Artificial Intelligence and Disaster Management Dr. Jaziar Radianti Teknologidagene 2018 Trondheim, 31 October 2018
Agenda 1. Artificial Intelligence (AI) and Disaster Management 2. Research on disaster management at CIEM 3. AI and Limitations 4. Concluding Remarks
1. AI and Disaster Management
Picture: Pixabay License CC0 Creative Commons
What is Artificial Intelligence? Picture: Pixabay License CC0 Creative Commons
McCarthy 1950s: Artificial Intelligence is the science and engineering of making intelligent machines. AI is the broad concept of machine being able to carry out tasks in a smart way. Picture: Pixabay License CC0 Creative Commons
AI Opportunity for the Environment Source: world Economic forum, 2018. Harnessing Artificial Intelligence for the Earth
Real-time Smart disaster risk Agriculture mapping Drones and Impacts and AI for realtime risk mitigation monitoring of analytics river quality AI-designed Natural Automated intelligent, catashtrophe connected flood center and early warning liveable cities Social media enabled disaster response Extreme Detect weather underground event modelling leaks water and supply prediction systems A community disasterresponse data and analytics platform prediction and forecasting, early warning system, resilience infrastructure, resilience planning Source: World Economic Forum, 2018. Harnessing Artificial Intelligence for the Earth
Earth Dashboard? Picture: Pixabay License CC0 Creative Commons
Common Global First Responders Capability Gaps The ability to : 1. know the location of responders and their proximity to risks and hazards in real time (i.e. accurate geolocation of responders) 2. detect, monitor and analyze passive and active threats and hazards at incident scenes in real time (e.g. Chemical, Biological, Radiation, Explosive, suspicious behavior, fast moving object). 3. identify hazardous agents and contaminants rapidly. 4. incorporate information from multiple and non-traditional sources (crowdsourcing, social media) into incident command operations. (Ihttps://internationalresponderforum.org/)
AI Research in Disaster Risk Management Figures: http://www.bbc.co.uk/science/earth/natural_disasters
AI, Big Data, Social Media and Emergency Response Crowdsourcing + machine learning Satellite Crowdsourcing Sensor and IoT Mobile GPS, Simulation Combination of various data Unmanned Aerial Vehicle
2. Research on disaster management at CIEM
Top priority research centre at University of Agder, established in 2011 Interdisciplinary/ multidisciplinary Collaboration between Faculty of Social Sciences and Faculty of Engineering and Science
Integrated system for real-time TRACKing and collective intelligence in civilian humanitarian missions (12 partners, 8 countries CIEM contributions on AI part : The threat detection module: detecting threat from social media feed, messages sent by personnel on the ground and news reports. The decision support module: choosing one of the alternative actions/mitigation plans based on the predicted threat. (Named-entity recognition-ner and neural network)
Fire Detection and Predition Smart Glasses + Deep Learning Resilience Simulations Facial Expression Data Visualization
AI and Social Media, Situational Awareness, Cybersecurity CIEMlab H2020-MSCA-RISE-2018 (Marie Skłodowska- Curie Research and Innovation Staff Exchange) 2019-2022
Summary CIEM Research Areas Developing community resilience Climate change, migration and disaster vulnerability Supporting the next generation operations centre Multi-level situational awareness Decision support for humanitarian logistics Cybersecurity and critical infrastructures
4. AI and limitations Source: TRY
Sources of Limitations Discriminating algorithms/bias (racial, gender) Low transparency Malevolent use of AI such as autonomous weapons Source: Angwin, J, et. al., 2016
5. Concluding Remarks
Conclusions Picture: Wikipedia To find a way to stay relevant in the face of AI as we realize that AI improves our capabilities in different areas, including decision making Opportunities for AI and disaster management AI decisions are only as good as the data that humans feed them (to understand AI s limitations) Encourage research on: AI algorithm transparency, Explainable AI, AI risk analysis in various application landscape, Ethics and AI ethics algorithm.
Perhaps we should all stop for a moment and focus not only on making our AI better and more successful but also on the benefit of humanity Picture: CC0 Creative Commons (Stephen Hawking at Web Summit, Lisbon, 2017)
Thank you! Jaziar Radianti (jaziar.radianti@uia.no)
Reference Datta, A., Tschantz, M. C., & Datta, A. (2015). Automated experiments on ad privacy settings. Proceedings on Privacy Enhancing Technologies, 2015(1), 92-112. Bolukbasi, T., Chang, K. W., Zou, J. Y., Saligrama, V., & Kalai, A. T. (2016). Man is to computer programmer as woman is to homemaker? debiasing word embeddings. In Advances in Neural Information Processing Systems (pp. 4349-4357). Fefegha, Alex. Racial Bias and Gender Bias Examples in AI systems, the Comuzi journal 2018. Hackett, Robin, 2018. CENTAUR: Upsizing sewer storage. https://wwtonline.co.uk/features/centaur-upsizing-sewer-storage Hornyak, T. (2017), Mapping dengue fever hazard with machine learning, Eos, 98, https://doi.org/10.1029/2017eo076019. Published on 14 June 2017. IFAFRI, Common capability gaps. International Forum to Advance First Responder Innovation, https://www.internationalresponderforum.org/ Angwin, J., Larson, J., Mattu, S., & Kirchner, L. (2016). Machine bias: There s software used across the country to predict future criminals. And it s biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing Lagerquist,R, Flannigan, MD., Wang,X., Marshal, G.A. Automated prediction of extreme fire weather from synoptic patterns in northern Alberta, CanadaCanadian Journal of Forest Research, 2017, 47:1175-1183,https://doi.org/10.1139/cjfr-2017-0063 Li, Karpathu and Johnson, 2016. Spatial Localization and Detection. http://cs231n.stanford.edu/slides/2016/winter1516_lecture8.pdf Michael Mills, 2016. Artificial Intelligence in Law: The State of Play 2016, http://legalexecutiveinstitute.com/artificial-intelligence-in-law-the-state-ofplay-2016-part-1/ New Scientist, 2018. Discriminating algorithms: 5 times AI showed prejudice. RMIT, 2017. Australia: Bushfire ready, thanks to artificial intelligence. PreventionWeb https://www.preventionweb.net/news/view/53102 Roxana L. Ciurean, Dagmar Schro ter and Thomas Glade (April 17th 2013). Conceptual Frameworks of Vulnerability Assessments for Natural Disasters Reduction, Approaches to Disaster Management John Tiefenbacher, IntechOpen, DOI: 10.5772/55538. Sharkey, Noel. The impact of gender and race bias in AI, Humanitarian Law and Policy, August 28, 2018 World Economic Forum, http://www3.weforum.org/docs/harnessing_artificial_intelligence_for_the_earth_report_2018.pdf Zismos, www.zismos.com