2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)最新文献

{"title":"A progressive k-d tree for approximate k-nearest neighbors","authors":"Jaemin Jo, Jinwook Seo, Jean-Daniel Fekete","doi":"10.1109/DSIA.2017.8339084","DOIUrl":"https://doi.org/10.1109/DSIA.2017.8339084","url":null,"abstract":"We present a progressive algorithm for approximate k-nearest neighbor search. Although the use of k-nearest neighbor libraries (KNN) is common in many data analysis methods, most KNN algorithms can only be run when the whole dataset has been indexed, i.e., they are not online. Even the few online implementations are not progressive in the sense that the time to index incoming data is not bounded and can exceed the latency required by progressive systems. This latency significantly restricts the interactivity of visualization systems especially when dealing with large-scale data. We improve traditional k-d trees for progressive approximate k-nearest neighbor search, enabling fast KNN queries while continuously indexing new batches of data when necessary. Following the progressive computation paradigm, our progressive k-d tree is bounded in time, allowing analysts to access ongoing results within an interactive latency. We also present performance benchmarks to compare online and progressive k-d trees.","PeriodicalId":308968,"journal":{"name":"2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130781639","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 client-based visual analytics framework for large spatiotemporal data under architectural constraints","authors":"Guizhen Wang, A. Malik, C. Surakitbanharn, J. Q. Neto, S. Afzal, Siqiao Chen, David Wiszowaty, D. Ebert","doi":"10.1109/DSIA.2017.8339088","DOIUrl":"https://doi.org/10.1109/DSIA.2017.8339088","url":null,"abstract":"A primary aim of visual analytics is to provide end-users interactive and scalable environments to facilitate their decision making tasks. Researchers have often utilized several server-client solutions to support interactive data exploration (e.g., building the data cube, parallelizing data processing). However, these solutions can suffer from scalability issues especially in the absence of adequate computation functionality provided by servers. Organizational policies can also prohibit the transfer of data to external data servers because of security or budgetary concerns; thereby, severely limiting the capability of the visual analytic systems. Therefore, in this paper, we propose an interactive client-based visual analytics framework for large-scale spatiotemporal data. The proposed framework follows a sampling based incremental visual analysis approach to sustain the real-time responsiveness, meanwhile, with affordable computation resources in a client machine. General sampling methods [34] preprocess the entire dataset to build data indexing, which can bring the client unaffordable computation overhead. Instead, our framework proposes a novel data management model, using the spatiotemporal clustering pattern to predictively organize and sample data based on historical data acquisition activities. We demonstrate the capabilities and usefulness of our framework by applying it on crime data and Twitter data. We also conduct several experimental evaluations to determine the efficacy of our framework.","PeriodicalId":308968,"journal":{"name":"2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115851533","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":"Coupling visualization, simulation, and deep learning for ensemble steering of complex energy models","authors":"B. Bush, Nicholas Brunhart-Lupo, Bruce Bugbee, V. Krishnan, Kristi Potter, Kenny Gruchalla","doi":"10.1109/DSIA.2017.8339087","DOIUrl":"https://doi.org/10.1109/DSIA.2017.8339087","url":null,"abstract":"We describe a new framework that allows users to explore and steer ensembles of energy systems simulations by coupling multiple energy models and interactive visualization through a dataflow API. Through the visual interface, users can interactively explore complex parameter spaces populated by hundreds, or thousands, of simulation runs and interactively spawn new simulations to “fill in” regions of interest in the parameter space. The computational and visualization capabilities reside within a general-purpose dataflow architecture for connecting producers of multidimensional timeseries data, such as energy simulations, with consumers of that data, whether they be visualizations, statistical analyses, or datastores. Fast computation and agile dataflow can enhance the engagement with energy simulations, allowing users to populate the parameter space in real time. However, many energy simulations are far too slow to provide an interactive response. To support interactive feedback, we are creating reduced-form simulations developed through machine learning techniques, which provide statistically sound estimates of the results of the full simulations at a fraction of the computational cost. These reduced-form simulations have response times on the order of seconds, suitable for real-time human-in-the-loop design and analysis. The approximation methods apply to a wide range of