A practical approach to novel class discovery in tabular data

• Published in: Data mining and knowledge discovery Vol. 38; no. 4; pp. 2087 - 2116
• Main Authors: Colin, Troisemaine, Alexandre, Reiffers-Masson, Stéphane, Gosselin, Vincent, Lemaire, Sandrine, Vaton
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2024; Springer Nature B.V; Springer
• Series: ECML PKDD 2024
• Subjects: Algorithms; Artificial Intelligence; Chemistry and Earth Sciences; Clustering; Computer Science; Computer vision; Data Mining and Knowledge Discovery; Information Storage and Retrieval; Knowledge; Methods; Physics; Statistics for Engineering; Tables (data); Open world learning; Clustering; Transfer learning; Novel class discovery; Tabular data; transfer learning; clustering; novel class discovery; novel class discovery clustering tabular data open world learning transfer learning; tabular data; open world learning

The problem of novel class discovery (NCD) consists in extracting knowledge from a labeled set of known classes to accurately partition an unlabeled set of novel classes. While NCD has recently received a lot of attention from the community, it is often solved on computer vision problems and under u...