Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining最新文献

{"title":"The Fake vs Real Goods Problem: Microscopy and Machine Learning to the Rescue","authors":"Ashlesh Sharma, Vidyuth Srinivasan, Vishal Kanchan, L. Subramanian","doi":"10.1145/3097983.3098186","DOIUrl":"https://doi.org/10.1145/3097983.3098186","url":null,"abstract":"Counterfeiting of physical goods is a global problem amounting to nearly 7% of world trade. While there have been a variety of overt technologies like holograms and specialized barcodes and covert technologies like taggants and PUFs, these solutions have had a limited impact on the counterfeit market due to a variety of factors - clonability, cost or adoption barriers. In this paper, we introduce a new mechanism that uses machine learning algorithms on microscopic images of physical objects to distinguish between genuine and counterfeit versions of the same product. The underlying principle of our system stems from the idea that microscopic characteristics in a genuine product or a class of products (corresponding to the same larger product line), exhibit inherent similarities that can be used to distinguish these products from their corresponding counterfeit versions. A key building block for our system is a wide-angle microscopy device compatible with a mobile device that enables a user to easily capture the microscopic image of a large area of a physical object. Based on the captured microscopic images, we show that using machine learning algorithms (ConvNets and bag of words), one can generate a highly accurate classification engine for separating the genuine versions of a product from the counterfeit ones; this property also holds for \"super-fake\" counterfeits observed in the marketplace that are not easily discernible from the human eye. We describe the design of an end-to-end physical authentication system leveraging mobile devices, portable hardware and a cloud-based object verification ecosystem. We evaluate our system using a large dataset of 3 million images across various objects and materials such as fabrics, leather, pills, electronics, toys and shoes. The classification accuracy is more than 98% and we show how our system works with a cellphone to verify the authenticity of everyday objects.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124196886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Planning Bike Lanes based on Sharing-Bikes' Trajectories","authors":"Jie Bao, Tianfu He, Sijie Ruan, Yanhua Li, Yu Zheng","doi":"10.1145/3097983.3098056","DOIUrl":"https://doi.org/10.1145/3097983.3098056","url":null,"abstract":"Cycling as a green transportation mode has been promoted by many governments all over the world. As a result, constructing effective bike lanes has become a crucial task for governments promoting the cycling life style, as well-planned bike paths can reduce traffic congestion and decrease safety risks for both cyclists and motor vehicle drivers. Unfortunately, existing trajectory mining approaches for bike lane planning do not consider key realistic government constraints: 1) budget limitations, 2) construction convenience, and 3) bike lane utilization. In this paper, we propose a data-driven approach to develop bike lane construction plans based on large-scale real world bike trajectory data. We enforce these constraints to formulate our problem and introduce a flexible objective function to tune the benefit between coverage of the number of users and the length of their trajectories. We prove the NP-hardness of the problem and propose greedy-based heuristics to address it. Finally, we deploy our system on Microsoft Azure, providing extensive experiments and case studies to demonstrate the effectiveness of our approach.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124444907","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","authors":"S. Matwin, Shipeng Yu, Faisal Farooq","doi":"10.1145/3097983","DOIUrl":"https://doi.org/10.1145/3097983","url":null,"abstract":"It is our great pleasure to welcome you to the 2017 ACM Conference on Knowledge Discovery and Data Mining -- KDD 2017. We hope that the content and the professional networking opportunities at KDD 2017 will help you to succeed professionally by enabling you to: identify new technology trends; learn from contributed papers, presentations, and posters; discover new tools, processes and practices; identify new job opportunities; and hire new team members. \u0000 \u0000The terms \"Data Science\", \"Data Mining\" and \"Big Data\" have, in the last few years, grown out of research labs and gained presence in the media and in everyday conversations. We also hear these terms on social media and from decision makers at various level of governments and corporations. The impact of these technologies is felt in almost every walk of life. Importantly, the current rapid progress in data science is facilitated by the timely sharing of newly discovered and developed representations and algorithms between those working in research and those interested in industrial deployment. It is the hallmark of KDD conferences in the past that they have been the bridge between theory and practise, the great facilitator and catalyst for this exchange. Researchers and practitioners meet in person and interact in a meaningful way over several days. 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. Participants are welcome to freely attend any track, and the events common for all tracks. \u0000 \u0000The 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. Three keynote talks, by Cynthia Dwork, Bin Yu, and Renee J. Miller touch on some of the hard, emerging issues before the field of data mining. With a growing industry around AI assistants, our KDD Panel brings together industry experts in this field to spawn discussions and an exchanges of ideas. We have an outstanding lineup of industry speakers sharing their experiences and expertise in deploying industrial data mining solutions. We continue a strong hands-on tutorial program, in which participants will learn how to use practical data science tools. In order to broaden the impact of KDD and to increase the participation of attendees who would greatly benefit from the conference but would have otherwise found it financially challenging to attend, we reserved a substantial budget for