferential information is delivered immediately. When the consumer has left A In order to reach the consumer, the discount message cannot reach anything.

In addition, in addition to the above-mentioned real-time information processing issues, in the use, query and storage of a large amount of historical data, as various data of financial institutions come from different systems, the data is also increasing day by day, which will make the information system architecture complicated and each system The information structure of the data content is inconsistent and there is no unified system management, which not only causes difficulties in the integration and application of cross-system data, but also depends on past experience for data analysis. Internal communication requires a lot of time, and the correctness of the data also requires a lot of manpower to confirm and review the operation It is also complicated and time consuming. This will generate a lot of time-consuming and labor-intensive work and increase operating costs.

The disclosure proposes an operating data convergence system and its operation method. The operating data convergence system mainly uses a three-tiered data storage architecture. It mainly includes the original data area, data value-added area implemented by software programs and databases of the server system, and Data storage area.

Wherein, according to an embodiment, the operational data converging system implements the original data area with an associated data database for efficient data processing to provide the original data service; the data value-added area is implemented by a big data platform with cost-effective storage and decentralized processing , Real-time synchronization of the data in the relational database to provide value-added integration services; and a data storage area implemented by a low-cost inventory file server to provide backup and audit services.

Further, the operating data convergence system uses a message system as a data transmission tool between the relational database and the big data platform, and a message provider is executed between the relational database and the message system, which can transfer the information in the relational database. The table is treated as a stream according to the key, and it can detect changes in the data stream in the relational database in real time and send the changes to the message system.

Furthermore, a message writing program can be executed between the information system and the big data platform to sequentially write the transaction data retained in the information system to the big data platform to achieve data synchronization.

The way of operating the data converging system to obtain data is to collect external or internal raw data, including data that can be synchronized in real time or collected in a batch of data. This can include procedures for converting and cleaning data to provide high-quality data.

Further, in the data service in the operation data converging system, it is possible to provide diversified content through the function of the data value-added area, and through the data provision function to provide the services of data inquiry, subscription, and the provision of a batch of data.

In the embodiment of the operation data converging method, external or internal information is obtained through the system, and data conversion or data cleaning is performed on the external or internal information through a batch of data collection functions to form a non-structure in a data value-added area. The data, or directly integrated with the data, becomes the source of autonomous analysis in business intelligence applications, and is presented to the user in a visualization. Among them, the internal information is synchronized to the original data area through the real-time synchronization operation, for the purpose of querying the original data, and also for the real-time streaming analysis of the data value-added area.

In order to further understand the technology, methods and effects adopted by the present invention to achieve the intended purpose, please refer to the following detailed description and drawings of the present invention. It is believed that the purpose, features and characteristics of the present invention can be deepened and specific It is understood, however, the drawings are provided for reference and description only, and are not intended to limit the present invention.

