Type of hierarchical clustering where clusters are built from the bottom up. This algorithm starts building clusters where each object is in its own cluster, then clusters are recursively merged (agglomerated) using a 'linkage strategy' such as minimizing the sum of squared distances within a cluster.

An AI Agent is an intelligent system, typically powered by a Large Language Model (LLM), that can autonomously perform tasks, make decisions, and interact with users.

Alteryx provides a user-friendly interface to perform tasks such as data cleansing, transformation, and predictive analytics without requiring extensive coding skills.

CARTO’s Analytics Toolbox is a set of user-defined functions (UDFs) and Stored Procedures to unlock Spatial Analytics directly on top of your cloud data warehouse platform.

Apache Iceberg is an open-source table format that brings the reliability and scalability of data warehouses to data lakes.

In spatial analytics, AI can be used to uncover trends in location data, automate decision-making, and optimize everything from delivery routes to urban planning.

Provides a way to calculate the probability of a hypothesis based on its prior probability, the probabilities of observing various data given the hypothesis, and the observed data itself.

Bring Your Own LLM (BYOLLM) allows organizations to connect and use their own Large Language Model (LLM), instead of being limited to default or proprietary models.

A specialized database system hosted and managed as a service in a cloud computing environment. It serves as a centralized repository for storing and managing structured, semi-structured, and unstructured data from various sources within an organization. 

Cloud native refers to a modern approach in data storage and architecture, as well as software development and deployment that harnesses the capabilities of cloud computing environments to build, deploy, and manage applications.

Model validation technique for assessing how the results of a statistical model will generalize to new data.

Databricks Data Intelligence Platform simplifies the process of building and deploying big data applications. Founded by the creators of Apache Spark, Delta Lake, and MLflow.

A data lake is a collection of data or files generally stored in a cloud file storage or blob infrastructure. This data may or may not be unprocessed or clean.

A data warehouse is an OLAP (Online Analytical Processing) that can distribute query workloads across compute infrastructure and where data is organized in a structured manner to make it ready to analyze.

It is a clustering method that groups data that are close to each other based on a distance metric and a minimum number of data points.

deck.gl is one of the most popular Open Source map visualization libraries and is the preferred library to use with CARTO.

Esri is a Geographic Information System (GIS) technology. It develops GIS software, provides geospatial solutions, and offers consulting services for various industries such as government, natural resources, utilities, and many others.

The EU AI Act is a legal framework for artificial intelligence. It classifies AI systems based on their risk level, with "high-risk" systems facing strict requirements related to data governance, transparency, and auditability.

For spatial data science, Generative AI allows to generate diverse and augmented datasets, improve model training, simulate scenarios, and assist in spatial analysis and modeling by producing realistic spatial information.

Geocoding is the process of assigning longitude and latitude values to street addresses, whilst reverse geocoding is the opposite: identifying the addresses of locations. You can use geocoding to position a marker on a map.

Geographically Weighted Regression (GWR) quantifies the strength of relationships across space between a target and correlation variables.

Geographic coordinates, also known as geographic or unprojected coordinates, use latitude and longitude to specify a location on the Earth’s curved surface. These are stored as a geography field type.

GeoJSON is a type of JSON format designed for encoding geographic data structures. It supports various geometry types such as points, lines, polygons, and multi-geometries.

GeoParquet is a project aimed at extending the Parquet file format to directly support geometry data, eliminating the need for later translation and reducing computational costs.

Geospatial foundation models are AI models trained on vast datasets like satellite imagery to understand the physical and human world, adapting to diverse tasks.

Geospatial Reasoning is the ability of an AI system to understand a complex spatial problem, independently devise a multi-step analytical plan, execute it, and interpret the results to provide an actionable solution.

Google Cloud's BigQuery is the serverless, cost-effective, and multi-cloud data warehouse offered by Google Cloud Platform (GCP).

Geopackage (GPKG) is an open standard vector file format developed by the Open Geospatial Consortium (OGC).

HEAVY.AI is a San Francisco-based company that specializes in providing advanced analytics solutions for businesses and government organizations.

Hotspot analysis enables the identification of clusters of high or low values of data on a map, going beyond visual representation.

HTML is used for creating the structure and content of web pages. It plays a crucial role in being used to embed interactive maps, geospatial visualizations, and geospatial data in web applications.

