Keynote @ Exploitation of Social Media for Emergency Relief and Preparedness (SMERP)
Co-located with: The Web Conference 2018 (formerly WWW)
Lyon, France. 23 April 2018
Abstract:
Crises are imposing massive costs to economies worldwide. Natural disasters caused record $306 billion in damage to the U.S. in 2017! Real-time gathering of relevant data through ubiquitous presence of mobile technologies and the ability to disseminate them through social media has forever changed how disaster and health crisis monitoring and response are now carried out. Both tradition crisis response organization as well as temporary, informal, self-organized and community-based organizations have come to increasingly rely on social media. Furthermore, ability to collect, repurpose and reuse data from past events is helping with preparedness and planning for future events.
In this talk, I will review our extensive experience on (a) interactions with variety of stakeholders involved in emergency response at city, county, country and international levels, (b) research on real-time social media analysis spanning spatio-temporal-thematic; people-content-network; linguistic-sentiment-emotion-intent analysis dimensions, (c) development and use of crisis response specific tools (location identification, demand-supply match) and the comprehensive Twitris semantic social intelligence system (which is also commercialized as Cognovi Labs), and (d) a variety of real-world evaluations and real-time uses (e.g., supplying data for Google Crisis map during Uttarakhand Floods, rescue during Kashmir Floods, neighborhood image map during Chennai Floods, providing information to FEMA during Oklahoma tornados), spread of disease and epidemiology (e.g., Zika spread), metro-level multi-agency disaster preparedness exercise, etc.
https://www.cse.iitk.ac.in/users/kripa/smerp2018/SMERP-at-Web2018-keynote.pdf
Twitris in Action - a review of its many applications Amit Sheth
Twitris is a System for Collective Social Intelligence. It has been used in a large number of and many types (disaster coordination, banding, epidemiology, public health, election/polical, social movement) of applications - often in real-time. This presentation gives a bird-eye review of some of these applications with links to explore them further.
Public Health Crisis Analytics for Gender ViolenceHemant Purohit
Research-progress talk on the use of data analytics methods for one of the major public health crisis in the world Gender-based Violence and the campaign engagement in the initiatives of Non-profit organizations.
The Internet of Things means not just that computing devices have connectivity to the cloud but that they themselves are connected to each other, and therefore that novel applications can be developed in this rich ecosystem. One area for development is linking quantified self wearable sensors with automotive sensors for applications including Fatigue Detection, Real-time Parking and Assistance, Anger/Stress Reduction, Keyless Authentication, and DIY Diagnostics.
Helping Crisis Responders Find the Informative Needle in the Tweet HaystackCOMRADES project
Leon Derczynski - University of Sheffield,
Kenny Meesters - TU Delft, Kalina Bontcheva - University of Sheffield, Diana Maynard- University of Sheffield
WiPe Paper – Social Media Studies
Proceedings of the 15th ISCRAM Conference – Rochester, NY, USA May 2018
From drones to old-fashioned phone calls, data come from many unlikely sources. In a disaster, such as a flood or earthquake, responders will take whatever information they can get to visualise the crisis and best direct their resources. Increasingly, cities prone to natural disasters are learning to better aid their citizens by empowering their local agencies and responders with sophisticated tools to cut through the large volume and velocity of disaster-related data and synthesise actionable information.
Twitris in Action - a review of its many applications Amit Sheth
Twitris is a System for Collective Social Intelligence. It has been used in a large number of and many types (disaster coordination, banding, epidemiology, public health, election/polical, social movement) of applications - often in real-time. This presentation gives a bird-eye review of some of these applications with links to explore them further.
Public Health Crisis Analytics for Gender ViolenceHemant Purohit
Research-progress talk on the use of data analytics methods for one of the major public health crisis in the world Gender-based Violence and the campaign engagement in the initiatives of Non-profit organizations.
The Internet of Things means not just that computing devices have connectivity to the cloud but that they themselves are connected to each other, and therefore that novel applications can be developed in this rich ecosystem. One area for development is linking quantified self wearable sensors with automotive sensors for applications including Fatigue Detection, Real-time Parking and Assistance, Anger/Stress Reduction, Keyless Authentication, and DIY Diagnostics.
Helping Crisis Responders Find the Informative Needle in the Tweet HaystackCOMRADES project
Leon Derczynski - University of Sheffield,
Kenny Meesters - TU Delft, Kalina Bontcheva - University of Sheffield, Diana Maynard- University of Sheffield
WiPe Paper – Social Media Studies
Proceedings of the 15th ISCRAM Conference – Rochester, NY, USA May 2018
From drones to old-fashioned phone calls, data come from many unlikely sources. In a disaster, such as a flood or earthquake, responders will take whatever information they can get to visualise the crisis and best direct their resources. Increasingly, cities prone to natural disasters are learning to better aid their citizens by empowering their local agencies and responders with sophisticated tools to cut through the large volume and velocity of disaster-related data and synthesise actionable information.
