August 2020 Summaries

Thinking Machines, a Data Science consulting firm specializing in geospatial analytics, has partnered with CARTO to provide location-based solutions to customers in the APAC region. The partnership aims to enable businesses to make better decisions using spatial data analytics tools such as Google BigQuery and QGIS. CARTO's platform will be used by Thinking Machines' team of Data Scientists and Consultants to create spatial models and applications that address challenges in site selection, territory planning, and network deployment. The partnership reflects both companies' commitment to harnessing the power of geospatial insights for better decision-making in a complex world.

Lisbon, Portugal, was chosen as the first pilot city for a new initiative by Philip Morris International (PMI) to combat cigarette butt littering. The `Our World Is Not an Ashtray' global campaign aims to raise awareness and change behavior around this environmental issue. A data-driven approach is being used to assess the prevalence of cigarette butt litter, identify hotspots, and monitor the impact of anti-littering activities. Machine learning algorithms will be applied to crowdsourced data from the Litterati app to gather insights on litter patterns, which are inherently spatial. The initiative aims to use technology to mitigate future risk by gathering behavioral insight and mapping problem areas.

The coffee supply chain's complexity and lack of transparency raise concerns about the social, environmental, and economic conditions of smallholder farmers. Tracing a coffee back to its source is crucial for understanding quality and taste profiles, but it's not enough; responsible buyers must also verify sustainability issues and incentivize positive change. Technology can help gather information on these issues, and the industry needs to be held accountable for data reliability and transparency. Ultimately, traceability is just one tool in a broader effort to make coffee more sustainable and equitable.

ATTOM in CARTO: School District Boundary Maps` is a collaboration between `CARTO`, a geospatial data platform, and `ATTOM Data Solutions`, a leading provider of boundary-based spatial data. The partnership aims to provide accurate and up-to-date school district boundary maps for the US and Canada, enriching users' data analysis with ATTOM's comprehensive datasets. These boundaries cover 13,462 districts and 67,274 schools, offering benefits such as precision, accuracy, and ease of access for decision-makers and researchers. The integration of ATTOM's data into CARTO's Data Observatory enables users to make informed decisions using data-driven insights on school utilization and planning objectives.

The Fintech Spatial Data Science Masterclass with J.P.Morgan covered three main areas: spatial data science, spatial analysis and data science workflows, and hand-on spatial analysis using Python. Spatial data science treats location distance and spatial interaction as core aspects of the data, focusing on the importance of "where" in traditional data science. It includes various techniques such as spatial clustering, regionalisation, and spatial modelling to leverage location in prediction. The workshop also covered site selection and logistics spatial optimisation, where spatial data science was used to predict potential revenues for new locations and design an optimal supply chain network with reduced distance travelled and average utilisation. CARTO's framework was used throughout the hands-on section of the workshop to explore, enrich, analyse, and share data. The masterclass aimed to provide practical insights into spatial data science and its applications in various industries.

Raster vs Vector Maps: What's the Difference & Which are Best?` Raster data is made up as a matrix of pixels, often square and regularly spaced, whereas vector data stores basic geometries such as points, lines, and polygons. Raster data is suitable for continuous spatial phenomena like elevation or soil type, while vector data is ideal for discrete features like buildings or roads. The choice between raster and vector data depends on the specific use case, with raster often used in remote sensing data and vector being preferred for analysing discrete spatial features. Both formats can be converted to each other, and there are new data types that combine characteristics of both. Spatial indexes, a global grid that renders individual features like polygons, offer fast analysis speeds while still supporting vector-based analysis.

The CARTO BigQuery Tiler is now available for Google Cloud Platform, providing a solution for visualizing large location datasets without moving data outside BigQuery. This tool allows users to create stunning maps in minutes using simple SQL queries and produces vector tilesets directly from SQL. The tilesets are stored in BigQuery and can be used in various tools such as CARTO, Mapbox GL, and QGIS, with the option to make them public or private. With this feature, users working with geospatial Big Data of billions of points can create detailed responsive maps, and it is expected to bring new possibilities for those needing to visualize large datasets.

The article explores where expats invest in property on the Spanish coast, using geospatial analysis to identify trends. The most visited regions by nationality include the British in Muxia and Rubite, Germans mainly on Mallorca, Americans in Barcelona and Vizcaya, French in Girona and Valencia, Italians in the Balearic Islands, Dutch in Alicante, Swiss in La Coruña, and Swedes in Nerja. The regions with the most expensive property prices are Formentera, Cadaqués, and Sant Joan de Labritja, while Xove is one of the cheapest areas, favored by the British. Other nationalities such as Russians, Indians, Hongkongers, Portuguese, Uruguayans, and Singaporeans also show interest in specific regions. The study uses IP address data to determine nationality, allowing for a more detailed analysis of expat preferences.

CARTO, a leading Location Intelligence platform, has partnered with Google Cloud to offer its platform as a solution on the Google Cloud Platform Marketplace. This partnership enables users to deploy CARTO's full stack in a virtual machine (VM) in just a few clicks, providing seamless integration with BigQuery and other cloud computing services offered by Google Cloud. The partnership aims to make it easier for clients to deploy next-generation spatial data infrastructure, and both companies are excited about the potential of this collaboration to enable more companies to use spatial analysis in their business in a more seamless fashion. CARTO's platform can seamlessly import data from BigQuery, allowing data scientists to start crunching geospatial data quickly and efficiently. The partnership also provides users with flexible deployment options, including Bring Your Own License (BYOL) and Pay as you go (PAYG), and leverages Google Cloud's scalable virtual instances to provide full value for every new vCPU used.