Environmental Sensing Data for Assessing the Role of Vegetation in Urban Water and Climate Sustainability
Environmental sensing has expanded rapidly for more than a decade. I will provide an overview of the dimensions of this data revolution within the ecological sciences. I will then describe a specific evaluation of the water-ecosystem service trade-offs for the use of urban vegetation to cool cities. Vegetation interacts strongly with urban water sustainability. Plants...
Placing Lobbyists in Legislative Ideological Space
I propose a new method to place lobbyists into standard common space measures for ideology scores, leveraging responses from former members of the U.S. Congress to a survey containing a battery of ideology attitude measures, along with a flexible Bayesian statistical model. The statistical model incorporates estimation uncertainty into the imputed lobbyist ideology measures and...
Measuring Trade Profiles with Two Billion Observations of Product Trade
The product composition of bilateral trade encapsulates complex relationships about comparative advantage, global production networks, and domestic politics. Yet, despite the availability of product-level trade data, most researchers rely on either the total volume of trade or certain sets of aggregated products. We develop a new dynamic clustering method to effectively summarize this massive amount...
Computational analysis of olfaction
Insects use the sense of smell to identify their host animals and plants. The ability to detect and discriminate thousands of odorants from their hosts uses a very large family of transmembrane odorant receptors and complex neuronal circuitry. The study of olfaction has benefited from computational approaches to identify important principles: protein sequences of receptors...
Time Series Data Mining Using the Matrix Profile: A Unified View of Motif Discovery, Anomaly, Detection, Segmentation, Classification, Clustering and Similarity Joins
Time series data mining is a perennially popular research topic, due to the ubiquity of time series in medical, financial, industrial, and scientific domains. There are about a dozen major time series data mining tasks, including: • Time Series Motif Discovery • Time Series Joins • Time Series Classification (shapelet discovery) • Time Series Density...
Computational Imaging: Finding Structure from Randomness
We face two broad challenges as we design the next generation of intelligent and interconnected devices: On one extreme, these systems will collect an enormous amount of data from a multitude of sources and require low-complexity, versatile algorithms that can make sense of all the data. On the other extreme, certain physical or system constraints...
You Got Data? We Got Tensors!!
Tensors and tensor decompositions have been very popular and effective tools for analyzing multi-aspect data in a wide variety of fields, ranging from Psychology to Chemometrics, and from Signal Processing to Data Mining and Machine Learning. In this talk, first I will motivate the use of tensors as an effective data analytic tool in a...
Big Spatial Data on Hadoop and Beyond
The explosion in the amount of spatial data in the recent years urged researchers to build specialized systems for big spatial data. This talk will have two parts. In the first part, we describe SpatialHadoop, the most comprehensive open source system for big spatial data. We describe how SpatialHadoop managed to achieve simplicity and efficiency...
How big data and computational models are changing the study of child language acquisition
For decades, cognitive psychologists and linguists have studied language development by testing theories of learning and development in highly controlled behavioral experiments. Much has been learned from this approach. However, Big Data and computational models allow us to investigate language development in a radically different way: by collecting large datasets of actual speech to children...
Big data opportunities in computational materials science
The modern electronic structure methods allow for a reasonably accurate computation of quantum mechanical behavior of atoms and electrons in a material with almost any chemical composition, using virtually no input parameters. These methods allow design of new materials even before they are made in the laboratory. The only input parameters for these methods are...