Talk given August 29, 2018 at the 1st Biannual Conference on Design of Experimental Search & Information Retrieval Systems (DESIRES 2018). Paper: https://www.ischool.utexas.edu/~ml/papers/lease-desires18.pdf
Big Data for Development: Opportunities and Challenges, Summary SlidedeckUN Global Pulse
Summary points from UN Global Pulse White Paper "Big Data for Development: Opportunities & Challenges." See: http://www.unglobalpulse.org/BigDataforDevelopment
Analysing Social Media Conversations to Understand Public Perceptions of Sani...UN Global Pulse
The United Nations Millennium Campaign and the Water Supply and Sanitation Collaborative Council partnered to deliver a comprehensive advocacy and communication drive on sanitation. Their efforts were in support of the UN Deputy Secretary General’s Call to Action on Sanitation to increase the number of people with access to better sanitation. Global Pulse provided an analysis of social media in order to provide insight on the baseline of public engagement, and explore ways to monitor a new sanitation campaign. Using a custom keyword taxonomy, English language tweets from January 2011 to December 2013 were extracted, sorted into categories and analysed.
Cite as: UN Global Pulse, 'Analysing Social Media Conversations to Understand Public Perceptions of Sanitation', Global Pulse Project Series, no.5, 2014.
Global Pulse: Mining Indonesian Tweets to Understand Food Price Crises copyUN Global Pulse
Sudden increases in the price of staple foodstuffs like rice can push whole families below the poverty line and cause regional economic instability; these changes can happen rapidly but food price statistics are generally published only monthly or even less frequently.
This project, in collaboration with the Indonesian Ministry of Development Planning, UNICEF and WFP in Indonesia seeks to use social media analysis to provide real-time information from the population that could enable faster responses to food price increases in the form of social protection policies. Global Pulse analysed tweet volumes relevant to food and fuel between March 2011 and April 2013 and found a significant correlation, suggesting that even potential (rather than realised) fuel price rises affect people’s perceptions of food security. Researchers also found a relationship between retrospective official food inflation statistics and the number of tweets referencing food price increases.
http://www.unglobalpulse.org/social-media-social-protection-indonesia
"Big Data for Development: Opportunities and Challenges" UN Global Pulse
This White Paper is the culmination of UN Global Pulse’s research, collaborations, and consultations with experts to begin a dialogue around Big Data for Development. See: http://www.unglobalpulse.org/BigDataforDevWhitePaper
Understanding Immunisation Awareness and Sentiment with Social Media - Projec...UN Global Pulse
This multi-country study aims to track and analyse online conversations related to immunisation on social media and mainstream media in India, Kenya, Nigeria and Pakistan. Findings from the study showed that in social media, Nigerian and Pakistani politicians are active and influential in the vaccination debate and the political dimension is often referred to when discussing the failure to eradicate diseases such as polio. However, in Kenya, religious and ideological aspects were more frequently discussed. Twitter activity is primarily driven by sharing of news stories in all countries whereas Facebook focuses on the 'distrust' and 'ideals' categorisation.
Cite as: UN Global Pulse, “Understanding Immunisation Awareness and Sentiment Through Social and Mainstream Media”, Global Pulse Project Series no. 19, 2015.
Living in Tech City: 50+ Technology Trends and Innovations Transforming Workp...cjbonk
Abstract: This session is geared toward trainers, managers, instructional designers, educators, learners, practitioners, and government officials who share an interest in contemporary advances in learning technologies that are shaping education for today’s and tomorrow’s learner. In this session, Professor Curt Bonk of Indiana University will discuss dozens of technologies and Web resources that have emerged over the past few years to transform corporate training as well as higher education and most other learning settings. Among these technologies tools are smartphones and smart watches, digital course resources, social books, social media, online talking dictionaries, video walls, virtual assistants, and Web conferencing. Also exploding at this time is enrollment in online or virtual learning, blended learning, massive open online courses (MOOCs), and the use of collaborative tools in such e-learning courses. While these 50+ technology trends and innovation are exciting and highly transformative, each has pros and cons in how they are used in different training and education spaces. To make it more personal, this session will, in part, be a presentation, and, in part, a conversation about learning technology trends and innovations. As such, there will be much opportunity for question and answer as well as personal reflection.
Given the growth of social media and rapid evolution of Web of Data, we have unprecedented opportunities to improve crisis response by extracting social signals, creating spatio-temporal mappings, performing analytics on social and Web of Data, and supporting a variety of applications. Such applications can help provide situational awareness during an emergency, improve preparedness, and assist during the rebuilding/recovery phase of a disaster. Data mining can provide valuable insights to support emergency responders and other stakeholders during crisis. However, there are a number of challenges and existing computing technology may not work in all cases. Therefore, our objective here is to present the characterization of such data mining tasks, and challenges that need further research attention for leveraging social media and Web of Data to assist crisis response coordination.
