Hi [[ session.user.profile.firstName ]]

MinneAnalytics Mondays: Finding Purposefully Hidden Sites with GPUs and ML

Finding purposely-hidden nuclear sites is hard. But new tools and datasets allow analysts to interactively explore huge geotemporal datasets. OmniSci has recently partnered with the Center for Nonproliferation Studies (CNS) and Planet to demonstrate how daily satellite imagery, machine learning for feature extraction, and interactive analytics can help make the world safer. CNS continually assesses potential nuclear missile production sites. It has found that in North Korea these are often hidden at the ends of new mountain roads. How can we turn this insight into actionable data?

OmniSci’s GPU database technology lets us combine several factors into a suitability model considering roads and their relationships to terrain. We leveraged an amazing new machine learning product from Planet - a monthly road change dataset at 5 meter resolution. We combined this with absolute elevation, percent slope and topographic position. Since there are less than 20 known sites, we elected to use a “human in the loop” process to empower analysts to assess the parameters of known sites semi-manually, and then to search for similar sites across the full country. This allowed us to discover hundreds of potential new sites, which CNS plans to further explore and then monitor.
Recorded Jul 16 2020 33 mins
Your place is confirmed,
we'll send you email reminders
Presented by
Mike Flaxman, Spatial Data Science Lead, OmniSci & Adam Edelam, Federal Solutions, OmniSci
Presentation preview: MinneAnalytics Mondays: Finding Purposefully Hidden Sites with GPUs and ML

Network with like-minded attendees

  • [[ session.user.profile.displayName ]]
    Add a photo
    • [[ session.user.profile.displayName ]]
    • [[ session.user.profile.jobTitle ]]
    • [[ session.user.profile.companyName ]]
    • [[ userProfileTemplateHelper.getLocation(session.user.profile) ]]
  • [[ card.displayName ]]
    • [[ card.displayName ]]
    • [[ card.jobTitle ]]
    • [[ card.companyName ]]
    • [[ userProfileTemplateHelper.getLocation(card) ]]
  • Channel
  • Channel profile
  • Accelerated Analytics Alehouse (1.2021) Recorded: Jan 14 2021 43 mins
    Erilk Schultz, Sales Engineer, OmniSci
    Let’s leave 2020 behind and celebrate the new year! Join the OmniSci team for a virtual happy hour at the Accelerated Analytics Alehouse, opening January 13, 2021.

    Mix and mingle with the OmniSci community, learn how to make a Converge Cadillac Margarita, cheers to the new year, and hear about what’s new at OmniSci. Erik Schultz, Senior Technical Sales Engineer and expert bartender, will teach you to make a world class Converge Cadillac Margarita. While you enjoy your new favorite cocktail, or a beverage of your choice, he will highlight some new demos and OmniSci updates, answer any questions you have about the platform, and open up the virtual alehouse for networking and discussion.

    Register today and you will be entered to win an OmniSci branded cocktail set!

    Ingredients (optional):
    - Ice cubes
    - 1 lime
    - Maldon sea salt or other flaky sea salt, for garnish
    - 2 1/2 ounces good quality Anejo tequila or any other style of 100% agave tequila that you like
    - 1/2 ounce Grand Marnier Cointreau or other orange liqueur
    - 3/4 ounce freshly squeezed lime juice
    - 1/2 ounce agave syrup or simple syrup
  • Costumer Story: TELUS + OmniSci with Chris Havlin Recorded: Jan 4 2021 24 mins
    Chris Havlin, Data Scientist, TELUS
    Hear Chris Havlin, Data Scientist at TELUS, discuss how TELUS uses OmniSci to overcome their biggest data challenges and become insight rich.

