AIOK: AI for Outdoor Knowledge

Application domain(s)

  • Big Data
  • Future Internet
  • Industry 4.0

Area(s) of expertise

  • Artificial Intelligence
  • Predictive Modelling

Description

AIOK is a predictive analytics framework, composed of machine learning and geospatial analysis modules trainable on heterogeneous data (geospatial enhanced data, time series, profiling data) produced in a controlled environment such as outdoor sport.

Visual representation of AIOK architecture: Inputs, AI engine and users (private, manager and Insurtech)

Advantages & Applications

  • AIOK supports a dynamic personalized risk profile for outdoor sports by a smart pluggable and scalable AI engine.
  • AIOK allows to setup a predictive pipeline for risk management and resource optimization.

Specifications

  • AIOK is developed in python using machine learning and deep learning modules at the state of art.
  • Predictors are trained on historical data.
  • External resources can be easily connected into the framework and added to the models.

Status

TRL 5 – technology validated in relevant environment (one case study in outdoor sport on big data)

Contacts

  • Reference person: Cesare Furlanello, Andrea Gobbi

    Research Unit: MPBA; contact [email protected]

    Research Center ICT: ict.fbk.eu