The Future of Life Sciences 2013 for Max Planck InstituteMelanie Swan
Top 10 List of Life Sciences Opportunities - The next wave of the biotechnology revolution is underway and promises to reshape the world in ways even more transformative than the agricultural, industrial and information revolutions that preceded it. It is not nimaginable that at some point, all biological processes, human and otherwise, will be understood and managed. Some of the most likely sources of life sciences discontinuities are genomic sequencing and synthesis, synthetic biology, nanoscience and aging.
Data privacy and security in ICT4D - Meeting Report UN Global Pulse
On May 8th, 2015 UN Global Pulse hosted a workshop on data privacy and security in technology-enabled development projects and programmes, as part of a series of events about the Nine Principles for Digital Development. This report summarizes the presentations and discussions from the workshop. http://unglobalpulse.org/blog/improving-privacy-and-data-security-ict4d-projects
Digital Demography - WWW'17 Tutorial - Part IIIngmar Weber
Second part of a tutorial given at WWW'17 (http://www2017.com.au/) on Digital Demography. More information about the tutorial at https://sites.google.com/site/digitaldemography/. Please reference the archival tutorial description (at http://papers.www2017.com.au.s3-website-ap-southeast-2.amazonaws.com/companion/p935.pdf) when using the material.
Processing Social Media Messages in Mass Emergency: A SurveyMuhammad Imran
Millions of people use social media to share information during disasters and mass emergencies. Information available on social media, particularly in the early hours of an event when few other sources are available, can be extremely valuable for emergency responders and decision makers, helping them gain situational awareness and plan relief efforts. Processing social media content to obtain such information involves solving multiple challenges, including parsing brief and informal messages, handling information overload, and prioritizing different types of information. These challenges can be mapped to information processing operations such as filtering, classifying, ranking, aggregating, extracting, and summarizing. This work highlights these challenges and presents state of the art computational techniques to deal with social media messages, focusing on their application to crisis scenarios.
In emerging markets, eight out of ten small businesses cannot access the loans they need to grow. USAID’s Development Credit Authority (DCA) uses risk-sharing agreements to mobilize local private capital to fill this financing gap. The goal of this collaboration between UN Global Pulse and USAID is to explore how big data could support the work of USAID’s Development Credit Authority. Kenya has become an established tech leader in Africa in recent years – generating greater volumes of digital data as a result. The goal of this study is to explore what new sources of digital data, and methods for analysis, could be helpful in answering the question: “What barriers to accessing loans do small businesses in Kenya face?” Accordingly, this report paints a picture of the big data landscape in Kenya, shows preliminary findings, and lays the groundwork for further investigation.
"Big Data for Development: Opportunities & Challenges” - UN Global PulseUN Global Pulse
Presentation from UN Global Pulse event to launch a new white paper "BIg Data for Development: Challenges and Opportunities" on July 10, 2012 event at UN Headquarters.
Details, and webcast, of the event can be found at: http://unglobalpulse.org/bd4dwebcast
Web 2.0 Technology Building Situational Awareness: Free and Open Source Too...Connie White
covers ways to use web apps, smart phones and free disaster management software like Sahana Eden, which offer agencies free and open source tools to customize and build situational awareness for their own agency or organizational needs.
Using Financial Transaction Data To Measure Economic Resilience To Natural Di...UN Global Pulse
This project explored how financial transaction data can be analysed to better understand the economic resilience of people affected by natural disasters. The project used the Mexican state of Baja California Sur as a case study to assess the impact of Hurricane Odile on livelihoods and economic activities over a period of six months in 2014. The project measured daily Point of Sale transactions and ATM withdrawals at high geospatial resolution to gain insight into the way people prepare for and recover from disaster.
The study revealed that people spent 50% more than usual on items such as food and gasoline in preparation for the hurricane and that recovery time ranged from 2 to 40 days depending on characteristics such as gender or income. Findings suggest that insights from transaction data could be used to target emergency response and to estimate economic loss at local level in the wake of a disaster.
Global Pulse is playing a leading role in helping UN and other development partners adopt more agile processes powered by Big Data to meet the challenges of driving sustainable development in a Post-2015 world. Our initiative has been closely involved in shaping the discussion of a Post-2015 development “data revolution.”
Over the past year, we have focused our efforts on advocating for the responsible use of Big Data, building partnerships for access to real-time data sources, cutting edge data mining tools and data science expertise. At the country level, we continued to expand our network of Pulse Labs to strengthen national and regional capacity for using Big Data. We are pleased to have begun operating our first regional innovation hub in the vibrant East African technology scene with the opening of Pulse Lab Kampala in late 2013. In 2013, our portfolio of innovation projects involved more than 25 partner organizations including UNICEF, UN Development Programme (UNDP), World Food Programme (WFP) and World Health Organisation (WHO).
