João Marcelo Borovina Josko

Avenida dos Estados, 5001 · Santo André · São Paulo · Brazil
Room 520-2 · Tower 2 · Building A
marcelo.josko@ufabc.edu.br

 

Further information

My ORC
My Scholar Google


Interests

Data Quality and Data Integration

Independent of your business, all data analysis initiative requires data at certain quality levels to produce reliable outcomes. So, data persisted in Relational or NoSQL is not enough. Data quality assessment and data integration complement each other to rise data quality at the required levels. The former is in charge of detecting data defects on several types of data (e.g., atemporal, temporal, spatial, semi-structured, non-structured), while the latter attempts to identify integrations points between different entities or schemas. Both use a broad set of resources (e.g., machine learning, data visualizations, natural language processing, probability) to reach its aims.

Data Visualization

Data Visualization, as field, searches for interactive visual representations that permit complex data analysis. Indeed, visualizations are required for a myriad of situations, including data quality visual assessment, education, database administration, cloud administration, financial analysis, scientific analysis and much more. New or improved visualizations techniques (metaphors), New or improved interactive techniques, evaluation of visualizations solutions represent some of the promising works regarding this subject.

Computer Science Education

Computer science is a diverse field that gathers several technical, personal and in group skills. Learning such complex and interconnected skills demand much more than gis, blackboard and presentations. The use of pedagogical methods with or without support of software resources (e.g., simulators, intelligent interactive tutors, visualizations) are a must to leverage computer science learning. In this context, learning analytics is in charge to help us to understand if and how methods and resources improve learning.

Database

Databases are ubiquitous to all kinds of relevant applications. Due to this relevance, databases require several procedures to work propely throughout its productive life. Certain management procedures (e.g., schema evolution or versioning, database tuning ) must have more resources to make their execution easier or even autonomous.