Book Description

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. First of a series of books, it covers methodological aspects, data acquisition, management and cleaning. It describes the CRISP DM methodology, the working phases, the success criteria, the languages and the environments that can be used, the application libraries. Since this book uses Orange for the application aspects, its installation and widgets are described. Dealing with data acquisition, the book describes data sources, the acceleration techniques, the discretization methods, the security standards, the types and representations of the data, the techniques for managing corpus of texts such as bag-of-words, word-count , TF-IDF, n-grams, lexical analysis, syntactic analysis, semantic analysis, stop word filtering, stemming, techniques for representing and processing images, sampling, filtering, web scraping techniques. Examples are given in Orange. Data quality dimensions are analysed, and then the book considers algorithms for entity identification, truth discovery, rule-based cleaning, missing and repeated value handling, categorical value encoding, outlier cleaning, and errors, inconsistency management, scaling, integration of data from various sources and classification of open sources, application scenarios and the use of databases, datawarehouses, data lakes and mediators, data schema mapping and the role of RDF, OWL and SPARQL, transformations. Examples are given in Orange. The book is accompanied by supporting material and it is possible to download the project samples in Orange and sample data.