Corso Data Science GRATUITO (settembre 2022)

Durata:	248 ore
Periodo:	Dal 26 settembre al 22 novembre 2022
Orario:	Dal lun al ven dalle 09.00 alle 16.00 (pausa dalle 13.00 alle 14.00)
Sede:	Il corso Data Science si svolgerà ONLINE tramite piattaforma webinar
Partecipanti:	Verranno selezionati massimo 15 partecipanti
Scopo:	ASSUNZIONE
Prezzo:	GRATUITO
Prerequisiti:	Laurea in indirizzi tecnici, basi di programmazione
Destinatari:	Tutti coloro che si occupano di dati.
In breve:	Il mondo di oggi produce più dati che mai! Riuscire a trasformarle in informazioni utili è una nuova abilità chiave. Questo corso ti introduce al Data Science pratico, in ti verrà svelato il mistero che circonda l'argomento. Spiegheremo i principi degli algoritmi popolari, mostrandoti come usarli nelle tue applicazioni pratiche. Il corso è composto da tre diversi moduli, base, intermedio e avanzato, per uno sforzo totale di quasi 250 ore (6 ore/giorno). Questo corso è rivolto a tutti coloro che si occupano di dati. Non richiede alcuna programmazione informatica, sebbene sia necessaria una certa esperienza nell'uso dei computer per le attività quotidiane. La matematica del liceo è un prerequisito, insieme ad alcuni concetti di statistica elementare (come medie e varianze). Nelle 8 settimane verrà utilizzato software gratuito.

Topics ( part-1 / weeks 1-2)

What data science is & Where it can be applied

How simple classification algorithms work

What their strengths and weaknesses are

In what ways real-life classification methods are more complex

How to evaluate a classifier’s performance

What “overfitting” is and how you can combat it

How ensemble techniques can combine the result of different algorithms

Data Science and ethical/legal consideration

Achievements ( part-1 / weeks 1-2)

Demonstrate use of Data Science for key data mining tasks

Evaluate the performance of a classifier on new, unseen, instances

Explain how data scientists can unwittingly overestimate the performance of their system

Identify learning methods that are based on different flavors of simplicity

Apply many different learning methods to a dataset of your choice

Interpret the output produced by classification methods

Describe the principles behind many modern machine learning methods

Compare the decision boundaries produced by different classification algorithms

Debate ethical issues raised by mining personal data

Topics ( part-2 / weeks 3-4-5)

Running large-scale data science experiments

Constructing and executing knowledge flow

Processing very large datasets

Analyzing collections of textual documents

Mining association rules

Preprocessing data using a range of filters

Automatic methods of attribute selection

Clustering data

Taking account of different decision costs

Producing learning curves

Optimizing learning parameters in data

Achievements ( part-2 / weeks 3-4-5)

Compare the performance of different data science
methods on a wide range of datasets

Solve data science problems on huge datasets

Identify the advantages of supervised vs unsupervised
discretization

Classify documents using various techniques

Explain how association rules can be generated and used

Perform attribute selection by wrapping a classifier inside a
cross-validation loop

Develop effective sets of attributes for text classification
problems

Design and evaluate multi-layer neural networks

Calculate optimal parameter values for a given learning
system

Demonstrate how to set up learning tasks as a knowledge flow

Apply equal-width and equal-frequency binning for discretizing
numeric attributes

Evaluate different trade-offs between error rates in 2-class
classification

Debate the correspondence between decision trees and
decision rules

Discuss techniques for representing, generating, and
evaluating clusters

Describe different techniques for searching through subsets of
attributes

Explain cost-sensitive evaluation, cost-sensitive classification,
and cost-sensitive learning and assess the volume of training
data needed for mining tasks

Topics ( part-3 / weeks 6-7-8)

Time series analysis

Data stream mining

Incremental classifiers

Evolving data streams

Support vector machines

Distributed data science

Map-reduce framework

Scripting data science in R, Python and Groovy

Applications: Soil analysis, Sentiment analysis,
Bioinformatics, MRI neuroimaging, Image classification

Achievements ( part-3 / weeks 6-7-8)

Discuss the use of lagged variables in time series forecasting

Identify several different applications of data science

Evaluate the performance of classifiers under conditions of
concept drift

Calculate optimal parameter values for non-linear support
vector machines

Design R, Python and Groovy scripts

Explore the use of overlay data in time series forecasting

Classify tweets using various techniques

Demonstrate how to set up learning tasks as a knowledge flow

Apply equal-width and equal-frequency binning for discretizing
numeric attributes

Experiment with distributed implementations of classifiers and
clusterers

Compare incremental and non-incremental implementations of
classifiers

Apply Python libraries to produce sophisticated visualizations

	Nome: Dr. Davide D'Aprile Esperienza: PhD e MSc in Informatica, con quasi 15 anni di esperienza pratica sulla ricerca e l’applicazione del Data Science in più settori.