11‧‧‧ raw data area

12‧‧‧ Data Added Area

13‧‧‧Data storage area

21‧‧‧Efficient data processing area

22‧‧‧ Cost-effective data processing area

23‧‧‧ Low-cost data processing area

31‧‧‧Related Database

32‧‧‧ Feeder

33‧‧‧Information System

34‧‧‧Message Writer

35‧‧‧ Big Data Platform

41‧‧‧Source

411‧‧‧Internal Information

412‧‧‧External Information

42‧‧‧Data Collection Unit

421‧‧‧Real-time synchronization

422‧‧‧Batch data collection

43‧‧‧Data Service Unit

431‧‧‧Data description area

432‧‧‧ raw data area

433‧‧‧ Data Added Area

434‧‧‧Data storage area

44‧‧‧ Information Supply Unit

441‧‧‧Inquiry

442‧‧‧Data Subscription Service

443‧‧‧ Full batch of information provided

45‧‧‧Data Application

451‧‧‧Business Intelligence

452‧‧‧Reporting Platform

453‧‧‧Newsletter Platform

454‧‧‧Application System

501‧‧‧ Source of raw materials

502‧‧‧Source data query

601‧‧‧Historical data query

602‧‧‧Data Model Establishment

603‧‧‧ Complex and efficient operations

604‧‧‧Unstructured data storage

605‧‧‧Data integration operation

606‧‧‧Automatic analysis preprocessing

607‧‧‧Real-time streaming analysis

608‧‧‧Data Model Operation

701‧‧‧Data storage

702‧‧‧ Audit inquiry

81‧‧‧External Information

82‧‧‧Internal Information

83‧‧‧Real-time synchronization

84‧‧‧Batch data collection

85‧‧‧ raw data area

86‧‧‧Data Value Added Area

87‧‧‧Inquiry

88‧‧‧Data description area

89‧‧‧Business Intelligence

90‧‧‧ Marketing Analyst

Steps 801 ~ 821‧‧‧‧ Example of operation data flow

Figure 1 shows a schematic diagram of a three-tiered architecture embodiment of the operational data convergence system; Figure 2 shows a cost and performance relationship diagram of the three-tiered data storage method in the operational data convergence system; and Figure 3 shows an example of a system architecture of the operational data convergence system. 1; FIG. 4 shows the second embodiment of the system architecture of the operational data convergence system; FIG. 5 shows the original data area embodiment of the operational data convergence system; FIG. 6 shows the data value-added area embodiment of the operational data convergence system; and FIG. 7 shows the operation An embodiment of the data storage area of the data converging system; FIG. 8 shows an embodiment of the method flow of operating the data converging system.

The disclosure relates to a method and system for converging operational data. The proposed operational data converging system conceptually implements a three-tier data architecture embodiment as shown in FIG. 1, which mainly includes a raw data area 11, which is located in a In the embodiment, it can be implemented by a relational database (RDB) for efficient data processing to provide raw data services. A data value-added area 12, which is a big data platform, can be cost-effective in one embodiment. Implementation of storage and decentralized Hadoop open source project. This is an open source project of the Apache Software Foundation. It has powerful decentralized computing capabilities and can handle analysis of various network usage records in the system, but it is not limited to the aforementioned use. Range, which is used to provide data value-added integration services; and a data storage area 13, which is mainly implemented by a low-cost inventory file server to provide data storage and auditing.

The operating data converging system can use a variety of technologies to combine data processes. The main means is to integrate the internal and external information of financial institutions (such as financial control and banks) through data processing solutions with different cost and efficiency considerations. It is divided into the original data area as shown in Figure 1. 11. The three-tier structure of data value-added area 12 and data storage area 13, according to its attributes, consider costs and storage performance. As shown in Figure 2, the cost and performance relationship diagram of the three-tier data storage method in the operating data converging system. .

Figure 2 shows the relationship between storage costs (vertical axis) and response speed (horizontal axis) of data processing for various data processing schemes.

The high-efficiency data processing area 21 is located at a location with high storage cost but fast response speed. It is a storage solution that often has a large amount of data influx. It is suitable for the raw data area 11 in the operating data converging system shown in Figure 1. The cost-effective data processing area 22 is a storage solution with a moderate storage cost but a good response speed, which is suitable for the data value added 12 in the operating data converging system, which often requires data analysis and provides customers with immediate needs. The low-cost data processing area 23 is a storage solution with low response speed but low storage cost. It is suitable for the needs of the data storage area 13 in the operating data converging system. Such data is not often changed, and it only needs to meet the needs of accessing files. .

Based on the three-tier data architecture described above with efficient data processing, cost-effective storage and decentralized processing, and low-cost inventory, the construction of the operational data converging system proposed in the disclosure book covers timeliness, diversity, integration, security, and security. High-quality five-oriented information services, with a single platform structure, meet the needs of financial institutions with high efficiency and stability and cost considerations.

According to the embodiment of the operational data converging system, reference may be made to the embodiment of the system architecture shown in FIG. 3. The original data area (FIG. 1, 11) is implemented by a relational database (RDB) 31, and the data value added area (FIG. 1, 12 ) Is a large data platform 35.