An isoline map is a way of presenting numerical data cartographically, by connecting points with similar values with lines or polygons, helping readers recognize geographical patterns and relationships, such as travel time catchments.

JavaScript is often employed for creating interactive and visually appealing web maps. It allows developers to manipulate and visualize spatial data using libraries such as React and deck.gl.

Kinetica is a real-time GPU-accelerated analytic database for big data. Kinetica leverages GPUs and modern many-core CPUs to improve performance.

KML stands for Keyhole Markup Language and it is an XML-based file format used to display geographic data.

Spatial interpolation method for predicting missing values, taking into account the distance and spatial correlation of known data points.

A lakehouse is a modern data architecture that combines the flexibility and low-cost storage of a data lake with the transactional capabilities and performance of a data warehouse.

A type of generative AI trained on massive datasets to understand and produce human-like language.

Location Intelligence (LI) is the methodology of deriving insights from location data to answer spatial questions. Location Intelligence goes beyond simple data visualization on maps, to analyzing location data as an integral part of a business or societal problem.

A subset of AI where algorithms learn from data to make predictions or decisions, improving over time without being explicitly programmed.

Mapbox enables users to create and integrate custom maps into their applications and websites. It provides a range of tools and APIs for map design, geocoding, search, navigation, and data visualization.

It refers to computer programs or applications designed to create, display, and manipulate spatial information in the form of maps.

Class of simulation methods used to approximate the posterior distribution by randomly sampling in a probabilistic space.

Model-Agnostic refers to CARTO's philosophy of not depending on a single AI provider. It gives the user the freedom to choose the one that best fits their needs and adheres to their governance requirements.

The Model Context Protocol (MCP) is an open standard that enables AI applications to connect securely and standardize their interactions with external systems, tools, and data sources.

A branch of AI that enables computers to understand, interpret, and generate human language.

Orchestration (in the context of AI) is the process of centrally coordinating and managing all system components (the LLM, tools, data, and user interface) to consistently execute user-requested tasks.

Projected Coordinates, also referred to as geometries, utilize a two-dimensional Cartesian coordinate system to represent locations on a flat surface, such as a map or a plane. These are stored as geometries.

Python is a popular programming language which can be used for spatial analysis by leveraging various libraries like GeoPandas, Shapely, and PySAL, which provide functionalities to manipulate, visualize, and analyze spatial data.

Raster data is represented as a grid of cells or pixels, with each cell containing a value or attribute. It has a grid-based structure and represents continuous values such as elevation, temperature, or satellite imagery.

Retrieval-Augmented Generation (RAG) is a technique that enhances LLM responses. This allows responses to be more accurate, reliable, and based on real-time data.

R is a powerful programming language specifically designed for statistical computing and data analysis.

Snowflake is a single, global platform that powers the Data Cloud. Its architecture separates computing and storage, allowing independent scaling and cost optimization. 

Spatial data, also known as geospatial data or location data, is any data that has a geographic component.

Use the Spatial Data Catalog to browse and filter our collection of public and premium datasets available in the Data Observatory.

Spatial data governance is a framework that provides a clear, structured approach to managing your spatial data, ensuring it is trustworthy, secure, and compliant.

Spatial data science (SDS) is a subset of Data Science that focuses on the unique characteristics of spatial data, moving beyond simply looking at where things happen to understand why they happen there.

Spatial Indexes are global grids - in that sense, they are a lot like raster data. However, they render a lot like vector data; each “cell” in the grid is an individual feature that can be interrogated. They are well suited to storing, analyzing, and rendering big spatial data as their location is encoded as a short string, rather than a large geometry field.

Spatial SQL uses the same elements & structure of normal SQL but allows you to work with geospatial data types such as geometries & geographies.

Structured Query Language (pronounced either as sequel or “ess-queue-ell,” depending on who you talk to) is used for managing and querying relational database management systems (RDBMS), as well as running analysis.

A tileset is a representation of spatial data optimized for map visualizations. Tilesets are designed to facilitate faster loading, optimized bandwidth usage, and seamless navigation in map visualizations. 

Vector data represents geographic features as discrete points, lines, and polygons.It has a geometry-based structure in which each element in vector data represents a discrete geographic object, such as roads, buildings, or administrative boundaries.