首页|Department of Computer Science and Information Technology Researcher Details New Studies and Findings in the Area of Machine Learning (Evaluation of machine lea rning models that predict lncRNA subcellular localization)
Department of Computer Science and Information Technology Researcher Details New Studies and Findings in the Area of Machine Learning (Evaluation of machine lea rning models that predict lncRNA subcellular localization)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Investigators discuss new findings in artificial intelligence. According to news reporting out of the Department of Co mputer Science and Information Technology by NewsRx editors, research stated, "T he lncATLAS database quantifies the relative cytoplasmic versus nuclear abundanc e of long non-coding RNAs (lncRNAs) observed in 15 human cell lines." Financial supporters for this research include National Science Foundation. Our news editors obtained a quote from the research from Department of Computer Science and Information Technology: "The literature describes several machine le arning models trained and evaluated on these and similar datasets. These reports showed moderate performance, e.g. 72-74% accuracy, on test subset s of the data withheld from training. In all these reports, the datasets were fi ltered to include genes with extreme values while excluding genes with values in the middle range and the filters were applied prior to partitioning the data in to training and testing subsets. Using several models and lncATLAS data, we show that this ‘middle exclusion' protocol boosts performance metrics without boosti ng model performance on unfiltered test data. We show that various models achiev e only about 60% accuracy when evaluated on unfiltered lncRNA data ."
Department of Computer Science and Infor mation TechnologyCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