When the system is operating, the relational database (RDB) 31 synchronizes data to the big data platform 35 in real time. The synchronization method, for example, adopts Change Data Capture (CDC) technology. The purpose is to process a large amount of data generated in the operational data converging system. The big data platform 35 used in it, such as the Hadoop open source project of the Apache Software Foundation, has strong decentralized computing capabilities and is responsible for processing and analyzing various network usage records in the system, but it is not limited to the aforementioned application examples. In the operating data converging system, the data is synchronized to the big data platform 35 implemented by Hadoop in real time by the relational database 31 through a CDC program.

The operational data converging system proposed in the disclosure uses a message system 33 as a data transmission tool between the relational database 31 and the big data platform 35. A message providing program 32 is executed between the relational database 31 and the message system 33. The table (TABLE) in the relational database 31 is regarded as a stream according to the key value (KEY). The message providing program 32 Real-time detection of changes in data flow in the associated database 31 and transmission of the changes to the message system 33. In addition, a message writing program 34 is executed between the message system 33 and the big data platform 35, and the transaction data stored in the message system 33 is sequentially written into the big data platform 35 through a stream processing program to achieve data synchronization.

The above-mentioned information system 33, such as Kafka, which handles real-time data in the Apache Software Foundation, is a data management and information system designed to process transaction data and operational data processing in a system. The system takes a website as an example. The website's activity data, such as the number of web page visits, viewed content, and search records, etc., are finally written into a file or database for subsequent analysis and statistics. Operational data such as Web server performance data, such as CPU operation, input / output interface usage, runtime, etc.

When the information system 33 is used to disclose the operational data converging system disclosed in the book, it is related to the work of publishing and subscribing to the message, allowing the publisher of the message to write the content, and the receiver of the message can read the data from the message system. Among them, the information system 33 implemented by, for example, Kafka, can guarantee that the message publisher in the operational data converging system sends the message to the designated object.

In the process of message transmission, the metadata in the data becomes important information for data management. The metadata can be used as historical review and analysis. If it is applied to the operation data convergence system, when the data is updated, the metadata is also synchronized. Update, basic metadata information such as table name, field name, field content, etc. Advanced metadata information such as data business description, field business description, technical logic, data update frequency, etc.

According to the data processing architecture of the operational data converging system described in the above embodiments, the system architecture is mainly shown in Figure 4. It can be roughly divided into five parts when applied to financial institutions, which can be servo systems, databases, and run in servo systems. Software programs.

The first part is the data source 41, including internal information 411 (deposits, credit cards, wealth, finance, etc.) and external information 412 (public information, community information, etc.) from financial institutions, or one of them.

As far as banking is concerned, for example, consumers' information on the use of financial cards and credit cards in various places will be imported into the system in real time. These data will be processed by the system's data collection unit 42 and require a high-performance data processing server system, including Real-time synchronization operations 421 that need to be considered for timeliness, such as checking credit card consumption, transferring funds, and responding to instant messages in real time. The system can generate high-quality data through a batch of data collection 422 operations, including data conversion, such as ETL (extract, transform, and load) data processing procedures, and data cleaning procedures, etc. The work can ensure the production of high-quality data, and integrate and uniformly process the data generated from different sources.

The system includes a data service unit 43, which can provide diversified content according to the collected data. For example, there is a data description area 431, which includes tools such as enterprise data dictionary, blood relationship and impact analysis. The collected data can be translated (semantic). After analysis, knowledge tools in the system are formed.

There is a raw data area 432, which is used to store the collected, converted and cleaned up raw data. Compared with the last saved data, this part is more recent data and can be related. Database (RDB) to provide services for querying raw data.

The data service unit 43 is provided with a data value-added area 433 to generate integrated data. For example, the big data servo system implemented by the aforementioned Hadoop (the project Hadoop® under the Apache Foundation) can be used to implement history in this system. Data query, data model establishment, complex and efficient calculation, non-structured data storage, data integration calculation, autonomous analysis pre-processing, real-time streaming analysis, and data model calculation and other functions.

The data service unit 43 is provided with a data storage area 434. The data storage area 434 is generally a low-cost storage solution for storing historical data and providing functions such as data storage, auditing and query.

