数据挖掘20位学术新秀, 16名华人入选, 科大亮丽在现今如火如荼的大数据时代,从学术研究到工业应用,从科学发现到医疗卫生服务,数据挖掘吸引了来自机器学习、统计、数据库、万维网、生物信息学、多媒体、自然语言处理、人机交互、社会网络计算、以及高性能计算等众多领域的专家和学者。ACM SIGKDD(简称KDD)作为数据挖掘领域的顶级国际会议,其评选出的20位KDD新星,也格外引人注目。 KDD论文录取率极低,百度表现不俗,但中国高校鲜有入选。不过20位KDD新星,却以华人学者为主,更有6位科大校友 近日,数据挖掘领域的顶级国际会议ACMSIGKDD (KDD2016)在美国旧金山召开。此次KDD大会堪称史上最大规模,有超过2700名来自学术界和工业界人士参与此次盛会,共吸引了1115篇投稿。其中研究专题投稿论文784篇,最终仅有142篇录用;应用数据科学专题投稿论文331篇,录用仅有66篇。可见大会论文竞争之激烈,水平之高超。此外,由于KDD的应用数据科学专题放开了过去只接收工业界投稿的限制,吸引了大量来自高校的学者投稿,因此今年的331篇投稿较之去年的189篇投稿有了大幅度的提升。 从以下录取论文的单位分布来看,Linkedin独领风骚,微软紧随其后,IBM和百度也表现不俗,显示工业界的强大实力。而领先的高校包括芝加哥大学、密歇根大学、和卡内基梅隆大学。国内高校没有一家入选两篇论文。 基于KDD大会的作者及论文等相关大数据信息,微软学术搜索评选出近6年里上升最快的二十位KDD学术新星。虽然国内研究单位在KDD入选论文方面尚有差距,引领数据挖掘潮流的学术新星,却有16位华人青年学者,其中更有6位科大校友,显示中国人在这一领域的十足后劲。下面是这些入选新星的介绍,和他(她)们的工作: 孙欢,加州大学圣塔芭芭拉分校 2010年毕业于中国科学技术大学 相关工作: Exploiting RelevanceFeedback in Knowledge Graph Search; Analyzing ExpertBehaviors in Collaborative Networks; Network Mining andAnalysis for Social Applications; Synthetic ReviewSpamming and Defense
Mohammad TahaBahadori University of SouthernCalifornia 相关工作: Multi-layerRepresentation Learning for Medical Concepts; FLASH: Fast BayesianOptimization for Data Analytic Pipelines; Deep ComputationalPhenotyping; FBLG: a simple andeffective approach for temporal dependence discovery from time series data 任翔,伊利诺伊大学香槟分校 2012年毕业于浙江大学 (本科) 相关工作: Label Noise Reduction inEntity Typing by Heterogeneous Partial-Label Embedding; ClusType: EffectiveEntity Recognition and Typing by Relation Phrase-Based Clustering; Automatic EntityRecognition and Typing from Massive Text Corpora: A Phrase and Network MiningApproach; ClusCite: effectivecitation recommendation by information network-based clustering
Bee-Chung Chen 1998年毕业于国立台湾大学 (本科) 相关工作: Ranking UniversitiesBased on Career Outcomes of Graduates; An Empirical Study onRecommendation with Multiple Types of Feedback; GLMix: GeneralizedLinear Mixed Models For Large-Scale Response Prediction; Personalizing LinkedInFeed; Activity ranking inLinkedIn feed; SpatiotemporalPersonalized Recommendation of Social Media Content
刘洪甫,东北大学 毕业于北京航空航天大学 相关工作: Infinite Ensemble forImage Clustering; Spectral EnsembleClustering; SEA: a system for eventanalysis on chinese tweets
曾春秋,佛罗里达国际大学 2009年毕业于四川大学 (硕士) 相关工作: Online Context-AwareRecommendation with Time Varying Multi-Armed Bandit; Applying data miningtechniques to address critical process optimization needs in advancedmanufacturing; FIU-Miner: a fast,integrated, and user-friendly system for data mining in distributed environment
付燕杰,罗格斯大学 2008年毕业于中国科学技术大学 (本科) 相关工作: Days on Market:Measuring Liquidity in Real Estate Markets; Real Estate Ranking viaMixed Land-use Latent Models; Exploiting geographicdependencies for real estate appraisal: a mutual perspective of ranking andclustering; Learning geographicalpreferences for point-of-interest recommendation
Alex Deng,微软 2006年毕业于浙江大学 (本科) 相关工作: Data-Driven MetricDevelopment for Online Controlled Experiments: Seven Lessons Learned; Seven rules of thumb forweb site experimenters; Online controlledexperiments at large scale; Trustworthy onlinecontrolled experiments: five puzzling outcomes explained
赵亮,弗吉尼亚理工大学 相关工作: Hierarchical IncompleteMulti-source Feature Learning for Spatiotemporal Event Forecasting; Multi-Task Learning forSpatio-Temporal Event Forecasting; STED: semi-supervisedtargeted-interest event detectionin in twitter
Evangelos E. Papalexakis Carnegie Mellon University 相关工作: Whither Social Networksfor Web Search?; Good-enough brain model:challenges, algorithms and discoveries in multi-subject experiments; GigaTensor: scalingtensor analysis up by 100 times - algorithms and discoveries
袁晶,微软 2012年毕业于中国科学技术学 (博士) 相关工作: Contextual IntentTracking for Personal Assistants; Collaborative KnowledgeBase Embedding for Recommender Systems; Regularity andConformity: Location Prediction Using Heterogeneous Mobility Data
