Work

Developing data mining methods for finding causal relationships involving multiple factors

Year: 2017
Type: dissertation
Abstract: In the past decades, causal discovery from observational data has been studied widely. However, most existing research has focused on detecting causal relationship between a single factor and the outcome, which may not be sufficient for reasoning about the cause of a particular effect. In this thesis, we study complex causal relationships with multiple factors involved, and have developed the methods for discovering combined causes, cause interactions and context specific causes from data. We have also developed a causality based classification model for predictions with causal interpretations.
Author: Ma, Saisai,
Language: English
Open Access status: green
OpenAlex ID: W7222370803
Cites: 125
Cited by:
Related to: 10