會(huì)議進(jìn)程薄疚，包括三個(gè)平行軌道：

The conference program, with its three parallel tracks - the Research Track, the Applied Data Science Track and the Applied Invited Speakers Track - brings the two groups together.

（1）研究軌道昧捷；

（2）應(yīng)用數(shù)據(jù)科學(xué)軌道劝贸；

（3）應(yīng)用邀請(qǐng)演講者軌道 - 將兩個(gè)小組合并在一起蝇狼。

The conference this year continues with its tradition of a strong tutorial and workshop program on leading edge issues of data mining during the first two days of the program. The last three days are devoted to contributed technical papers, describing both novel, important research contributions, and deployed, innovative solutions.

前兩天：教程+數(shù)據(jù)挖掘前沿問(wèn)題

最后三天：重要文獻(xiàn)——描述新穎透敌、重要的研究成果，以及創(chuàng)新解決方案定页。

內(nèi)容：

paper List : http://www.kdd.org/kdd2017/accepted-papers

Three keynote talks, by Cynthia Dwork, Bin Yu, and Renée J. Miller touch on some of the hard, emerging issues before the field of data mining.

1. 三個(gè)主題演講——數(shù)據(jù)挖掘領(lǐng)域面臨的新興難題趟薄。

（1）What’s Fair?——Cynthia Dwork (Microsoft Research & Harvard University)

（2）The Future of Data Integration?

數(shù)據(jù)集成的未來(lái)——Renée J. Miller (University of Toronto)

（3）Three Principles of Data Science: Predictability, Stability and Computability

數(shù)據(jù)科學(xué)的三個(gè)原則：可預(yù)測(cè)性，穩(wěn)定性和可計(jì)算性——Bin Yu (University of California, Berkeley)

2. 12個(gè) Applied Invited Talks

（1）Foreword to the Applied Data Science – Invited Talks Track at KDD-2017

應(yīng)用數(shù)據(jù)科學(xué)前言 - KDD-2017特邀報(bào)告

（2）More than the Sum of its Parts: Building Domino Data Lab

不僅僅是相加：構(gòu)建Domino數(shù)據(jù)實(shí)驗(yàn)室——Eduardo Ari?o de la Rubia (Domino Data Lab)

（3）Mining Big Data in Neuro Genetics to Understand Muscular Dystrophy

挖掘神經(jīng)遺傳學(xué)中的大數(shù)據(jù)來(lái)了解肌營(yíng)養(yǎng)不良癥——Andy Berglund (University of Florida

（4）Industrial Machine Learning

工業(yè)機(jī)器學(xué)習(xí)——Josh Bloom (GE)

（5）Behavior Informatics to Discover Behavior Insight for Active and Tailored Client Management

行為信息學(xué)進(jìn)行行為洞察典徊，用于主動(dòng)和定制的客戶端管理——Longbing Cao (University of Technology Sydney）

（6）It Takes More than Math and Engineering to Hit the Bullseye with Data

擊中數(shù)據(jù)靶心不僅需要數(shù)學(xué)和工程——Paritosh Desai (Target)

（7）Planning and Learning under Uncertainty: Theory and Practice

不確定性下的規(guī)劃與學(xué)習(xí)：理論與實(shí)踐——Jonathan P. How (Massachusetts Institute of Technology)

（8）Big Data in Climate: Opportunities and Challenges for Machine Learning

氣候大數(shù)據(jù)：機(jī)器學(xué)習(xí)的機(jī)遇和挑戰(zhàn)——Anuj Karpatne, Vipin Kumar (University of Minnesota)

（9）Addressing Challenges with Big Data for Media Measurement

應(yīng)對(duì)大數(shù)據(jù)媒體測(cè)量挑戰(zhàn)——Mainak Mazumdar (Nielsen)

（10）Machine Learning Software in Practice: Quo Vadis?