The Annual Report 2013 summarizes this activity and explains how the UN's data science labs operate and innovate.
Memo for the Danish Emergency Management Agency by student Anna Boye Koldaas, Master of Science (MSc)-student in Security Risk Management at Copenhagen University.
Social Media & Web Mining for Public Services of Smart Cities - SSA TalkHemant Purohit
This talk at Data Science Seminar of SSA presents challenges and methods to model behavior on social media & Web for application opportunities for public services. The talk also demonstrates an in-depth case study of mining intentional behavior from the noisy natural language text of social media messages during disasters and how it could assist emergency services of future smart cities.
Social Media in Crisis Management: ISCRAM Summer School 2011Connie White
This is a lecture for PhD students at a summer school hosted by Information Systems for Crisis Response and Management, ISCRAM www.iscram.org. This lecture covers social media and the information systems concepts that show how social media can support emergency management.
Talk given August 29, 2018 at the 1st Biannual Conference on Design of Experimental Search & Information Retrieval Systems (DESIRES 2018). Paper: https://www.ischool.utexas.edu/~ml/papers/lease-desires18.pdf
Big Data for Development: Opportunities and Challenges, Summary SlidedeckUN Global Pulse
Summary points from UN Global Pulse White Paper "Big Data for Development: Opportunities & Challenges." See: http://www.unglobalpulse.org/BigDataforDevelopment
Analysing Social Media Conversations to Understand Public Perceptions of Sani...UN Global Pulse
The United Nations Millennium Campaign and the Water Supply and Sanitation Collaborative Council partnered to deliver a comprehensive advocacy and communication drive on sanitation. Their efforts were in support of the UN Deputy Secretary General’s Call to Action on Sanitation to increase the number of people with access to better sanitation. Global Pulse provided an analysis of social media in order to provide insight on the baseline of public engagement, and explore ways to monitor a new sanitation campaign. Using a custom keyword taxonomy, English language tweets from January 2011 to December 2013 were extracted, sorted into categories and analysed.
Cite as: UN Global Pulse, 'Analysing Social Media Conversations to Understand Public Perceptions of Sanitation', Global Pulse Project Series, no.5, 2014.
Global Pulse: Mining Indonesian Tweets to Understand Food Price Crises copyUN Global Pulse
Sudden increases in the price of staple foodstuffs like rice can push whole families below the poverty line and cause regional economic instability; these changes can happen rapidly but food price statistics are generally published only monthly or even less frequently.
This project, in collaboration with the Indonesian Ministry of Development Planning, UNICEF and WFP in Indonesia seeks to use social media analysis to provide real-time information from the population that could enable faster responses to food price increases in the form of social protection policies. Global Pulse analysed tweet volumes relevant to food and fuel between March 2011 and April 2013 and found a significant correlation, suggesting that even potential (rather than realised) fuel price rises affect people’s perceptions of food security. Researchers also found a relationship between retrospective official food inflation statistics and the number of tweets referencing food price increases.
http://www.unglobalpulse.org/social-media-social-protection-indonesia
"Big Data for Development: Opportunities and Challenges" UN Global Pulse
This White Paper is the culmination of UN Global Pulse’s research, collaborations, and consultations with experts to begin a dialogue around Big Data for Development. See: http://www.unglobalpulse.org/BigDataforDevWhitePaper
Understanding Immunisation Awareness and Sentiment with Social Media - Projec...UN Global Pulse
This multi-country study aims to track and analyse online conversations related to immunisation on social media and mainstream media in India, Kenya, Nigeria and Pakistan. Findings from the study showed that in social media, Nigerian and Pakistani politicians are active and influential in the vaccination debate and the political dimension is often referred to when discussing the failure to eradicate diseases such as polio. However, in Kenya, religious and ideological aspects were more frequently discussed. Twitter activity is primarily driven by sharing of news stories in all countries whereas Facebook focuses on the 'distrust' and 'ideals' categorisation.
Cite as: UN Global Pulse, “Understanding Immunisation Awareness and Sentiment Through Social and Mainstream Media”, Global Pulse Project Series no. 19, 2015.
Living in Tech City: 50+ Technology Trends and Innovations Transforming Workp...cjbonk
Abstract: This session is geared toward trainers, managers, instructional designers, educators, learners, practitioners, and government officials who share an interest in contemporary advances in learning technologies that are shaping education for today’s and tomorrow’s learner. In this session, Professor Curt Bonk of Indiana University will discuss dozens of technologies and Web resources that have emerged over the past few years to transform corporate training as well as higher education and most other learning settings. Among these technologies tools are smartphones and smart watches, digital course resources, social books, social media, online talking dictionaries, video walls, virtual assistants, and Web conferencing. Also exploding at this time is enrollment in online or virtual learning, blended learning, massive open online courses (MOOCs), and the use of collaborative tools in such e-learning courses. While these 50+ technology trends and innovation are exciting and highly transformative, each has pros and cons in how they are used in different training and education spaces. To make it more personal, this session will, in part, be a presentation, and, in part, a conversation about learning technology trends and innovations. As such, there will be much opportunity for question and answer as well as personal reflection.
