This work follows the 2021 curriculum of the Association for Computing Machinery for specialists in Data Sciences, with the aim of producing a manual that collects notions in a simplified form, facilitating a personal training path starting from specialized skills in Computer Science or Mathematics or Statistics. It has a bibliography with links to quality material but freely usable for your own training and contextual practical exercises.
Part of a series of texts, it first of all summarizes the standard CRISP DM working methodology used in this book and in Data Science projects.
Since this text uses Orange, MySQL and SQLite for application aspects, it describes briefly their installation and their elementary use. In any case this book should not be considered a user manual for these programs, and if by chance you were wrong in choosing it because you expected something different, you are free to return it but probably it is not the case to give a negative rating.
Then types of data and their representations, including images and documents, are analysed. The concepts of System, Information System and Data Base and the most common data base security and privacy practices are introduced.
The relational model and SQL is explained with application examples with MySQL and SQLite. We then analyze the various types of Joins, sorting, aggregation and grouping queries, integrity constraints, GRANT and REVOKE security features, views, indexing, Normal Forms and Normalization.
Multi-user access to Data Bases, interference and deadlock, locking techniques and transactions are then considered. Distributed Databases and the options possible with MySQL and SQLite are then described.
The limits of the relational model and the most common non-relational models (NOSQL) are outlined.
The most used conceptual models, Entity-Relationship and Object Model according to ISO/UML, are considered, also describing the process for stepping from the text of the problem to the associated conceptual and logical relational model.
The data integration process is outlined with the use of data warehouses, data lakes and mediators. The processes for data cleaning, management of missing, repeated, anomalous and incorrect values, coding of categorical values are described. Finally, Data Science project objectives are distinguished suggesting the best model, whether relational or non-relational.
The text is accompanied by supporting material and it is possible to download the examples and test data.
