光華講壇——社會(huì)名流與企業(yè)家論壇第6590期
主題：A unified performance measurement framework for classification algorithms 一種針對(duì)分類(lèi)算法的統(tǒng)一性能度量框架
主講人：美國(guó)特拉華大學(xué) 陳濱桐教授
主持人：工商管理學(xué)院 李偉教授
時(shí)間：10月11日 14:30-16:30
舉辦地點(diǎn)：頤德樓H308
主辦單位：工商管理學(xué)院 科研處
主講人簡(jiǎn)介：
Bintong Chen graduated from Shanghai Jiaotong University with dual B.S. degrees in ship-building/naval architecture and electrical engineering. He received M.S. in systems engineering and Ph.D. in operations management/research from the Wharton School, the University of Pennsylvania. He is currently a professor of the Lerner College of Business and Economics and the director of the Institute for Financial Services Analytics at University of Delaware. He published many high quality papers in the area of optimization theory, data-driven analytics, and business applications. He received many outstanding research and teaching awards in institutions he worked. Professor Chen consulted many international companies, including JP Morgan Chase, Agriculture Bank of China, AT&T, Burlington Northern Rail, Delaware Department of Transportation, Nordstrom, and AstraZeneca, etc. He was a board member for APICS, the largest supply chain professional association in North American.
陳濱桐，特拉華大學(xué)勒納商學(xué)與經(jīng)濟(jì)學(xué)院教授，于1985年獲上海交通大學(xué)船舶結(jié)構(gòu)與海洋工程系、電子工程系雙學(xué)士學(xué)位，1987年獲賓夕法尼亞大學(xué)工程與應(yīng)用科學(xué)學(xué)院系統(tǒng)工程系碩士學(xué)位，1990年獲賓夕法尼亞大學(xué)沃頓商學(xué)院運(yùn)籌與信息管理系博士學(xué)位?，F(xiàn)為特拉華大學(xué)勒納商學(xué)與經(jīng)濟(jì)學(xué)院金融服務(wù)分析中心主任、博士項(xiàng)目主任。陳濱桐教授在管理科學(xué)、運(yùn)籌學(xué)和運(yùn)營(yíng)管理領(lǐng)域取得了豐碩的研究成果，有多項(xiàng)研究成果發(fā)表在相關(guān)領(lǐng)域全球頂級(jí)學(xué)術(shù)期刊，包括《Management Science》和《Operations Research》，曾在諸多國(guó)際期刊編輯委員會(huì)中任職，包括《POM》和《Omega》期刊，并在所工作的大學(xué)中獲得了許多杰出的研究和教學(xué)獎(jiǎng)項(xiàng)。陳教授曾為許多國(guó)際公司提供咨詢(xún)服務(wù)，包括摩根大通、中國(guó)農(nóng)業(yè)銀行、美國(guó)電話(huà)電報(bào)公司、伯靈頓北方鐵路、特拉華州交通部、諾德斯特龍和阿斯利康等。他曾是北美最大的供應(yīng)鏈專(zhuān)業(yè)協(xié)會(huì)——美國(guó)生產(chǎn)與庫(kù)存管理協(xié)會(huì)(APICS)的董事會(huì)成員。
內(nèi)容簡(jiǎn)介:
Many classification algorithm performance measures have been independently proposed and studied. Two questions arise about these measurements: (1) When do they measure the maximum potential of a classification algorithm? (2) How to efficiently identify and calculate the maximum performance for each measurement? We propose a unified theoretical framework that includes all existing performance measures and curves as special cases. To answer the first question, we investigate two variable transformations and apply theoretical findings to various measures and performance curves. To answer the second question, we classify all performance measures into three categories: monotone measures, unimodal measures, and multi-modal measures, based on the process to search for the optimal threshold. The unified framework allows us to systematically analyze the properties of classification algorithm performance measures and provides guidance to design new performance measures.
許多分類(lèi)算法的性能度量方法都是被獨(dú)立提出和研究。關(guān)于這些度量方法存在兩個(gè)問(wèn)題：（1）它們?cè)诤螘r(shí)度量分類(lèi)算法的最大潛力？（2）如何高效識(shí)別并計(jì)算每個(gè)度量方法的最大性能？我們提出了一個(gè)統(tǒng)一的理論框架，其中包括所有現(xiàn)有的性能度量方法和曲線(xiàn)作為特例。為回答第一個(gè)問(wèn)題，我們研究了兩個(gè)變量變換，并將理論發(fā)現(xiàn)應(yīng)用到了各種度量方法和性能曲線(xiàn)上。為回答第二個(gè)問(wèn)題，我們基于搜索最優(yōu)閾值的過(guò)程將所有性能度量方法分為了三類(lèi)：?jiǎn)握{(diào)度量、單峰度量和多峰度量。這個(gè)統(tǒng)一的框架使我們能夠系統(tǒng)地分析分類(lèi)算法性能度量的屬性，并為設(shè)計(jì)新的性能度量方法提供指導(dǎo)。
