AGU Tuesday 12/13/2016

0800-1000

First Speaker: Chris Lynnes (NASA ESDIS)

Heuristics for relevancy ranking of earth dataset search results

• Earth Observing System Data and Information System

• variety in EOSDIS data: 3700 science products.

• https://search.earthdata.nasa.gov/ 

• Variety: instrument evolution, satellites, processing, algorithm, temporal

• Return relevant results 

• Community use hueristics to find most useful

• Applications user, students, climate modellers.

Second Speaker: Mark Reese (NASA)

Earthdata Search: scaling assessing and improving relevancy.

• search, discover, and vizualize earth science data

• sitting on NASAs client

• 32,960 data products

• Facets to compartmentalize search results 

Third speaker: Stephen Richard (SDSC)

Integrating semantic information in metadata descriptions for a geoscience-wide resource inventory

• Community inventory for earthcube resources for geoscience interoperability (CINERGI)

• Using standard metadata formats 

• MongoDB, JSON format

• http://cinergi.sdsc.edu

• provenance recording: w3c prov and neo4j

• spatial enhancer (bounding box)

• keyword enhancer (faceted search)

• organization enhancer (associate with virtual authority identifiers)

• harvest from a number of data sources (600k processed / 1.5 mil total)

Fourth speaker: Anu Devaraju (CSIRO)

Data recommender system for research data discovery

• recommender system, data access portal (dap)

• https://data.csiro.au/

• DAP Server logs -->Google Analytics, SQL database-->HDF

• data.gov.au

Fifth speaker: Mo Wang (Chinese Academy of Sciences)

A hybrid personalized data recommendation approach for geoscience data sharing

• http://www.geodata.cn/

• spatial similarity models for returning search results.

• Showed a lot of equations for computing spatio-temporal metrics.

Sixth speaker: Chao Yang (George Mason)

Advancing geoscience resource discovery with cutting edge cyberinfrastructure technology

• http://cisc.gmu.edu/

• http://www.stcenter.net

• http://www.sciencedirect.com/science/article/pii/S0198971516303106

• sweet 2.0

• geosearch.cloudapp.net

• elastic

• https://github.com/mudrod

Seventh speaker: Tristan Wellman (USGS)

Cross linkage of multiple criteria to enhance selection aplicability and use of biogeographic occurance data.

• http://www.iobis.org/mapper/

• optimize data quality

• data assessment module (dam) - regulate data across an interface (like water over a dam)

Eighth speaker: Sky Bristol (USGS)

Measuring the impact of an api-first mentality with ScienceBase after 4.5 years * API use exceeds portal traffic with 70+ API-driven apps * https://nccwsc.usgs.gov/tools * www.sciencebase.gov * iPython notebook 

1020-1200 Dropped into Data session

geoplatform.gov

CZO Townhall

OpenSource software & data

Computational infrastructure for geodynamics

• workshops, tutorials, webinars, hackathons

• http://aspect.dealii.org

NASA DEVELOP

• software carpentry

Agile Softwarwe management for Successful open source software projects

• Open Geospatial Consortium

• http://www.opengeospatial.org/

• Testbed 13 coming

• Github, TestNG, Maven, CTL, XSLT

DARPA Memex

• Apache Nutch, Sparkler, Tika, Solr

• Lucene Geo-Gazetteer

  • GeoTopicParser, GeoReverse

scalable parallel 'tile streaming' architecture

www.rasdaman.org, www.jacobs-university.de/lsis osgeolive planetserver.eu ISO array SQL candidate standard Multi-dimensional Arrays (SQL/MDA)

OpenSource Architecture

datasources --> normalize and extract --> Data Lake (CDR)

apache nutch --> apache tika, deepdive --> lucene, apache solr, elastic

Data Engeineer at the architecture level, Data scientist at the analysis level, Web developer at the interface.