For more than a century, the methods for data representation and the exploration of the intrinsic structures of data have developed remarkably and consist of supervised and unsupervised methods. However, recent years have witnessed the flourishing of big data, where typical dataset dimensions are high and the data can come in messy, incomplete, unlabeled, or corrupted forms. Consequently, discovering the hidden structure buried inside such data becomes highly challenging. From this perspective, exploratory data analysis (EDA) plays a substantial role in learning the hidden structures that encompass the significant features of the data in an ordered manner by extracting patterns and testing hypotheses to identify anomalies. Unsupervised generative learning (UGL) models are a class of Machine Learning (ML) models characterized by their potential to reduce the dimensionality, discover the exploratory factors, and learn representations without any predefined labels; moreover, such models can generate the data from the reduced factors’ domain. The beginner researchers can find in this survey the recent UGL models for the purpose of data exploration and learning representations; specifically, this paper covers three families of methods based on their usage in the era of big data: blind source separation (BSS), manifold learning (MfL), and Neural Networks (NNs), from shallow to deep architectures.

A survey of unsupervised generative models for exploratory data analysis and representation learning / M. Abukmeil, S. Ferrari, A. Genovese, V. Piuri, F. Scotti. - In: ACM COMPUTING SURVEYS. - ISSN 0360-0300. - 54:5(2021), pp. 99.1-99.40. [10.1145/3450963]

A survey of unsupervised generative models for exploratory data analysis and representation learning

M. Abukmeil
Primo
;
S. Ferrari
Secondo
;
A. Genovese;V. Piuri
Penultimo
;
F. Scotti
Ultimo
2021

Abstract

For more than a century, the methods for data representation and the exploration of the intrinsic structures of data have developed remarkably and consist of supervised and unsupervised methods. However, recent years have witnessed the flourishing of big data, where typical dataset dimensions are high and the data can come in messy, incomplete, unlabeled, or corrupted forms. Consequently, discovering the hidden structure buried inside such data becomes highly challenging. From this perspective, exploratory data analysis (EDA) plays a substantial role in learning the hidden structures that encompass the significant features of the data in an ordered manner by extracting patterns and testing hypotheses to identify anomalies. Unsupervised generative learning (UGL) models are a class of Machine Learning (ML) models characterized by their potential to reduce the dimensionality, discover the exploratory factors, and learn representations without any predefined labels; moreover, such models can generate the data from the reduced factors’ domain. The beginner researchers can find in this survey the recent UGL models for the purpose of data exploration and learning representations; specifically, this paper covers three families of methods based on their usage in the era of big data: blind source separation (BSS), manifold learning (MfL), and Neural Networks (NNs), from shallow to deep architectures.
Blind Source Separation; Manifold Learning; Neural Networks; Exploratory Data Analysis; Representation Learning; Explainable Machine Learning; Unsupervised Deep Learning
Settore INF/01 - Informatica
Settore ING-INF/05 - Sistemi di Elaborazione delle Informazioni
   Multi-Owner data Sharing for Analytics and Integration respecting Confidentiality and Owner control (MOSAICrOWN)
   MOSAICrOWN
   EUROPEAN COMMISSION
   H2020
   825333

   High quality Open data Publishing and Enrichment (HOPE)
   HOPE
   MINISTERO DELL'ISTRUZIONE E DEL MERITO
   2017MMJJRE_003

   Machine Learning-based, Networking and Computing Infrastructure Resource Management of 5G and beyond Intelligent Networks (MARSAL)
   MARSAL
   EUROPEAN COMMISSION
   H2020
   101017171
2021
Article (author)
File in questo prodotto:
File Dimensione Formato  
csur21main.pdf

accesso aperto

Tipologia: Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione 769.91 kB
Formato Adobe PDF
769.91 kB Adobe PDF Visualizza/Apri
3450963.pdf

accesso riservato

Tipologia: Publisher's version/PDF
Dimensione 553.74 kB
Formato Adobe PDF
553.74 kB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/815200
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 31
  • ???jsp.display-item.citation.isi??? 20
social impact