Discovering Latent Structures: Experience with the CoIL Challenge 2000 Data Set

We present a case study to demonstrate the possibility of discovering complex and interesting latent structures using hierarchical latent class (HLC) models. A similar effort was made earlier [6], but that study involved only small applications with 4 or 5 observed variables. Due to recent progress in algorithm research, it is now possible to learn HLC models with dozens of observed variables. We have successfully analyzed a version the CoIL Challenge 2000 data set that consists of 42 observed variable. The model obtained consists of 22 latent variables, and its structure is intuitively appealing.

• Type: Conference
• Year: 2007
• Where: ICCS
• Authors: Nevin Lianwen Zhang
• Keywords: Applied Computing, Hierarchical Latent Class, ICCS 2007, Interesting Latent Structures, Observed Variable