The operating data converging system includes a data providing unit 44 that provides various services through software programs in the operating data converging system. The services provided include data query 441. For example, security measures with permission management can be provided through the application programming interface (API). Related operations of data query 441 include database virtualization integration, query permission management (security), and data retrieval. Identification. The data providing unit 44 provides a data subscription service 442, which can provide functions such as data subscription and push broadcast according to user requests. The data providing unit 44 provides a whole batch of data providing 443, including methods of data conversion, ETL data processing, and batch delivery.

Finally, the data collected by the operating data converging system is output by the data service unit 43 and the data providing unit 44 to form various data applications 45. According to the embodiment, the application of financial institutions can include business intelligence 451, which can provide visualization And self-analysis; report platform 452 provides report query; SMS platform 453 provides newsletter sending service; and application system 454 includes performance management and risk management functions, but the implementation of application system 454 is not limited to the aforementioned scope of use.

The core of the operating data convergence system lies in the original data area (Figure 4, 432), data value-added area (Figure 4, 433), and data storage area (Figure 4, 434) in the data service unit (Figure 4, 43). .

FIG. 5 first shows an embodiment of the original data area (FIG. 4, 432) in the data service unit when the operation data converging system is running. The figure shows that the raw data area 432 mainly includes two functions: a raw data supply 501 and a raw data query 502 implemented by software programs running on a computer system and cooperating with a database. The original data supply 501 is data that is collected, converted, and cleaned by a synchronous data source, and stored in a relational database. In addition to providing high-quality data, it also uses the function of the original data query 502 to form queryable structured data.

FIG. 6 shows an embodiment of a data value-added area of an operation data converging system. The figure shows the data value-added area 433. After various data are collected by the system, the data are integrated through the operation of software programs and databases in the system to generate integrated big data, which enables the system to provide historical data query 601, data Model creation 602, complex and efficient operations 603, unstructured data storage 604, data integration operations 605, autonomous analysis pre-processing 606, real-time streaming analysis 607, and data model operations 608.

Historical data query 601 enables the data collected by the system to become queryable historical data.

The data model creation 602 function provides the system to create a data model based on the collected data, define inputs and outputs, cooperate with the data model calculation 608, and reply to the results after the input data is calculated in real time.

The complex and efficient operation 603 provides the system with the ability to efficiently operate various data because of the data virtualization architecture. The unstructured data storage 604 allows the system to store unstructured data such as audiovisual data, external open data, and the like. Data integration operation 605 provides a system that can quickly integrate cross-system resources without being limited to a single data source. It is suitable for financial institutions and other applications that need to process data from complex sources. The autonomous analysis pre-processing 606 provides the system with the ability to process internal information and external information in advance, such as data conversion, cleaning, and synchronization, before autonomous analysis.

Real-time streaming analysis 607 provides the system to process real-time data, which can broadcast information in real-time using technologies such as newsletters and push broadcasts, and can form subscription content.

Figure 7 shows an example of the data storage area of the operating data converging system. The main function of the data storage area 434 is to provide data storage 701. It uses a low-cost storage solution to store recent data for future reference, and it is also used in financial institutions. Audit query 702.

It can be seen that the operation data integration system provides data integration operation procedures through the data collection unit 42, the data service unit 43, and the data providing unit 44, so that the original data collected by the system can form knowledge, queryable data, and data models. In addition to the data storage established by the cost storage scheme, it is still possible to provide data query, subscription and batch services, and generate the final diversity of data applications45.

The above examples show that the operating data converging system allows the entire system to take into account the timeliness (real-time, real-time, batch) through the application of software programs and databases on the server system side, mainly referring to the timeliness of data query and application. The function of the data collecting unit 42 achieves the effect of synchronizing the data with the data source, so as to facilitate quick response to customer needs.

The system provides data diversity, including original / aggregated / historical data query, real-time data push, batch data provision, and other functions. It provides a three-tiered data storage architecture that integrates a relational database with a large-scale parallel operation system. After the integration of data platforms and low-cost storage solutions, data storage management will be strengthened to provide data services to improve the use efficiency.

The high quality means that the system provides functions such as data cleaning (such as ETL) and data quality inspection, which can promote data integration.

