Program
Keynote
Prof. Lex Comber
Leeds Institute for Data Analytics (LIDA) and the School of Geography
University of Leeds, UK
Title: Geosensor network optimisation to support decisions at multiple scales
Abstract
Geosensor networks are often used to monitor processes at different spatial scales. Existing approaches for configuring geosensor locations (i.e. sample design) do not address two key challenges: 1) they are limited to a single scale of analysis and do not support multiple scales of evaluation, and 2) they assume that the geosensor network, once established at whatever scale, does not change either in terms of location or number of geosensors. While approaches exist in part for 1) and 2) they do not for both combined. This paper describes a novel approach for optimising geosensor locations in support of multi-scale decisions. It uses the local variation in a gradient as a cost surface that approximates the process of interest, based on the assumption that relevant data are increasingly available across different domains. This is used to weight potential solutions to a p-median problem, the approach quantifies the information loss / gain of geosensor designs when evaluated over different scales. An agricultural case study is used to illustrate the trade-offs when geosensors locations are optimised and then evaluated over different scales (field, farm and catchment). In this parsimonious approach, cross-scale evaluations of geosensor spatial configurations are supported by evaluations of the information loss about the siting of new and existing geosensors within spatially nested decision scales. In so doing, the methods described in this paper fill an important gap as they are i) suggest appropriate sample and geosensor network designs to support cross-scale monitoring, ii) inform on how current network or geosensor coverage could be enhanced by filling gaps, and iii) quantify the information trade-offs (information loss) associated with designs when they are evaluated from the perspective of different decision scales.
Biography
Lex Comber (http://goo.gl/CPgDRj) is the Professor of Spatial Data Analytics at Leeds Institute for Data Analytics (LIDA) and the School of Geography at the University of Leeds, UK. He worked previously at the University of Leicester where he held a chair in Geographical Information Science, and has collaborated with a number of researchers in both France and Canada, originally through the REV!GIS project led by Robert Jeansoulin. Lex has a background in Plant and Crop Science and undertook a PhD in Computer Science developing expert systems for land cover monitoring. This brought him into the world of spatial data, spatial analysis, and mapping. Lex’s interests span many different application areas including land cover / land use, demographics, public health, agriculture, bio-energy and accessibility, all of which require multi-disciplinary approaches. His research draws from geocomputation, mathematics, statistics and computer science and he has extended techniques in operations research / location-allocation (what to put where), graph theory (cluster detection in networks), heuristic searches (how to move intelligently through highly dimensional big data), remote sensing (novel approaches for classification), handling divergent data semantics (uncertainty handling, ontologies, text mining) and spatial statistics (quantifying spatial and temporal process heterogeneity). With the wider use of spatial data in other disciplines (actually all data are spatial – they are collected some-where!), Lex’s collaborations are increasingly with researchers in non-geographical domains. Recent examples include mental health, consumer analytics, market segment dynamics, bio-informatics and spatial transcriptomics. Lex has co-authored (with Chris Brunsdon) the first ‘how to book’ for spatial analyses and mapping in R the open source statistical software, An Introduction to R for Spatial Analysis and Mapping, now in its second edition (https://uk.sagepub.com/en-gb/eur/an-introduction-to-r-for-spatial-analysis-and-mapping/book258267). The follow on to this is Geographical Data Science and Spatial Data Analysis: and Introduction in R, (ISBN: 978-1526449351, was published in 2021 (https://us.sagepub.com/en-us/nam/geographical-data-science-and-spatial-data-analysis/book260671). Lex is currently the Chair of the AGILE, the largest European GIS grouping (https://agile-online.org) and the co-Chair for the forthcoming GIScience 2023 conference that will be held at Leeds (https://giscience2023.github.io). Outside of academic work and in no particular order, Lex enjoys his vegetable garden, walking the dog and playing pinball (he is the proud owner of a 1981 Bally Eight Ball Deluxe).
Conference Program
Day1: June 12th 2023
09:00 – 09:30 Opening
09:30 – 10:30 Session 1: Keynote
Alexis Comber and Paul Harris
Geosensor network optimisation to support decisions at multiple scales
10:30 – 11:00 Coffee break
11:00 – 12:00 Session 2: Sensor Networks and Data Steaming
Saeid Doodman, Mir Abolfazl Mostafavi and Raja Sengupta
Towards Integration of Spatial Context in Building Energy Demand Assessment Supported by CityGML Energy Extension
Jinlong Cui and Zhixiang Fang
A Three-stage Framework to Estimate Pedestrian Path by Using Signaling Data and Surveillance Video
12:00 – 14:00 Lunch
14:00 – 15:00 Session 3: Mobility and Human environment interactions
Sanaz Azimi, Mir Abolfazl Mostafavi, Krista Lynn Best and Aurélie Dommes
Investigating the navigational behavior of wheelchair users in urban environments using eye movement data
Maryam Naghdizadegan Jahromi, Najmeh Neysani Samany, Mir-Abolfazl Mostafavi and Meysam Argany
A new approach for accessibility assessment of side-walks for wheelchair users considering the sidewalk traffic
15:00 – 15:30 Coffee break
15:30 – 17h00 Session 4: Panel discussion: Digital Twins for Mobility and Navigation
18h30 – 23h00 Social event and Gala Dinner
Day2: June 13th 2023
09:00 – 10:30 Session 1: AI for Mobility Data Analytics
Sergio Di Martino, Nicola Mazzocca, Franca Rocco di Torrepadula and Luigi Libero Lucio Starace.
Mobility Data Analytics with KNOT: the KNime mObility Toolkit
Laura Dunne, Franca Rocco Di Torrepadula, Sergio Di Martino, Gavin McArdle and Davide Nardone.
Bus Journey Time Prediction with Machine Learning: An Empirical Experience in Two Cities
Ali Afghantoloee, Mir Abolfazl Mostafavi and Bertrand Gélinas.
A Novel GIS-Based Machine Learning Approach for the Classification of Multi-Motorized Transportation Modes
10:30 – 11:00 Coffee break
11:00 – 12:00 Session 2: Volunteered Geographic Information (VGI)
Maryam Lotfian, Jens Ingensand, Adrien Gressin and Christophe Claramunt.
CIMEMountainBot: A Telegram bot to collect mountain images and to communicate information with mountain guides
Milad Moradi, Stéphane Roche and Mir Abolfazl Mostafavi
A New Feature Matching Method for Matching OpenStreetMap Buildings with Those of Reference Dataset
12:00 – 14:00 Lunch
14:00 – 15:30 Session 3 Network Analysis and Geovisualization
Matthieu Viry and Marlène Villanova
Geovisualisation generation from semantic models: a state of the art
Jialiang Gao, Peng Peng, Christophe Claramunt and Feng Lu
A Heterogeneous Information Attentive Network for the Identification of Tourist Attraction Competitors
Lasith Niroshan, James Carswell
Poly-GAN: Regularizing Polygons with Generative Adversarial Networks
15:30 – 16:00 Closing Remarks