機(jī)器學(xué)習(xí)軟件的實(shí)踐：Quo Vadis杭煎？——Szilárd Pafka (Epoch)

（11）Designing AI at Scale to Power Everyday Life

設(shè)計(jì)人工智能以幫助日常生活——Rajesh Parekh (Facebook)

（12）Spaceborne Data Enters the Mainstream

星載數(shù)據(jù)進(jìn)入主流——David Potere (Tellus Laboratories)

3. KDD 2017 Panels（人工智能相關(guān)）

（1）Benchmarks and Process Management in Data Science: Will We Ever Get Over the Mess?

數(shù)據(jù)科學(xué)中的基準(zhǔn)測(cè)試和流程管理：我們能否克服困難？——Usama M. Fayyad (Open Insights), Arno Candel (H2O.ai, Inc.), Eduardo Ari?o de la Rubia (Domino Data Lab),Szilárd Pafka (Epoch), Anthony Chong (IKASI), Jeong-Yoon Lee (Microsoft)

（2）The Future of Artificially Intelligent Assistants

人工智能助手的未來(lái)——Muthu Muthukrishnan (Rutgers University), Andrew Tomkins, Larry Heck (Google), Alborz Geramifard (Amazon), Deepak Agarwal (LinkedIn)

4.KDD 2017 Research Papers (Oral Papers) 研究文獻(xiàn)

（1）Learning Certifiably Optimal Rule Lists

學(xué)習(xí)可證明的最優(yōu)規(guī)則列表——Elaine Angelino (University of California, Berkeley),Nicholas Larus-Stone, Daniel Alabi, Margo Seltzer (Harvard University), Cynthia Rudin (Duke University)

（2）Improved Degree Bounds and Full Spectrum Power Laws in Preferential Attachment Networks

在優(yōu)先附著網(wǎng)絡(luò)中改進(jìn)度邊界和全譜冪律——Chen Avin, Zvi Lotker (Ben Gurion University of the Negev),Yinon Nahum, David Peleg (Weizmann Institute of Science)

（3）Unsupervised Network Discovery for Brain Imaging Data

腦成像數(shù)據(jù)的無(wú)監(jiān)督網(wǎng)絡(luò)發(fā)現(xiàn)——Zilong Bai (University of California, Davis), Peter Walker, Anna Tschiffely (Naval Medical Research Center),Fei Wang (Cornell University), Ian Davidson (University of California, Davis)

（4）Patient Subtyping via Time-Aware LSTM Networks

病人分類卒落，通過(guò)時(shí)間感知的LSTM網(wǎng)絡(luò)——Inci M. Baytas (Michigan State University), Cao Xiao (IBM T. J. Watson Research Center),Xi Zhang, Fei Wang (Cornell University), Anil K. Jain, Jiayu Zhou (Michigan State University)

（5）Robust Top-k Multiclass SVM for Visual Category Recognition

穩(wěn)健Top-k多類SVM羡铲，用于視覺(jué)分類識(shí)別——Xiaojun Chang (Carnegie Mellon University), Yao-Liang Yu (University of Waterloo),Yi Yang (University of Technology Sydney)

（6）KATE: K-Competitive Autoencoder for Text

KATE：文本K-競(jìng)爭(zhēng)自動(dòng)編碼器——Yu Chen, Mohammed J. Zaki (Rensselaer Polytechnic Institute)

（7）A Minimal Variance Estimator for the Cardinality of Big Data Set Intersection

大數(shù)據(jù)集交叉基數(shù)的最小方差估計(jì)——Reuven Cohen, Liran Katzir, Aviv Yehezkel (Technion)

（8）HyperLogLog Hyperextended: Sketches for Concave Sublinear Frequency Statistics

HyperLogLog Hyperextended：用于凹次線性頻率統(tǒng)計(jì)的草圖——Edith Cohen (Google Research)

（9）Fast Enumeration of Large k-Plexes

Large k-Plexes的快速枚舉——Alessio Conte (University of Pisa), Donatella Firmani (Roma Tre University),Caterina Mordente (Be Think Solve Execute), Maurizio Patrignani, Riccardo Torlone (Roma Tre University)

（10）Matrix Profile V: A Generic Technique to Incorporate Domain Knowledge into Motif Discovery