Given the growth of social media and rapid evolution of Web of Data, we have unprecedented opportunities to improve crisis response by extracting social signals, creating spatio-temporal mappings, performing analytics on social and Web of Data, and supporting a variety of applications. Such applications can help provide situational awareness during an emergency, improve preparedness, and assist during the rebuilding/recovery phase of a disaster. Data mining can provide valuable insights to support emergency responders and other stakeholders during crisis. However, there are a number of challenges and existing computing technology may not work in all cases. Therefore, our objective here is to present the characterization of such data mining tasks, and challenges that need further research attention for leveraging social media and Web of Data to assist crisis response coordination.
The Future of Life Sciences 2013 for Max Planck InstituteMelanie Swan
Top 10 List of Life Sciences Opportunities - The next wave of the biotechnology revolution is underway and promises to reshape the world in ways even more transformative than the agricultural, industrial and information revolutions that preceded it. It is not nimaginable that at some point, all biological processes, human and otherwise, will be understood and managed. Some of the most likely sources of life sciences discontinuities are genomic sequencing and synthesis, synthetic biology, nanoscience and aging.
Data privacy and security in ICT4D - Meeting Report UN Global Pulse
On May 8th, 2015 UN Global Pulse hosted a workshop on data privacy and security in technology-enabled development projects and programmes, as part of a series of events about the Nine Principles for Digital Development. This report summarizes the presentations and discussions from the workshop. http://unglobalpulse.org/blog/improving-privacy-and-data-security-ict4d-projects
Digital Demography - WWW'17 Tutorial - Part IIIngmar Weber
Second part of a tutorial given at WWW'17 (http://www2017.com.au/) on Digital Demography. More information about the tutorial at https://sites.google.com/site/digitaldemography/. Please reference the archival tutorial description (at http://papers.www2017.com.au.s3-website-ap-southeast-2.amazonaws.com/companion/p935.pdf) when using the material.
Processing Social Media Messages in Mass Emergency: A SurveyMuhammad Imran
Millions of people use social media to share information during disasters and mass emergencies. Information available on social media, particularly in the early hours of an event when few other sources are available, can be extremely valuable for emergency responders and decision makers, helping them gain situational awareness and plan relief efforts. Processing social media content to obtain such information involves solving multiple challenges, including parsing brief and informal messages, handling information overload, and prioritizing different types of information. These challenges can be mapped to information processing operations such as filtering, classifying, ranking, aggregating, extracting, and summarizing. This work highlights these challenges and presents state of the art computational techniques to deal with social media messages, focusing on their application to crisis scenarios.
In emerging markets, eight out of ten small businesses cannot access the loans they need to grow. USAID’s Development Credit Authority (DCA) uses risk-sharing agreements to mobilize local private capital to fill this financing gap. The goal of this collaboration between UN Global Pulse and USAID is to explore how big data could support the work of USAID’s Development Credit Authority. Kenya has become an established tech leader in Africa in recent years – generating greater volumes of digital data as a result. The goal of this study is to explore what new sources of digital data, and methods for analysis, could be helpful in answering the question: “What barriers to accessing loans do small businesses in Kenya face?” Accordingly, this report paints a picture of the big data landscape in Kenya, shows preliminary findings, and lays the groundwork for further investigation.
"Big Data for Development: Opportunities & Challenges” - UN Global PulseUN Global Pulse
Presentation from UN Global Pulse event to launch a new white paper "BIg Data for Development: Challenges and Opportunities" on July 10, 2012 event at UN Headquarters.
Details, and webcast, of the event can be found at: http://unglobalpulse.org/bd4dwebcast
Web 2.0 Technology Building Situational Awareness: Free and Open Source Too...Connie White
covers ways to use web apps, smart phones and free disaster management software like Sahana Eden, which offer agencies free and open source tools to customize and build situational awareness for their own agency or organizational needs.
Using Financial Transaction Data To Measure Economic Resilience To Natural Di...UN Global Pulse
This project explored how financial transaction data can be analysed to better understand the economic resilience of people affected by natural disasters. The project used the Mexican state of Baja California Sur as a case study to assess the impact of Hurricane Odile on livelihoods and economic activities over a period of six months in 2014. The project measured daily Point of Sale transactions and ATM withdrawals at high geospatial resolution to gain insight into the way people prepare for and recover from disaster.