Yang Zhou Georgia Institute of Technology 相关工作: IntegratingVertex-centric Clustering with Edge-centric Clustering for Meta Path GraphAnalysis; Activity-edge centricmulti-label classification for mining heterogeneous information networks; Social influence basedclustering of heterogeneous information networks; Multimedia features forclick prediction of new ads in display advertising
陈稳霖,华盛顿大学圣路易斯分校 2011年毕业于中国科学技术大学 (本科) 相关工作: CompressingConvolutional Neural Networks in the Frequency Domain; Optimal ActionExtraction for Random Forests and Boosted Trees; Fast flux discriminantfor large-scale sparse nonlinear classification; Density-based logisticregression; An integrated datamining approach to real-time clinical monitoring and deterioration warning
Shuo Xiang,亚利桑那州立大学 2010年毕业于南京航空航天大学 (硕士) 相关工作: Simultaneous feature andfeature group selection through hard thresholding; Multi-source learningwith block-wise missing data for Alzheimer's disease prediction; Robust principalcomponent analysis via capped norms; Optimal exact leastsquares rank minimization
祝恒书,百度 2014年毕业于中国科学技术大学 (博士) 相关工作: Recruitment Market TrendAnalysis with Sequential Latent Variable Models; Days on Market:Measuring Liquidity in Real Estate Markets; Talent Circle Detectionin Job Transition Networks; Point-of-InterestRecommendations: Learning Potential Check-ins from Friends; Taxi Driving BehaviorAnalysis in Latent Vehicle-to-Vehicle Networks: A Social Influence Perspective; Real Estate Ranking viaMixed Land-use Latent Models; Discerning TacticalPatterns for Professional Soccer Teams: An Enhanced Topic Model with Applications; A cost-effectiverecommender system for taxi drivers; Mobile apprecommendations with security and privacy awareness
张伟楠,伦敦大学学院 2011年毕业于上海交通大学 (本科) 相关工作: Bid-aware GradientDescent for Unbiased Learning with Censored Data in Display Advertising; Statistical ArbitrageMining for Display Advertising; Annotating Needles inthe Haystack without Looking: Product Information Extraction from Emails; Optimal real-timebidding for display advertising; Joint optimization ofbid and budget allocation in sponsored search
Paulo Shakarian Arizona State University, Military Academy 相关工作: Early Identification ofViolent Criminal Gang Members; Mining for CausalRelationships: A Data-Driven Study of the Islamic State; Reducing gang violencethrough network influence based targeting of social programs; Mining forgeographically disperse communities in social networks by leveraging distancemodularity
刘传仁,罗格斯大学 2008年毕业于中国科学技术大学 (本科) 相关工作: Catch Me If You Can:Detecting Pickpocket Suspects from Large-Scale Transit Records; UnifiedPoint-of-Interest Recommendation with Temporal Interval Assessment; Data-driven AutomaticTreatment Regimen Development and Recommendation; Temporal Phenotyping fromLongitudinal Electronic Health Records: A Graph Based Framework; Temporal skeletonizationon sequential data: patterns, categorization, and visualization; Proactive workflowmodeling by stochastic processes with application to healthcare operation and management; A taxi businessintelligence system
Yasuko Matsubara Kyoto University, Kumamoto University 相关工作: Regime Shifts inStreams: Real-time Forecasting of Co-evolving Time Sequences; FUNNEL: automatic miningof spatially coevolving epidemics; Rise and fall patternsof information diffusion: model and implications; Fast mining andforecasting of complex time-stamped events
蒋朦,伊利诺伊大学香槟分校 2010年毕业于清华大学 (本科) 相关工作: CatchTartan:Representing and Summarizing Dynamic Multicontextual Behaviors; CatchSync: catchingsynchronized behavior in large directed graphs; FEMA: flexibleevolutionary multi-faceted analysis for dynamic behavioral pattern discovery 下一篇: "中国诺贝尔"揭晓, 薛其坤获奖
|