    To learn more about OmniSci, visit http://www.omnisci.com.
  • OIS 2020: Keynote Speech by Ilham A. Habibie, Chief Executive WanTikNas Recorded: Nov 30 2020 37 mins
    Ilham A. Habibie, Chief Executive WanTikNas, Deputy Chairman Telematics
    OmniSci Indonesia Summit 2020
  • OIS 2020: Keynote Speech Setia Pramana, Head of Sub-directorate of Statistical Recorded: Nov 23 2020 31 mins
    Setia Pramana, Head of Sub-directorate of Statistical Model Development BPS
    OmniSci Indonesia Summit
  • OIS 2020: Discussion with Prof Eko K Budiardjo, Chairman of IPKIN Recorded: Nov 23 2020 50 mins
    Prof Eko K Budiardjo, Chairman of IPKIN
    OminiSci Indonesia Summit 2020
  • OIS 2020: Keynote Speech by Dr Ettikan K Karuppiah, Director/Technologist NVIDIA Recorded: Nov 23 2020 28 mins
    Dr Ettikan K Karuppiah, Director/Technologist NVIDIA
    OmniSci Indonesia Summit 2020
  • OIS 2020: Keynote Speech by Prof. Ir. Nizam. M.Sc. Ph.D, Director General Higher Recorded: Nov 18 2020 26 mins
    Prof. Ir. Nizam. M.Sc. Ph.D, Director General Higher
    OmniSci Indonesia Summit 2020
  • Building Performant GEOINT Dashboards with OmniSci Immerse & Jupyter Notebooks Recorded: Oct 15 2020 60 mins
    Mike Flaxman, Spatial Data Science Lead, OmniSci & Adam Edelam, Federal Solutions, OmniSci
    Today’s geospatial professionals are increasingly called upon to deliver content in the form of interactive web apps. This can be challenging with traditional architectures, both because of the inflexibility of conventional template-based approaches, and due to major performance issues when delivering large geotemporal data across the web. OmniSci is a GPU database with built-in high performance geographic rendering (capable of handling literally billions of features). OmniSci’s Immerse web application allows end users to build or customize a wide range of dashboards themselves. This workshop will present some typical GEOINT applications delivered as interactive dashboards, and will drill down into the construction of one of them in detail. Attendees will learn all three levels of a modern web stack. We will start by reviewing the capabilities of OmniSci Immerse. Next, we will dig into the capabilities of OmniSQL, which uses postGIS-like syntax to expose powerful geoprocessing tools. Lastly, we will walk through Jupyter Notebook integration, and show how Python and the pymapd library can be used to construct powerful workflows.

    Training Goals:

    1. What is GPU analytics and why should I care? How can I get it?

    2. How to create powerful geotemporal dashboards in Immerse

    3. Using omniSQL to handle geotemporal data at scale

    4. Leverage Jupyter notebook integration to optimize data pipelines supporting interactive use cases
  • Leveraging Big Data for Digital Transformation Using GPU-accelerated Analytics Recorded: Oct 15 2020 43 mins
    Jon Stresing, Account Manager DoD, NVIDA & David Goodwin, Director DoD, OmniSci
    The presence of extraordinary amounts of data, to train on, to learn from, and to explore, represents a golden age of computing. But beneath this incredible opportunity lies a massive challenge. Traditional CPU compute cannot keep pace with the growth in data, and as a result, even the most sophisticated organizations are unable to unlock its value. There is a new approach to computing and analytics that solves this yawning gap, involving the application of GPU compute from NVIDIA and analytical software from OmniSci Join OmniSci and NVIDIA to learn about GPUs, GPUs AI/ML, and how you can use NVIDIA and OmniSci for big data analytics and data science initiatives.
  • How Charter Uses GPU-accelerated Analytics to Give Them a Competitive Speed Recorded: Oct 15 2020 28 mins
    Herfini Haryono, VP Vertical Industry, OmniSci & Jared Ritter, Sr. Director Analytics & Automation, Charter Communications
    As one of the earliest adopters and collectors of big data, Telcos are now leading the world in addressing the challenges that come from having lots of data, but oftentimes too few insights. Some of those challenges include:

    - Multiple Sources of Data
    - Lengthy ETL Processes
    - Limited Data Science Teams
    - Growing Data Science Teams
    - Multiple Service Offerings

    In this talk, Charter Communications and OmniSci will show how they are addressing these challenges for the telecom industry. We’ll demonstrate using GPUs and OmniSci’s accelerated analytics platform for fast queries and joins of multiple sources of data, interactive visualizations of spatiotemporal data, and integration with the latest AI/ML tools for running complete data science pipelines without the need for a large team of data scientists.
  • Demo with an OmniSci Expert: 5G Network Optimization Recorded: Oct 15 2020 13 mins
    Dr. Mike Flaxman, Spatial Data Science Lead, OmniSci
    5G Is on its way to consumer markets in the coming years. 5G network infrastructure is expected to completely revolutionize network connectivity. With proper 5G network planning and optimization, telecommunications companies will be able to deliver better customer experiences, solve complex problems, and move their business forward. 5G and big data go hand-in-hand. Experts predict 5G data usage could increase by 10-14 times current figures. This astronomical influx of 5G data creates an incredible opportunity for telco companies to explore real-time 5G insights and leverage 5G data analytics for 5G network optimization that gives your customers fast and consistent network connectivity at all times.
  • Predicting Census Undercount Areas with Massive GPS Movement Pattern Data Recorded: Oct 14 2020 27 mins
    Abhishek Damera, Data Scientist, Product Management & Dr. Mike Flaxman, Spatial Data Science Practice Lead, OmniSci
    Undercounting within the US National Census has always been a significant issue for specific populations, such as those with complex family structures. With the rise of Covid-19 and a shortened counting period, this year’s census is probably the most challenging to date. How can data science and new datasets help?