The study revealed that people spent 50% more than usual on items such as food and gasoline in preparation for the hurricane and that recovery time ranged from 2 to 40 days depending on characteristics such as gender or income. Findings suggest that insights from transaction data could be used to target emergency response and to estimate economic loss at local level in the wake of a disaster.
Global Pulse is playing a leading role in helping UN and other development partners adopt more agile processes powered by Big Data to meet the challenges of driving sustainable development in a Post-2015 world. Our initiative has been closely involved in shaping the discussion of a Post-2015 development “data revolution.”
Over the past year, we have focused our efforts on advocating for the responsible use of Big Data, building partnerships for access to real-time data sources, cutting edge data mining tools and data science expertise. At the country level, we continued to expand our network of Pulse Labs to strengthen national and regional capacity for using Big Data. We are pleased to have begun operating our first regional innovation hub in the vibrant East African technology scene with the opening of Pulse Lab Kampala in late 2013. In 2013, our portfolio of innovation projects involved more than 25 partner organizations including UNICEF, UN Development Programme (UNDP), World Food Programme (WFP) and World Health Organisation (WHO).
The Annual Report 2013 summarizes this activity and explains how the UN's data science labs operate and innovate.
Memo for the Danish Emergency Management Agency by student Anna Boye Koldaas, Master of Science (MSc)-student in Security Risk Management at Copenhagen University.
Social Media & Web Mining for Public Services of Smart Cities - SSA TalkHemant Purohit
This talk at Data Science Seminar of SSA presents challenges and methods to model behavior on social media & Web for application opportunities for public services. The talk also demonstrates an in-depth case study of mining intentional behavior from the noisy natural language text of social media messages during disasters and how it could assist emergency services of future smart cities.
Social Media in Crisis Management: ISCRAM Summer School 2011Connie White
This is a lecture for PhD students at a summer school hosted by Information Systems for Crisis Response and Management, ISCRAM www.iscram.org. This lecture covers social media and the information systems concepts that show how social media can support emergency management.
Computing for Human Experience: Semantics empowered Cyber-Physical, Social an...Amit Sheth
Keynote at On the Move conference, October 2011, Greece.
Abstract:
Traditionally, we had to artificially simplify the complexity and richness of the real world to constrained computer models and languages for more efficient computation. Today, devices, sensors, human-in-the-loop participation and social interactions enable something more than a “human instructs machine” paradigm. Web as a system for information sharing is being replaced by pervasive computing with mobile, social, sensor and devices dominated interactions. Correspondingly, computing is moving from targeted tasks focused on improving efficiency and productivity to a vastly richer context that support events and situational awareness, and enrich human experiences encompassing recognition of rich sets of relationships, events and situational awareness with spatio-temporal-thematic elements, and socio-cultural-behavioral facets. Such progress positions us for what I call an emerging era of “computing for human experience” (CHE). Four of the key enablers of CHE are: (a) bridging the physical/digital (cyber) divide, (b) elevating levels of abstractions and utilizing vast background knowledge to enable integration of machine and human perception, (c) convert raw data and observations, ranging from sensors to social media, into understanding of events and situations that are meaningful to humans, and (d) doing all of the above at massive scale covering the Web and pervasive computing supported humanity. Semantic Web (conceptual models/ontologies and background knowledge, annotations, and reasoning) techniques and technologies play a central role in important tasks such as building context, integrating online and offline interactions, and help enhance human experience in their natural environment.
In this talk I will discuss early enablers of CHE including semantics-empowered social networking and sensor Web, and computation of higher level abstractions from raw and phenomenological data. An article in IEEE Internet Computing provides background information: http://bit.ly/HumanExperience
Keynote at: https://www.springer.com/us/book/9783642251054
Event Date: Oct 18, 2011
A Holistic Approach to Evaluating Social Media's Successful Implementation in...Connie White
As emergency management agencies and organizations implement social media and web technology to support crisis information and communication efforts, many question if present strategies are beneficial. This is especially true if social media is being implemented for the first time or has not been experienced in a live disaster. Studies have been conducted providing information on a variety of interactions between Social Media and Emergency Management (SMEM). However, few have taken a formal scientific approach as a means of measurement providing a 'Comprehensive Performance Metric.' Performance metrics need to have consistency while providing room for implementing unique measurement criteria for individualized efforts. We offer a research design using field studies of real world cases, evaluating rural and metropolitan areas. The result produces a set of 'Best Practices' through implementation. By offering a means of measuring success, SMEM can continue to evolve by using a methodologically sound approach using social media.
This is a brief a brief review of current multi-disciplinary and collaborative projects at Kno.e.sis led by Prof. Amit Sheth. They cover research in big social data, IoT, semantic web, semantic sensor web, health informatics, personalized digital health, social data for social good, smart city, crisis informatics, digital data for material genome initiative, etc. Dec 2015 edition.