System integration refers to functions such as data integration calculations and data model calculations to achieve data aggregation and form a business model.

Security refers to permission management, providing query permission management mechanism, which can provide appropriate information according to permissions, including the use of data to identify.

FIG. 8 shows an embodiment of a method flow running in an operational data converging system. In this process embodiment, the data sources are external information 81 and internal information 82. The system uses the obtained information to form the processing data converging system processing target. In the process of real-time synchronization 83, batch data collection 84, original data area 85, data value-added area 86, data query 87 is provided, and the content description area 88 is formed. It is also used as business intelligence 89 and marketing analysts 90 Applications that perform analysis.

In the process, external public map information is obtained from the external information 81 (step 801), and the map data is converted in the batch of data collection 84 (step 805) to form unstructured data stored in the data value added area 86 (step 809). ), And after data integration calculation (step 810), it becomes a part of the business intelligence 89 application, which is to provide autonomous analysis (step 818), which can be presented to users, such as marketing analysts 90.

On the other hand, the system obtains the internal information of the enterprise (such as a financial institution) (step 802). These data can be provided to the data value-added area 86 after the data is cleared (step 806), and then provide independent analysis (step 818).

In the process used by financial institutions, the internal information 82, for example, provides consumers with preferential data (step 803). These data are synchronized in real-time synchronization operation 83 (step 804) in order to be synchronized to the original data area 85 to become Part of the original data supply (step 807) can be used as the original data query (step 808), and also used in the data value-added area 86 for real-time streaming analysis, analyzing big data data streaming (step 811), and also providing autonomous analysis (Step 818).

When the original data supply (step 807) becomes the original data query (step 808), the system judges whether these are required data by the function of the data query 87? (Step 813), if the system needs data (Yes), perform autonomous analysis (Step 818), otherwise, the data can become a data dictionary in the data description area 88 (Step 817). The content of the data dictionary allows marketing analysts to identify the data they need to analyze. One of the purposes is to allow users to filter the data that they don't need to analyze.

When the system obtains the converted external information, the cleaned internal information, the internal information analyzed by real-time streaming, or the data obtained by querying the original data, after the independent analysis step in the Business Intelligence 89 application (step 818), the information It can be visualized (step 819), allowing marketing analysts 90 to use visual reporting tools, combining real-time data streaming analysis and batch organization of data, and analyze the effectiveness of marketing activities based on business dimensions until the end (step 820). Process.

On the other hand, the marketing analyst 90 raises the need for data analysis and the system performs the analysis (step 821). First, the data description area 88 can be queried for the data dictionary (step 817). The marketing analyst 90 can confirm the data to be analyzed. Screen for data that does not require analysis.

Next, in the data query 87 function provided by the operation data confluence system, first perform the query permission management (step 816) step to determine whether the user who logs in to the system query (that is, the marketing analyst 90 in this example) has permission? (Step 814) If it is judged that it has authority (Yes), the data de-identification will be performed (Step 812). The purpose of the data de-identification is to allow the system to remove the information that can identify the source or identity in the data. The query of the original data (step 808), on the other hand, allows the user to confirm the analyzed data. Conversely, if the user does not have the query authority (No), the query is rejected (step 815).

The following lists the advantages of the operational data converging system proposed in the disclosure.

The operational data converging system combines real-time data synchronization technology, big data data streaming, ETL and other technologies to consolidate the entire bank's data, which can greatly simplify the complex data flow in the past, reduce system maintenance costs, and meet the real-time value-added aggregation and model. Build data service requirements for operations, historical bulk data, and unstructured data processing. Through the data virtualization architecture, the system quickly integrates cross-system resources, provides a single data query channel and data de-identification function, realizes centralized permission management and improves data security.

Running in specific enterprises, the operational data converging system provides corporate data definitions and facilitates information sharing across departments through appropriate authority management, reducing the barriers to communication and application between business systems, which can not only improve system analysis efficiency, but also drive The business analyzes energy autonomously, breaks through traditional digital management, and drives business development with data. At the same time, the accuracy of the data is maintained at all times through data quality management.

