作者: Damiano Verda , Stefano Parodi , Enrico Ferrari , Marco Muselli
DOI: 10.1186/S12859-019-2953-8
关键词: Set (abstract data type) 、 Machine learning 、 Logic learning machine 、 Artificial intelligence 、 Artificial neural network 、 Support vector machine 、 Cancer 、 Decision tree 、 Cross-validation 、 Computer science
摘要: Logic Learning Machine (LLM) is an innovative method of supervised analysis capable constructing models based on simple and intelligible rules. In this investigation the performance LLM in classifying patients with cancer was evaluated using a set eight publicly available gene expression databases for diagnosis. accuracy assessed by summary ROC curve (sROC) estimated area under sROC (sAUC). Its compared cross validation that standard methods, namely: decision tree, artificial neural network, support vector machine (SVM) k-nearest neighbor classifier. showed excellent (sAUC = 0.99, 95%CI: 0.98–1.0) outperformed any other except SVM. new powerful tool data Simple rules generated could contribute to better understanding biology, potentially addressing therapeutic approaches.