    We found that modern machine learning techniques can provide the basis for much-improved census undercount prediction. We also found that the addition of GPS data and point of interest data increased model accuracy and provided further insight into behavioral factors beyond demographics which significantly affect undercount.
  • Connecting the Holoviz Visualization Ecosystem to OmniSci Recorded: Oct 14 2020 30 mins
    Tony Fast, Developer Advocate & Kim Pevey, Software Engineer, Quansight
    Holoviz is a framework for visualization and application development that encourages annotating data to generate rich interactive visualizations and dashboards. Holoviz provides interfaces to multiple plotting backends in python including Bokeh and Matplotlib. This talk, will demonstrate brand new functionality showing how Omnisci, through the ibis framework, can now plot data using hvplot or holoviews. We’ll demonstrate these novel integrations with Omnisci in a series of computational notebooks that visualize different projections of Omnisci data, and deploy them as applications.
  • Modin powered by OmniSci Recorded: Oct 14 2020 25 mins
    Areg Melik-Adamyan, Engineering Manager, Intel
    Modin powered by OmniSci
  • Dynademographics: Mapping Human Population Patterns at Scale Recorded: Oct 14 2020 32 mins
    Abhishek Damera, Data Scientist & Dr. Mike Flaxman, Spatial Data Science Practice Lead, OmniSci
    Human mobility patterns have been studied for millennia and are of obvious interest to transportation infrastructure and retail planners among others. It is only in the last few years that consumer GPS and cell phones have allowed observations of large samples over wide areas and long periods of time. Yet the mechanisms used by these services vary widely and should be expect by default to contain demographic sampling biases. We have conflated these GPS-derived spatial event stream data sources with traditional census block group data in order to quantify and calibrate them. We have used these adjustments to produce the first known map of US populations at an hourly basis before and during the COVID-19 epidemic. In order to protect individual privacy, we have deployed an explicit aggregation system using variable-sized hexagonal bins. Nonetheless, our maps represent much higher resolution data than conventionally available, ranging from 50m in dense urban areas to 1km in non density rural areas. Because sampling along roads is much higher than elsewhere, we are able to provide traffic estimates at high spatial precision nationally. In addition to overall population density, we have used overnight device dwell times in residential areas to infer the hourly demographic characteristics of areas based on device trajectories.
  • Covid Risk Assessment from Network Logs Recorded: Oct 14 2020 20 mins
    Mike Hinchey, Solutions Architect, OmniSci & Dr. Mike Flaxman, Spatial Data Science Practice Lead, OmniSci
    As employees go back to work in offices in some countries, employers can use Network and Wifi connection logs to assess risk of disease transmission for high-traffic areas and proximity to individuals that are found to be infected.
  • Data Science with OmniSci - Year One Recorded: Oct 14 2020 36 mins
    Venkat Krishnamurthy, VP Product, OmniSci
    In this overview talk, we'll showcase how Data Science workflows are a key part of the OmniSci platform, since the launch last year. We will showcase new integrations of OmniSci within the open PyData ecosystem, new foundational capabilities within the open source OmniSci engine useful for data science, and discuss joint work with our collaborators at Intel and Quansight, as well as some exciting future plans
Massively Accelerated Analytics and Data Science
Interactively query, visualize, and power location intelligence workflows over billions of records.

Embed in website or blog

Successfully added emails: 0
Remove all
  • Title: MinneAnalytics Mondays: Finding Purposefully Hidden Sites with GPUs and ML
  • Live at: Jul 16 2020 8:00 pm
  • Presented by: Mike Flaxman, Spatial Data Science Lead, OmniSci & Adam Edelam, Federal Solutions, OmniSci
  • From:
Your email has been sent.
or close