computational models, including supply-chain models, electric power grid simulations, and building models. Such reduced-form representations do not replace or re-implement existing simulations, but instead supplement them by enabling rapid scenario design and exploration for large ensembles of simulations. The improved understanding, facilitated by the reduced-form models, dataflow API, and visualization tools, allows researchers to better allocate computational resources to capture informative relationships within the system as well as provide a low-cost method for validating and quality-checking large-scale modeling efforts.","PeriodicalId":308968,"journal":{"name":"2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128557460","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":"Position statement: The case for a visualization performance benchmark","authors":"L. Battle, Remco Chang, Jeffrey Heer, M. Stonebraker","doi":"10.1109/DSIA.2017.8339089","DOIUrl":"https://doi.org/10.1109/DSIA.2017.8339089","url":null,"abstract":"Visualizations are an invaluable tool in the data analysis process, as they enable scientists to explore and interpret billions of datapoints quickly, and with just a few rendered images. However, many visualization systems are unable to keep up with the unprecedented accumulation of data through remote sensors, field sensors, medical and personal devices, social networks, and more. This is due to certain assumptions that many of these tools rely on, such as the assumption that these systems can store entire datasets directly in main memory. With so many datasets massive datasets available, ranging from the NASA MODIS satellite imagery dataset[3] to the Internet Movie Database [4] to Twitter streams [1], this assumption no longer matches reality.","PeriodicalId":308968,"journal":{"name":"2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)","volume":"144 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132184595","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":"High-dimensional scientific data exploration via cinema","authors":"J. Woodring, J. Ahrens, J. Patchett, C. Tauxe, D. Rogers","doi":"10.1109/DSIA.2017.8339086","DOIUrl":"https://doi.org/10.1109/DSIA.2017.8339086","url":null,"abstract":"Large-scale scientific simulations and experiments generate enormous volumes of data. Data analytics may become a bottleneck to scientific discovery without scalable tools for interactive exploration. Cinema was developed as a way to overcome hurdles by providing an exploratory, image database approach for analyzing large scientific data sets. In the following, we present several new methods for Cinema: 1) a structured data model that lends itself to querying and database support, 2) support for arbitrary data products beyond images, and 3) parameter exploration through high-dimensional visualization. These changes enrich the types of exporatory visualizations and discoveries that are naturally supported by Cinema-style analyses, further enabling data-driven science.","PeriodicalId":308968,"journal":{"name":"2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129528270","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":"Xplorer: A system for visual analysis of sensor-based motor activity predictions","authors":"M. Cavallo, Çağatay Demiralp","doi":"10.1109/DSIA.2017.8339085","DOIUrl":"https://doi.org/10.1109/DSIA.2017.8339085","url":null,"abstract":"Due to the large diffusion of wearable devices, the task of detecting motor activities from sensor data is becoming increasingly common in a wide range of applications. During the development of predictive models for activity recognition, data scientists generally rely on performance metrics (such as accuracy score) for evaluating and comparing the performance of classification algorithms. While these numerical estimates represent a straightforward way to summarize the effectiveness of a model, they convey little insights on the causes of misclassified events, not offering enough clues for data scientists to improve their algorithms. In this paper we present BlueSky Xplorer, an interactive visualization system to analyze, debug and compare the output of multiple predictive models at different levels of granularity. We combine classification results on multi-sensor data with the context of usage of each sensor and with ground truth information (such as textual labels and videos), representing them as temporally-aligned linear tracks. We then define an algebraic language over these tracks that enables users to quickly identify classification errors and to visually reason on the performance of classifiers. We demonstrate the usefulness of our tool by applying it to a real-world example, involving the development of models for assessing the symptoms of Parkinsons disease. In particular, we show how Xplorer was used to improve the performance of classification models and to discover problems in data temporal alignment.","PeriodicalId":308968,"journal":{"name":"2017 IEEE Workshop on Data Systems for Interactive Analysis (DSIA)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115033295","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}