Digital development and Online Gender-Based ViolenceAnand Sheombar
Online talk held for Cordaid 18th November 2021, on the concept of digital development, and what online gender-based violence (GBV or eVAW) means for the activities of international development NGOs.
Meenakshi Nagarajan,Amit Sheth,Selvam Velmurugan, "Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web Applications," Tutorial at WWW2011, Hyderabad, India, March 28, 2011.
More info at:
http://knoesis.org/library/resource.php?id=1030
http://www2011india.com/tutorialstr27.html
Statement for the Record of Heather Blanchard, Co Founder of CrisisCommons before the Ad Hoc Subcommittee on Disaster Recovery and Intergovernmental Affairs, Homeland Security and Governmental Affairs Committee, United States Senate on May 19, 2011
Smart Data - How you and I will exploit Big Data for personalized digital hea...Amit Sheth
Amit Sheth's keynote at IEEE BigData 2014, Oct 29, 2014.
Abstract from:
http://cci.drexel.edu/bigdata/bigdata2014/keynotespeech.htm
Big Data has captured a lot of interest in industry, with the emphasis on the challenges of the four Vs of Big Data: Volume, Variety, Velocity, and Veracity, and their applications to drive value for businesses. Recently, there is rapid growth in situations where a big data challenge relates to making individually relevant decisions. A key example is personalized digital health that related to taking better decisions about our health, fitness, and well-being. Consider for instance, understanding the reasons for and avoiding an asthma attack based on Big Data in the form of personal health signals (e.g., physiological data measured by devices/sensors or Internet of Things around humans, on the humans, and inside/within the humans), public health signals (e.g., information coming from the healthcare system such as hospital admissions), and population health signals (such as Tweets by people related to asthma occurrences and allergens, Web services providing pollen and smog information). However, no individual has the ability to process all these data without the help of appropriate technology, and each human has different set of relevant data!
In this talk, I will describe Smart Data that is realized by extracting value from Big Data, to benefit not just large companies but each individual. If my child is an asthma patient, for all the data relevant to my child with the four V-challenges, what I care about is simply, “How is her current health, and what are the risk of having an asthma attack in her current situation (now and today), especially if that risk has changed?” As I will show, Smart Data that gives such personalized and actionable information will need to utilize metadata, use domain specific knowledge, employ semantics and intelligent processing, and go beyond traditional reliance on ML and NLP. I will motivate the need for a synergistic combination of techniques similar to the close interworking of the top brain and the bottom brain in the cognitive models.
For harnessing volume, I will discuss the concept of Semantic Perception, that is, how to convert massive amounts of data into information, meaning, and insight useful for human decision-making. For dealing with Variety, I will discuss experience in using agreement represented in the form of ontologies, domain models, or vocabularies, to support semantic interoperability and integration. For Velocity, I will discuss somewhat more recent work on Continuous Semantics, which seeks to use dynamically created models of new objects, concepts, and relationships, using them to better understand new cues in the data that capture rapidly evolving events and situations.
Smart Data applications in development at Kno.e.sis come from the domains of personalized health, energy, disaster response, and smart city.
Emergency Management in the age of social convergencePatrice Cloutier
Conference on social media use in emergency management given at the Social Media in Government Conference on Oct. 3, 2011 for the Conference Board of Canada.
Social Media and Forced Displacement: Big Data Analytics and Machine Learning...UN Global Pulse
UN Global Pulse and UNHCR Innovation Service, an interdepartmental initiative of the Office of the United Nations High Commissioner for Refugees (UNHCR) used data from Twitter to monitor protection issues and the safe access to asylum of migrants and refugees in Europe. The experimental project investigated interactions among refugees, between refugees and host communities, and between refugees and service providers along the way into Europe. This paper summarises the initial findings and lessons learned, and describes the results of ten mini-studies that were developed as part of the project. It outlines the process, questions and methodology used to develop the studies, and presents preliminary observations on how aspects of the Europe Refugee Emergency are related on social media.
Disasters Happen. We need to manage them to minimize the loss to life and property. Disaster management has been received much attention, but has not been touched much by the latest technology. This paper presents an approach to manage disasters using latest and popular technology. We are interested in building a community of researchers who are interested in developing such tools.
Slideshare lost the previous upload which had nearly 70K views. Re-uploading. http://knoesis.org/?q=node/2633
With the explosion in social media (1B+ Facebook users, 500M+ Twitter users) and ubiquitous mobile access (6B+ mobile phone subscribers) sharing their observations and opinions, we have unprecedented opportunities to extract social signals, create spatio-temporal mappings, perform analytics on social data, and support applications that vary from situational awareness during crisis response, preparedness and rebuilding phases to advanced analytics on social data, and gaining valuable insights to support improved decision making.This tutorial weaves three themes and corresponding relevant topics- a.) citizen sensing and crisis mapping, b.) technical challenges and recent research for leveraging citizen sensing to improve crisis response coordination, and c.) experiences in building robust and scalable platforms/systems. It will couple technical insights with identification of computational techniques and algorithms along with real-world examples. We will also do exemplary demos of the features in the Sahana, CrowdMap (Ushahidi's version) and Twitris platforms while elaborating on the practical issues and pitfalls of the development and operation of these large-scale platforms, especially during the real-time crisis response
Keynote talk for NCRM Stream Analytics workshop, 19 January 2017, Manchester.
