讲座主题：Machine Learning and Transport Simulations for Groundwater Anomaly Detection

主讲人：Jiangguo Liu

工作单位：Colorado State University

讲座时间：2019年5月29日10点30

讲座地点：数学院341

主办单位：烟台大学数学与信息科学学院

内容摘要：

In this talk, we present studies on models and algorithms for groundwater anomaly detection. Specifically, conductivity along with four other surrogates are used for identifying anomaly in groundwater, the one-class support vector machine (SVM) technique is utilized for model training, and real data from “Colorado Water Watch” is used for testing the models and algorithms. Design of code modules in Python will be briefly discussed. Since groundwater contamination rarely happens in reality, we also use data from numerical simulations of flow and transport in porous media to test the anomaly detection code modules. This is a joint work with Ken Carlson, Jianli Gu, and Huishu Li at Colorado State University.

主讲人介绍：

Jiangguo (James) Liu刘江国，美国科罗拉多州立大学数学系教授，博士生导师。曾任美国工业与应用数学学会中部地区分会主席，现任 Journal of Computational and Applied Mathematics 杂志编辑。主要研究兴趣为数值分析，科学计算，及生物数学。已在 SIAM Journal on Numerical Analysis, SIAM Journal on Scientific Computing, Journal of Computational Physics 等杂志上发表论文40多篇。所主持研究项目受美国国家自然科学基金资助。