travel grants. KDD 2017 awarded a record USD 145k for student travel and also set aside USD 25k to enable smaller startups to attend. With the new \"Meet the Experts\" sessions, KDD 2017 also gives researchers and practitioners a unique opportunity to form professional networ","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114550670","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Hybrid Framework for Text Modeling with Convolutional RNN","authors":"Chenglong Wang, Feijun Jiang, Hongxia Yang","doi":"10.1145/3097983.3098140","DOIUrl":"https://doi.org/10.1145/3097983.3098140","url":null,"abstract":"In this paper, we introduce a generic inference hybrid framework for Convolutional Recurrent Neural Network (conv-RNN) of semantic modeling of text, seamless integrating the merits on extracting different aspects of linguistic information from both convolutional and recurrent neural network structures and thus strengthening the semantic understanding power of the new framework. Besides, based on conv-RNN, we also propose a novel sentence classification model and an attention based answer selection model with strengthening power for the sentence matching and classification respectively. We validate the proposed models on a very wide variety of data sets, including two challenging tasks of answer selection (AS) and five benchmark datasets for sentence classification (SC). To the best of our knowledge, it is by far the most complete comparison results in both AS and SC. We empirically show superior performances of conv-RNN in these different challenging tasks and benchmark datasets and also summarize insights on the performances of other state-of-the-arts methodologies.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116283185","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Practical Algorithm for Solving the Incoherence Problem of Topic Models In Industrial Applications","authors":"Amr Ahmed, James Long, Daniel Silva, Y. Wang","doi":"10.1145/3097983.3098200","DOIUrl":"https://doi.org/10.1145/3097983.3098200","url":null,"abstract":"Topic models are often applied in industrial settings to discover user profiles from activity logs where documents correspond to users and words to complex objects such as web sites and installed apps. Standard topic models ignore the content-based similarity structure between these objects largely because of the inability of the Dirichlet prior to capture such side information of word-word correlation. Several approaches were proposed to replace the Dirichlet prior with more expressive alternatives. However, this added expressivity comes with a heavy premium: inference becomes intractable and sparsity is lost which renders these alternatives not suitable for industrial scale applications. In this paper we take a radically different approach to incorporating word-word correlation in topic models by applying this side information at the posterior level rather than at the prior level. We show that this choice preserves sparsity and results in a graph-based sampler for LDA whose computational complexity is asymptotically on bar with the state of the art Alias base sampler for LDA cite{aliasLDA}. We illustrate the efficacy of our approach over real industrial datasets that span up to billion of users, tens of millions of words and thousands of topics. To the best of our knowledge, our approach provides the first practical and scalable solution to this important problem.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129494641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Post Processing Recommender Systems for Diversity","authors":"Arda Antikacioglu, R. Ravi","doi":"10.1145/3097983.3098173","DOIUrl":"https://doi.org/10.1145/3097983.3098173","url":null,"abstract":"Collaborative filtering is a broad and powerful framework for building recommendation systems that has seen widespread adoption. Over the past decade, the propensity of such systems for favoring popular products and thus creating echo chambers have been observed. This has given rise to an active area of research that seeks to diversify recommendations generated by such algorithms. We address the problem of increasing diversity in recom- mendation systems that are based on collaborative filtering that use past ratings to predict a rating quality for potential recommendations. Following our earlier work, we formulate recommendation system design as a subgraph selection problem from a candidate super-graph of potential recommendations where both diversity and rating quality are explicitly optimized: (1) On the modeling side, we define a new flexible notion of diversity that allows a system designer to prescribe the number of recommendations each item should receive, and smoothly penalizes deviations from this distribution. (2) On the algorithmic side, we show that minimum-cost network flow methods yield fast algorithms in theory and practice for designing recommendation subgraphs that optimize this notion of diversity. (3) On the empirical side, we show the effectiveness of our new model and method to increase diversity while maintaining high rating quality in standard rating data sets from Netflix and MovieLens.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132898702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Learning to Generate Rock Descriptions from Multivariate Well Logs with Hierarchical Attention","authors":"Bin Tong, Martin Klinkigt, Makoto Iwayama, T. Yanase, Yoshiyuki Kobayashi, Anshuman Sahu, Ravigopal Vennelakanti","doi":"10.1145/3097983.3098132","DOIUrl":"https://doi.org/10.1145/3097983.3098132","url":null,"abstract":"In the shale oil & gas industry, operators are looking toward big data analytics to optimize operations and reduce cost. In this paper, we mainly focus on how to assist operators in understanding the subsurface formation, thereby helping them make optimal decisions. A large number of geology reports and well logs describing the sub-surface have been accumulated over years. Issuing geology reports is more time consuming and depends more on the expertise of engineers than acquiring the well logs. To assist in issuing geology reports, we propose an encoder-decoder-based model to automatically generate rock descriptions in human-readable format from multivariate well logs. Due to the different formats of data, this task differs dramatically from image and video captioning. The challenges are how to model structured rock descriptions and leverage the information in multivariate well logs. To achieve this, we design a hierarchical structure and two forms of attention for the decoder. Extensive validations are conducted on public well data of North Dakota in the United States. We show that our model is effective in generating rock descriptions. The two forms of attention enable the provision of a better insight into relations between well-log types and rock properties with our model from a data-driven perspective.