### System Mindmap ![dbscore-mindmap.png](https://cdn.triangulos.tech/assets/dbscore_mindmap_321f9ecf50.png) ### Research Summary #### Problem definition The world is becoming increasingly digital through technological innovations and the growing volume of data. As a society, we need to prepare ourselves and be ready for the changes that emerge in the use of data in decision-making and, above all, in the efficient and responsible sharing of data [Article1]. Therefore, data validation and quality will become more and more present to build or develop a model. However, there are no tools or systems that can assist in the automatic or semi-automated verification of a database, assisting analysts or data publishers [Article2]. Thus, the DBScore project is a system developed in Python that automates the analysis to define the quality of a database using two indicators: FAIR and Openness. Data can be inserted in various formats, such as "free" (.csv, .xlsx, or .pdf), compilation formats of sets of databases (.zip, .rar), or even in API / HTTPS format. #### Methodology DBScore is a system that can analyze large volumes of data and validate their quality, using indicators and generating detailed reports for model development in decision-making. Thus, the proposed system is generally structured and presented in Figure 1, where we can understand all the connections between the layers and their importance in the context of evaluation. For a better understanding, each layer will be better exemplified in the next subsections: ![DBScore](https://cdn.triangulos.tech/assets/dbscore_flow_9bac594b03.png) Importation/Standardization of the Database for Analysis For the processing of DBScore, the first step was to understand and systematize the types of databases that could be added to the platform for analysis. Therefore, the main types are [Article4]: - Relational Database: This is the most common type of database used to store related data in tables. - NoSQL Database: This type of database is designed to handle large amounts of unstructured data and is often used in large-scale applications. - Object-Oriented Database: This type of database is designed to handle structured data similarly to objects in programming. **Evaluation of Indicators** DBScore was implemented using best practices to validate and measure the quality of a database automatically, with the help of libraries and functions that efficiently replicate rules that allow verifying crucial points of a database. The indicators in the system complement each other in the goal of offering a broader view of minimum conditions that respect data sharing, data maturity, and integrity. Thus, the indicators that compose and join the system are: - Openness: This indicator checks the openness of the data, validating its availability, accessibility, description, usage license, and especially its resource for access. - FAIR: It is a principle created to guide the production, storage, and use of scientific and technological data in the digital age. They aim to ensure that data is produced and stored consistently, reliably, and accessibly, in order to maximize its value and reuse. **Generation of Reports** After executing the steps of Importation/Standardization of the Database for Analysis and Evaluation of Indicators, described in subsections 2.1 and 2.2, respectively. The report generation step is responsible for compiling and reconciling all the results achieved in the evaluation of the database. This report brings together the following information: - Introduction: Context of the evaluation, objectives, and adopted methodology, with a detailed summary of the indicators and their relevance. - Exploratory Analysis: Graphs that illustrate the shape and score of the databases for each analyzed indicator. - Conclusion: Summary of the results, identification of strengths and weaknesses of the database, recommendations to improve data quality for each analyzed indicator. This report aims to provide clear and reliable information about data quality, enabling decision-making for both managers and researchers who will use the data to create models. #### References Article1 - SIQUEIRA, Ethevaldo. Para compreender o mundo digital. Globo Livros, 2008. Article2 - SARLET, Ingo Wolfgang. Inteligência Artificial, Proteção de Dados Pessoais e Responsabilidade na Era Digital-Série Direito, Tecnologia, Inovação e Proteção de Dados num Mundo em Transformação. Saraiva Educação SA, 2022. Article3 - HOMCI, JANAINA VIEIRA. A PROTEÇÃO DOS DADOS PESSOAIS NO CONSUMO DIGITAL. Editora Thoth, 2023. Article4 - MACHADO, Felipe Nery Rodrigues. Banco de Dados Projeto e Implementação. Saraiva Educação SA, 2020.

Welcome to

DEMO OUTREACH