矩陣Profile V：將領(lǐng)域知識(shí)合并到Motif發(fā)現(xiàn)中的一種通用技術(shù)

（10）metapath2vec: Scalable Representation Learning for Heterogeneous Networks

metapath2vec：異構(gòu)網(wǎng)絡(luò)的可擴(kuò)展表示學(xué)習(xí)

（11）Ego-Splitting Framework: from Non-Overlapping to Overlapping Clusters

自分割框架：從非重疊到重疊的集群

（12）Contextual Motifs: Increasing the Utility of Motifs using Contextual Data

上下文的圖案：使用上下文數(shù)據(jù)增加圖案效用

（13）Unsupervised P2P Rental Recommendations via Integer Programming

無(wú)監(jiān)督的P2P出租推薦，通過(guò)整數(shù)編程

（14）The Co-Evolution Model for Social Network Evolving and Opinion Migration

社會(huì)網(wǎng)絡(luò)演進(jìn)和意見遷移的共同演化模型

（15）Groups-Keeping Solution Path Algorithm for Sparse Regression with Automatic Feature Grouping

歸分組解決路徑算法导绷，用于基于自動(dòng)特征分組的稀疏回歸

（16）Clustering Individual Transactional Data for Masses of Users

用戶群體的單個(gè)交易數(shù)據(jù)聚類

（17）Network Inference via the Time-Varying Graphical Lasso

通過(guò)時(shí)變圖套索進(jìn)行網(wǎng)絡(luò)推斷

（18）Efficient Correlated Topic Modeling with Topic Embedding

有效的相關(guān)主題建模與主題嵌入

（19）Accelerating Innovation Through Analogy Mining

通過(guò)類比挖掘加速創(chuàng)新

（20）Communication-Efficient Distributed Block Minimization for Nonlinear Kernel Machines

通信高效分布?jí)K最小化犀勒，用于非線性核機(jī)制

（21）A Hierarchical Algorithm for Extreme Clustering

一種極端聚類的分層算法

（21）Estimating Treatment Effect in the Wild via Differentiated Confounder Balancing

通過(guò)差異配種平衡屎飘，評(píng)估野外治療效果

（22）The Selective Labels Problem: Evaluating Algorithmic Predictions in the Presence of Unobservables

選擇性標(biāo)簽問(wèn)題：在不可觀察情況下評(píng)估算法預(yù)測(cè)

（23）Constructivism Learning: A Learning Paradigm for Transparent Predictive Analytics

建構(gòu)主義學(xué)習(xí)：透明預(yù)測(cè)分析的學(xué)習(xí)范式

（24）Is the Whole Greater Than the Sum of Its Parts?

（25）Collaborative Variational Autoencoder for Recommender Systems

用于推薦系統(tǒng)的協(xié)作變分自動(dòng)編碼器

（26）Linearized GMM Kernels and Normalized Random Fourier Features

線性化GMM核與歸一化隨機(jī)傅立葉特征

（27）Discrete Content-aware Matrix Factorization

感知內(nèi)容的矩陣分解

（28）Effective and Real-time In-App Activity Analysis in Encrypted Internet Traffic Streams

加密的互聯(lián)網(wǎng)業(yè)務(wù)流中有效和實(shí)時(shí)的應(yīng)用內(nèi)活動(dòng)分析

（29）Functional Annotation of Human Protein Coding Isoforms via Non-convex Multi-Instance Learning

人類蛋白質(zhì)編碼亞型的非凸多實(shí)例學(xué)習(xí)功能注釋

（30）Discovering Reliable Approximate Functional Dependencies

發(fā)現(xiàn)可靠的近似函數(shù)依賴

（21）Towards an Optimal Subspace for K-Means

（22）SPARTan: Scalable PARAFAC2 for Large & Sparse Data

用于大型稀疏數(shù)據(jù)的可擴(kuò)展的PARAFAC2

（23）struc2vec: Learning Node Representations from Structural Identity

（24）Similarity Forests

（25）Structural Deep Brain Network Mining

（26）On Finding Socially Tenuous Groups for Online Social Networks

（27）A Local Algorithm for Structure-Preserving Graph Cut