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130909814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Distributed Local Outlier Detection in Big Data","authors":"Yizhou Yan, Lei Cao, C. Kuhlman, Elke A. Rundensteiner","doi":"10.1145/3097983.3098179","DOIUrl":"https://doi.org/10.1145/3097983.3098179","url":null,"abstract":"In this work, we present the first distributed solution for the Local Outlier Factor (LOF) method -- a popular outlier detection technique shown to be very effective for datasets with skewed distributions. As datasets increase radically in size, highly scalable LOF algorithms leveraging modern distributed infrastructures are required. This poses significant challenges due to the complexity of the LOF definition, and a lack of access to the entire dataset at any individual compute machine. Our solution features a distributed LOF pipeline framework, called DLOF. Each stage of the LOF computation is conducted in a fully distributed fashion by leveraging our invariant observation for intermediate value management. Furthermore, we propose a data assignment strategy which ensures that each machine is self-sufficient in all stages of the LOF pipeline, while minimizing the number of data replicas. Based on the convergence property derived from analyzing this strategy in the context of real world datasets, we introduce a number of data-driven optimization strategies. These strategies not only minimize the computation costs within each stage, but also eliminate unnecessary communication costs by aggressively pushing the LOF computation into the early stages of the DLOF pipeline. Our comprehensive experimental study using both real and synthetic datasets confirms the efficiency and scalability of our approach to terabyte level data.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121037701","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Simpler The Better: A Unified Approach to Predicting Original Taxi Demands based on Large-Scale Online Platforms","authors":"Yongxin Tong, Yuqiang Chen, Zimu Zhou, Lei Chen, Jie Wang, Qiang Yang, Jieping Ye, Weifeng Lv","doi":"10.1145/3097983.3098018","DOIUrl":"https://doi.org/10.1145/3097983.3098018","url":null,"abstract":"Taxi-calling apps are gaining increasing popularity for their efficiency in dispatching idle taxis to passengers in need. To precisely balance the supply and the demand of taxis, online taxicab platforms need to predict the Unit Original Taxi Demand (UOTD), which refers to the number of taxi-calling requirements submitted per unit time (e.g., every hour) and per unit region (e.g., each POI). Predicting UOTD is non-trivial for large-scale industrial online taxicab platforms because both accuracy and flexibility are essential. Complex non-linear models such as GBRT and deep learning are generally accurate, yet require labor-intensive model redesign after scenario changes (e.g., extra constraints due to new regulations). To accurately predict UOTD while remaining flexible to scenario changes, we propose LinUOTD, a unified linear regression model with more than 200 million dimensions of features. The simple model structure eliminates the need of repeated model redesign, while the high-dimensional features contribute to accurate UOTD prediction. We further design a series of optimization techniques for efficient model training and updating. Evaluations on two large-scale datasets from an industrial online taxicab platform verify that LinUOTD outperforms popular non-linear models in accuracy. We envision our experiences to adopt simple linear models with high-dimensional features in UOTD prediction as a pilot study and can shed insights upon other industrial large-scale spatio-temporal prediction problems.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115262693","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"STAR: A System for Ticket Analysis and Resolution","authors":"Wubai Zhou, Wei Xue, Ramesh Baral, Qing Wang, Chunqiu Zeng, Tao Li, Jian Xu, Zheng Liu, L. Shwartz, G. Grabarnik","doi":"10.1145/3097983.3098190","DOIUrl":"https://doi.org/10.1145/3097983.3098190","url":null,"abstract":"In large scale and complex IT service environments, a problematic incident is logged as a ticket and contains the ticket summary (system status and problem description). The system administrators log the step-wise resolution description when such tickets are resolved. The repeating service events are most likely resolved by inferring similar historical tickets. With the availability of reasonably large ticket datasets, we can have an automated system to recommend the best matching resolution for a given ticket summary. In this paper, we first identify the challenges in real-world ticket analysis and develop an integrated framework to efficiently handle those challenges. The framework first quantifies the quality of ticket resolutions using a regression model built on carefully designed features. The tickets, along with their quality scores obtained from the resolution quality quantification, are then used to train a deep neural network ranking model that outputs the matching scores of ticket summary and resolution pairs. This ranking model allows us to leverage the resolution quality in historical tickets when recommending resolutions for an incoming incident ticket. In addition, the feature vectors derived from the deep neural ranking model can be effectively used in other ticket analysis tasks, such as ticket classification and clustering. The proposed framework is extensively evaluated with a large real-world dataset.","PeriodicalId":314049,"journal":{"name":"Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115320971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}