Business Intelligence (BI) Architecture
Thebestindonesia.com – In Business Intelligence (BI), the architecture is composed of six main components, namely:
1. Data Sources
In this first stage, a process is needed to collect and integrate data stored in various sources that differ from each other both in origin and type. This data source comes from data contained in the operating system but can also come from unstructured documents such as e-mail and data sent by external parties.
2. Data Warehouses
At this stage, the process uses an extraction and transformation tool known as ETL (Extract, Transform, Load). Data originating from a variety of different sources is stored in a database that is intended to support the business intelligence analysis process.
3. Data Exploration
At this stage, tools that function for passive business intelligence analysis are used. These tools consist of query and reporting systems, as well as statistical methods. This methodology is passive because decision-makers must make decisions based on their hypotheses or define criteria from data extraction, then use analysis tools to find answers and match them with their initial views.
Read also : What is Business Intelligence?
4. Data Mining
At this stage, the process consists of active BI methodologies that aim to extract information and knowledge from the data. This methodology contains several mathematical models for pattern recognition, machine learning, and data mining techniques. In improving the work of the BI Dashboard, especially in making business decision automation, then in designing BI, it is necessary to add tools called data mining. The potential of data mining in handling large amounts of data and its ability to extract knowledge from big data makes data mining can be used to improve the performance of BI. The role of data mining is turning into a companion in automated decision support systems or decision automation.
5. Optimization
At this stage, it produces a solution where the best solution must be chosen from among the existing alternative solutions, and usually, there are many and varied or varied.
6. Decisions
At this stage, the main problem is determining the final decision, known as the decision-making process. Even though the BI methodology has been successfully implemented, the decision choice remains with the decision-makers.
Read also : Application of Business Intelligence (BI) in E-Banking Transactions
In addition, there are various types of BI, including:
1. Enterprise Reporting
It is used to generate static reports that are distributed to many people. This type of report is perfect for operational reports and dashboards.
2. Cube Analysis
It provides multidimensional Online Transaction Processing (OLTP) analysis for business managers in constrained environments.
3. Ad Hoc Query and Analysis
Used to provide users access to perform queries on databases and dig up information to the most basic level of transactional data. This query serves to explore the report carried out by the user.
4. Statistical Analysis and Data Mining
Used to perform predictive analysis or determine a causal correlation between two matrices.
5. Delivery Reports and Alerts
Used proactively to send complete reports or provide alerts to large or large populations of users.
