Course Syllabus

Export

Il corso intende fornire le competenze (metodologiche e tecnici) e gli strumenti per la comprensione e la realizzazione di soluzioni  per il processamento di Big Data (strutturati e non), attraverso l'uso degli algoritmi e tool di AI per l'estrazione e rappresentazione della  conoscenza da dati reali. Inoltre, il corso intende fornire gli strumenti tecnici per la modellazione e realizzazione di data model in accordo con il paradigma NoSQL, focalizzando principalmente sui graph-database e databaseNoSQL. Infine, si forniranno competenze tecniche e metodologiche relativamente ad algoritmi di explainable AI per la comprensione di algoritmi di machine learning black box

Introduction to AI and Big Data Analytics  

Getting knowledge from data

Modelling and Querying the Resulting knowledge



  • Introduction to AI and Big Data Analytics  
  1. Goal and rationale of AI. The relation between Big Data and AI
  2. The value of knowledge – digital economy and data-driven decision making
  • Getting knowledge from data
    1. Word Embedding (Word2Vec, Doc2Vec,GLOVE, FastText, StarSpace)
    2. Evaluate word embedding models (intrinsic vs extrinsic evaluation)
    3. Topic Modelling through Python
  • Modelling and Querying the Resulting knowledge through NoSQL
  1. introduction to NoSQL data stores
  2. graph-databases and graph-traversal query languages (Cypher)
  3. Document Databases
  • Explainable AI (global and local interpretation models)
  1. Introduction to XAI, local/global interpretation models. model agnostic-specific algorithms
  2. XAI techniques as in the state of the art (eg. LIME, SHAP, etc)

Nessuno

Lezioni frontali, seminari monotematici, esercitazioni, assegnamenti da svolgere a casa.

Lezioni con l'ausilio di slide, laboratorio e casi applicativi. Articoli scientifici di riferimento saranno forniti dal docente. Il Software utilizzato sarà open-source

1 marzo - 15 aprile 2021

La modalità di verifica si basa su una prova scritta ed una eventuale prova orale. 

La prova scritta si svolge al computer ed è composta da domande aperte e chiuse e risposta multipla su tutti gli argomenti del corso. 

In sede di valutazione viene considerata la capacità dello studente di rispondere a quesiti specifici facendo riferimento agli aspetti teorici e pratici (mediante esempi) connessi all'argomento richiesto.

La prova scritta è comune sia per gli studenti frequentanti sia per i non frequentanti.

La prova orale è mirata ad accertare la conoscenza teorica dello studente sugli argomenti del corso. Saranno quindi valutate le capacità di ragionare e approfondire  le tematiche proposte in sede di esame e il rigore metodologico del loro sviluppo. 


su Appuntamento

Export
The course aims at providing skills (both methodological and technical) and tools for understanding and implementing solutions for Big Data processing (structured and unstructured data), through the use of AI algorithms and tools for the extraction and knowledge representation from real data. In addition, the course intends to provide the technical tools for modelling and realising data models following the  NoSQL paradigm, focusing mainly on the graph-database and NoSQL database. Finally, competencies related to XAI will be provided to explain the behaviour of black box algorithms

Introduction to AI and Big Data Analytics  

Getting knowledge from data

Modelling and Querying the Resulting knowledge



  • Introduction to AI and Big Data Analytics  
  1. Goal and rationale of AI. The relation between Big Data and AI
  2. The value of knowledge – digital economy and data-driven decision making
  • Getting knowledge from data
    1. Word Embedding (Word2Vec, Doc2Vec,GLOVE, FastText, StarSpace)
    2. Evaluate word embedding models (intrinsic vs extrinsic evaluation)
    3. Topic Modelling through Python
  • Modelling and Querying the Resulting knowledge through NoSQL
  1. introduction to NoSQL data stores
  2. graph-databases and graph-traversal query languages (Cypher)
  3. Document Databases
  • Explainable AI (global and local interpretation models)
  1. Introduction to XAI, local/global interpretation models. model agnostic-specific algorithms
  2. XAI techniques as in the state of the art (eg. LIME, SHAP, etc)


None

The course will be provided by means of lessons, seminars, and laboratory sessions and homeworks.

Lectures with the support of slides, laboratory and real-life case studies. Scientific Papers and books indicated by the lecturer. The software used is either available as open source or through academic license

III ciclo

The verification method is based on a written test whilst the oral examination will be provided on request.

The written test takes place at the computer and it consists of open and closed questions with multiple answers on all course topics.

The evaluation is focused on the student's ability to answer to specific questions by referring both to the theoretical and practical aspects (through examples) connected to the requested topic.

The written test is common for both attending students and non-attending students.

The oral exam is aimed at assessing the theoretical knowledge of the student on the topics of the course. The ability to reason and deepen the issues proposed during the examination and the methodological rigor of their development will be evaluated.



By Appointment

Enter

Key information

Field of research
ING-INF/05
ECTS
12
Term
Second semester
Activity type
Mandatory to be chosen
Course Length (Hours)
94
Language
Italian

Staff

    Teacher

  • LM
    Lorenzo Malandri
  • Fabio Mercorio
    Fabio Mercorio

Students' opinion

View previous A.Y. opinion

Bibliography

Find the books for this course in the Library

Enrolment methods

Manual enrolments
self