By integrating the entire bank's data sources, combined with real-time data synchronization (CDC) data synchronization technology, and through big data data streaming technology and ETL processing technology, in addition to reducing the complexity of each system interface, it also greatly simplifies the data at the same time. Although the number of systems in the industry will increase in the future, the overall system complexity will not be affected, which effectively reduces the system maintenance costs.

Finally, the system integrates online and offline data in the industry, and combines public information to understand real-time customer dynamics and grasp real-time business opportunities.

In this way, according to the embodiment of the operation data converging method and system above, it can be known that when the core of the operation data converging system constructs a three-tier data structure, the data is divided into the original data area, the data value-added area, and the data storage area, achieving a variety of Combining technology with data processes, integrating information from various systems within financial institutions, and constructing five major information services covering timeliness, diversity, integration, security and high quality, which can be combined with real-time data synchronization, big data streaming analysis, Techniques such as sensitive data masking, through data virtualization architecture, use the concept of application programming interface (API) to achieve a single platform of diverse data services, so that real-time streaming technology for commercial databases and open parallel computing environments can be created, achieving three layers The system interface between the two types of data structure, strengthen the data storage management, supplement the data definition, and improve the data use efficiency. Therefore, it has the following advantages: First, to meet the requirements of high stability of financial institutions; Second, the platform structure can be carried out at a low cost; Third, provide a variety of data services such as real-time / real-time / batch type; Fourth, the traditional Integrated structured and unstructured data on a single platform; and 5. Provide integrated and de-identified query services.

However, the above description is only a preferred and feasible embodiment of the present invention, and therefore does not limit the patent scope of the present invention. Therefore, any equivalent structural changes made by using the description and illustrated contents of the present invention are also included in the present invention. Within the scope, joint Chen Ming.

Claims (10)

    An operation data converging system includes: a raw data area implemented by a relational database for storing raw data collected from an external information and an internal information, or one of them, and data conversion and data cleaning Data; a data value-added area, realizing a large data platform, real-time synchronization of the data of the related database, to provide data value-consolidation services; and a data storage area, implemented by a file server, for Provide data storage and audit.         The operating data converging system according to claim 1, wherein an information system is used as a data transmission tool between the relational database and the big data platform.         The operating data convergence system as described in claim 2, wherein an information provider is executed between the relational database and the information system, and the tables in the relational database are treated as a stream according to the key value, and the The message provider can detect changes in the data flow in the associated database in real time and send the changes to the message system.         The operation data converging system according to claim 2, wherein a message writing program is executed between the information system and the big data platform, and the transaction data retained in the information system is sequentially written through a stream processing program This big data platform achieves data synchronization.         The operation data converging system according to claim 1, wherein the way of collecting the original data includes a real-time synchronization operation and a batch of data collection operations.         The operating data converging system as described in claim 5, wherein the operation of collecting the entire batch of data includes extraction, transposition and loading of data processing procedures, and a data cleaning procedure.         The operational data converging system as described in claim 1, wherein the data value added area includes functions including: historical data query, data model establishment, complex and efficient calculation, non-structured data storage, data integration calculation, autonomous analysis pre-processing, real-time Stream analysis and data model operations.         The operational data converging system as described in claim 1, wherein the data storage area uses a lower cost storage solution to store historical data, and the functions provided include data storage and audit query.         The operation data converging system according to any one of claims 1 to 8, further comprising a data providing unit, and services provided through software programs in the operation data converging system include data inquiry, data subscription services, and batch data provide.         An operational data converging method applying the operational data converging system according to claim 1, comprising: obtaining an external information or an internal information; performing a data conversion or data on the external information or the internal information through a batch of data collection function Clean up; form an unstructured data in the data value-added area, or directly integrate the data and calculate it to become an independent analysis in business intelligence applications and present it to the user in a visualization; where the internal information is synchronized in real time Synchronize to the original data area for the original data query and also for the real-time streaming analysis of the data value-added area. Among them, a data query function provided by the operational data converging system includes the steps of performing query authority management To determine the permissions of the user who logs in to the system.