報(bào)告題目：Debugging hierarchical relations in large biomedical ontologies: finding needles in a haystack

報(bào)告人: 崔麗聰

講座日期：2021-09-23

講座時(shí)間：11:00

報(bào)告地點(diǎn)：騰訊會(huì)議871656108

主辦單位：數(shù)學(xué)與統(tǒng)計(jì)學(xué)院

講座人簡(jiǎn)介：

崔麗聰(https://sbmi.uth.edu/cuilab/index.htm)，學(xué)士和碩士畢業(yè)于陜西師范大學(xué)，博士畢業(yè)于美國(guó)凱斯西儲(chǔ)大學(xué)(Case Western Reserve University)，曾任美國(guó)肯塔基大學(xué)(University of Kentucky)計(jì)算機(jī)系助理教授，現(xiàn)任德克薩斯大學(xué)休斯頓健康科學(xué)中心(University of Texas Health Science Center at Houston)助理教授, 博士生導(dǎo)師，研究方向?yàn)樯镝t(yī)學(xué)信息。在Journal of the American Medical Informatics Association, Journal of Biomedical Informatics, Bioinformatics, IEEE Journal of Biomedical and Health Informatics等國(guó)際期刊和會(huì)議發(fā)表論文70余篇，Google Scholar引用超過(guò)1200次。主持美國(guó)國(guó)家自然科學(xué)基金項(xiàng)目3項(xiàng)以及國(guó)立衛(wèi)生研究院項(xiàng)目4項(xiàng)，并參與國(guó)立衛(wèi)生研究院項(xiàng)目3項(xiàng)。

講座簡(jiǎn)介：

Ontologies have been used in a wide variety of biomedical applications including information extraction and retrieval, data integration and management, clinical decision support. Quality defects of biomedical ontologies, if not addressed, could affect all downstream applications that use them as knowledge sources. However, identification of potential quality defects is challenging due to the ever-growing size and complexity of biomedical ontologies (i.e., large and evolving graphs). We develop principled and machine learning-based approaches to effectively detect missing or erroneous hierarchical relations in large ontologies including SNOMED CT and Gene Ontology.