We evaluate the use of multivariate timeseries modeling with the multi-task Gaussian process (GP) models using noisy, incomplete, sparse, heterogeneous and unevenlysampled clinical data, including both physiological signals and clinical notes. The learned multi-task GP (MTGP) hyperparameters are then used to assess and forecast patient acuity. Presenter: Tristan Naumann Paper: ghassemi_AAAI2015_multivariate_timeseries_modeling Poster: MIT_AAAI_MTGP_lg

S-Store is the world’s first streaming OLTP engine, which seeks to seemlessly combine online transactional processing with push-based stream processing for real-time applications. Presenter: Nesime Tatbul Papers: s-store-demo sstore_vldb16 sstore-vldb2015 Website: sstore.cs.brown.edu/

Data Warehousing using in a balanced partitioning tree, instead of blocks Presenter: Anil Shanbhag on Github Poster: hp_poster

Swarm: a novel architecture that exploits ordered irregular parallelism, which is abundant but hard to mine with current software and hardware techniques. In this architecture, programs consist of short tasks with programmer-specified timestamps. Swarm executes tasks speculatively and out of order, and efficiently speculates thousands of tasks ahead of the earliest active task to uncover […]

ForeCache: a general-purpose tool for exploratory browsing of large datasets. ForeCache utilizes a clientserver architecture, where the user interacts with a lightweight clientside interface to browse datasets, and the data to be browsed is retrieved from a DBMS running on a back-end server. Presenter: Leilani Battle Poster: forecache_small Paper: MIT-CSAIL-TR-2015-031

BlueDBM: a new system architecture which has flash- based storage with in-store processing capability and a low- latency high-throughput inter-controller network. BlueDBM presents an attractive point in the cost-performance trade-off for Big Data analytics. Paper: ISCA15_Sang-Woo_Jun Presenter: Sang-Woo Jun Poster: poster

Question answering can help us make sense of Big Data by reducing a massive collection of documents into a relevant answer. Knowledge bases constructed automatically from the Web are rich sources of answers, but they suffer from sparsity in the relations that connect entities. For example, we can express that a person studied at a […]

DataHub is a web application, and the center of a collaborative ecosystem for data processing, including ingestion, curation, integration, discovery, query, analytics, visualization, and machine learning. Presenters: Justin Anderson and Albert Carter Paper:datahubcidr datahub Poster: MIT Living Lab Poster – special export Demo: datahub.csail.mit.edu

Decibel: A data management systems that natively support the versioning or branching of datasets to enable concurrent analysis, cleaning, integration, manipulation, or curation of data across teams of individuals. Presenter: David Goehring Paper: Decibel: The Relational Dataset Branching System Poster: Decibel Poster Slides: Decibel Presentation

NVL is a simple and expressive intermediate language for data parallel computations that targets vector units, multicores, GPUs, and other high-performance features of modern machines, allowing developers of data parallel applications to achieve excellent performance without extensive development effort. NVL’s simplicity enables powerful program analysis and aggressive optimization. When different NVL-enabled libraries are composed by […]