Politecnico di Torino (logo)

Implementation of a data analysis tool, including data mining algorithms for advanced diagnostic purposes

Domenico Giuseppe Gatto

Implementation of a data analysis tool, including data mining algorithms for advanced diagnostic purposes.

Rel. Elena Maria Baralis. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering), 2019


This thesis work takes place in General Motors, one of the leading farm in the automotive global industry. Being an industry leader, nowadays, means to progress constantly and make innovation; for these reasons, the thesis is thought to be divided in two parts, each one focused on different technological fields . In the first part, I deal with the process of improvement and development of a tool for data analysis and visualization. Chapter 1, first, discloses the inspection work of the system already existing; the main task was to detect and to fix the weak points and the errors. Then, presents the work of enhancement of the tool by way of the development of new features. Chapter 2, instead, introduces the concept and the realization of a new tool to analyze a different type of data, requiring also a different OLAP database. In the second part, instead, the approach gets more experimental involving multivariate statistic analysis and artificial intelligence concepts. Here, I discuss how it could offer new opportunities to get more useful information from the data. In particular, I describe the exploitation of some clustering techniques to analyze more thoroughly data concerning with a specific trouble. In particular, Chapter 3 presents a brief introduction to the basic concepts and application of artificial intelligence, clustering especially, and describes which are the final purposes GM want to pursuit by means of it. In the Chapter 4, hence, I will explain the preliminary operations to make the data suitable for clustering analysis. Along the Chapter 5, I will describe in detail the several steps that constitute the whole process of cluster discovery, providing, in the end, for an analytic comparison among different cluster techniques. In the end, the Chapter 6 shows the integration of the clustering results in the tool developed in the first part of the thesis. The chapter continues with an explanation of the results and, at the close, I will touch upon the extended horizons for further analysis and some hints for future improvements.

Relators: Elena Maria Baralis
Academic year: 2018/19
Publication type: Electronic
Number of Pages: 99
Additional Information: Tesi secretata. Fulltext non presente
Corso di laurea: Corso di laurea magistrale in Ingegneria Informatica (Computer Engineering)
Classe di laurea: New organization > Master science > LM-32 - COMPUTER SYSTEMS ENGINEERING
Aziende collaboratrici: GM Global Propulsion Systems – Torino S.r.l.
URI: http://webthesis.biblio.polito.it/id/eprint/11538
Modify record (reserved for operators) Modify record (reserved for operators)