Advancing Machine Learning and AI with Geography and GIS Robert Kircher rkircher@esri.com
Welcome & Thanks
GIS is expected to do more, faster. see where find where predict where locate, connect WHERE route where map the unmapped machines can be trained to do more for us
Topics Today Emerging Geo AI Business Cases Geo-enabling Machine Learning Landscape and Factory Notable Geo AI challenges and limitations
Emerging Geo AI Business Cases Accidents, Anomaly Prediction Predictive Asset Allocation Predictive Routing, ETA, Traffic Predictive GeoMarketing Advanced Feature Extraction Autos Police Landcover Landuse Classification Crops Object Detection Map the Unmapped Protection Risk Management Pollution Exposure Augmented Reality (AR)
Industry Advancing Fast (with Geo) Focused Computing GPUs, Devices SaaS, Cloud-based Spatial Statistics Organizations Platforms, On-premises
"Gives computers the ability to learn without being explicitly programmed"
AI > ML > DL Artificial Intelligence Machine Learning Deep Learning Supervised Learning Deep Supervised Learning Reasoning Knowledge Representation Perception 1. Training features Labels NLP Robotics Machine Learning 2. Predicting Unsupervised Learning Reinforcement Learning Dog
ML Algorithms
Data + ML (algo s) (statistics) Models Predictions (inferences) AI AR decisions understanding insights classify
ML changes things in exciting ways for GIS evidence based models faster automated data driven accurate tunable platforms pervasive
Machine Learning Capabilities Today Emerging ArcGIS
Geo ML: Spatial Analytics, Geostatistics, Spatial Statistics, Surfaces
Geo-enabled ML or Geo-centric ML
ML Landscape, Factory and Pattern Data Wrangling/Preparing Model Training/Tuning Share/Deploy Model(s) Use Model(s) Classification Predict, Infer Data Discovery/Create imagery vector business Exploratory Data Analysis (EDA) Features/Labels training Clustering Regression Deep Learning (neural networks) 5 4 3 2 1 0 1 2 3 4 regression Augmented Reality (AR) Computer Vision Visualize, Map It Data Gathering/ Wrangling Recommend Data Preparation ML algorithms neural networks Find/Detect Features data sources Model Validation apps, maps, services, devices
Geo ML Landscape, Factory and Pattern geo-enabling and advancing Data Wrangling/Preparing Model Training/Tuning Share/Deploy Model(s) Use Model(s) Classification Predict, Infer Vast Data Collections, Discovery/Create Expertise imagery Geo Data Sources Structured/Unstructured vector GeoAnalytics business Server Exploratory Maps, Visualization Data Analysis Jupyter/Python/R (EDA) Data Geo-centric Gathering/ Wrangling Datasets Geo-referenced Datasets Localized Datasets Data Geo Preparation CLT Geocoding training Spatial Statistics Spatial Analysis Clustering GP Tools Geo-certify Regression Deep Learning (neural networks) ML algorithms 5 4 3 2 1 0 1 2 3 4 regression neural networks Augmented Reality (AR) Computer Vision Integrated GIS Platform Mapping Platform Visualize, Map It Recommend Find/Detect Features data sources Model Validation apps, maps, services, devices
Demo: CAFO Detection Using TensorFlow CNNs to Detect CAFO sites from Satellite Imagery + Consuming the model from ArcGIS Pro Demo Code: https://github.com/qberto/ml_objectdetection_cafo
Please notice process details Geo Data Preparation Model Training Duration Integrating Model Across GIS Platform Augmented Reality (AR) Recent Demo at Esri (ignore some references) imagine your business problem, your data, your model
Some Challenges & Limitations
It s complex. Let s simplify it technology, automation, platforms, computing, etc.
Look past hype it s distracting prevent another Machine Learning Winter
Pred = W 1 A 1 + W 2 A 2 + + W loc (Location) not fully integrated yet
Feature 1 Behavior 1 Traditional Model (Geo-less) Location
Limited expertise among Geo Community in training and tuning models math, statistics, spatial statistics Introduce Geo Data Science into industry
Limited exposure to Geo in academia Data Science, Computer Science
Proving Model Trustworthiness, Authoritativeness
(Geo) Data Bias
Next Steps
Machine Learning needs human analysis to succeed. GIS Community, People, Professionals, Academia.
Advancing Geo AI Build on Emerging Business Cases and Geo-enabled ML Patterns Find your place in ML landscape and factory Together, let s continue to overcome Geo AI challenges and limitations