My talk is called "New and Emerging Forms of Data: Past, Present, and Future” and I will be giving a perspective from my role as one of the ESRC Strategic Advisers for Data Resources, in which I was responsible for new and emerging forms of data and realtime analytics. The talk also includes some of the current work in the Oxford e-Research Centre on Social Machines (the SOCIAM project) and an introduction to the PETRAS Internet of Things project.
The talk raises a number of important issues looking ahead, including massive scale of data that is already being supplied by Internet of Things, the implications of automation in our research, reproducibility and confidence in research results. I will also ask, how can the new forms of data and new research methods enable social scientists to work in new ways, and can we move on from the dependence on the traditional investment in longitudinal studies?
Establishing Global Rules for the Ethical Use of Artificial Intelligence in H...tinokreutzer
Versions of this presentation were given in March 2019 at the International Studies Association 2019 and the PREA Ethics and Humanitarian Research Conference. It briefly presents the techniques and technologies used to understand people affected by humanitarian emergencies. It then introduces ongoing work to demonstrate the feasibility of deploying Natural Language Processing in order to scale up the use of qualitative data in emergencies. Finally, I discusses the ethical implications of this work, and what rules, principles, and other ethical guidance is needed before AI can be used in humanitarian response.
06-04-2024 - NYC Tech Week - Discussion on Vector Databases, Unstructured Data and AI
Discussion on Vector Databases, Unstructured Data and AI
https://www.meetup.com/unstructured-data-meetup-new-york/
This meetup is for people working in unstructured data. Speakers will come present about related topics such as vector databases, LLMs, and managing data at scale. The intended audience of this group includes roles like machine learning engineers, data scientists, data engineers, software engineers, and PMs.This meetup was formerly Milvus Meetup, and is sponsored by Zilliz maintainers of Milvus.
Adjusting primitives for graph : SHORT REPORT / NOTESSubhajit Sahu
Graph algorithms, like PageRank Compressed Sparse Row (CSR) is an adjacency-list based graph representation that is
Multiply with different modes (map)
1. Performance of sequential execution based vs OpenMP based vector multiply.
2. Comparing various launch configs for CUDA based vector multiply.
Sum with different storage types (reduce)
1. Performance of vector element sum using float vs bfloat16 as the storage type.
Sum with different modes (reduce)
1. Performance of sequential execution based vs OpenMP based vector element sum.
2. Performance of memcpy vs in-place based CUDA based vector element sum.
3. Comparing various launch configs for CUDA based vector element sum (memcpy).
4. Comparing various launch configs for CUDA based vector element sum (in-place).
Sum with in-place strategies of CUDA mode (reduce)
1. Comparing various launch configs for CUDA based vector element sum (in-place).
06-04-2024 - NYC Tech Week - Discussion on Vector Databases, Unstructured Data and AI
Round table discussion of vector databases, unstructured data, ai, big data, real-time, robots and Milvus.
A lively discussion with NJ Gen AI Meetup Lead, Prasad and Procure.FYI's Co-Found
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdfGetInData
Recently we have observed the rise of open-source Large Language Models (LLMs) that are community-driven or developed by the AI market leaders, such as Meta (Llama3), Databricks (DBRX) and Snowflake (Arctic). On the other hand, there is a growth in interest in specialized, carefully fine-tuned yet relatively small models that can efficiently assist programmers in day-to-day tasks. Finally, Retrieval-Augmented Generation (RAG) architectures have gained a lot of traction as the preferred approach for LLMs context and prompt augmentation for building conversational SQL data copilots, code copilots and chatbots.
In this presentation, we will show how we built upon these three concepts a robust Data Copilot that can help to democratize access to company data assets and boost performance of everyone working with data platforms.
Why do we need yet another (open-source ) Copilot?
How can we build one?
Architecture and evaluation
Adjusting OpenMP PageRank : SHORT REPORT / NOTESSubhajit Sahu
For massive graphs that fit in RAM, but not in GPU memory, it is possible to take
advantage of a shared memory system with multiple CPUs, each with multiple cores, to
accelerate pagerank computation. If the NUMA architecture of the system is properly taken
into account with good vertex partitioning, the speedup can be significant. To take steps in
this direction, experiments are conducted to implement pagerank in OpenMP using two
different approaches, uniform and hybrid. The uniform approach runs all primitives required
for pagerank in OpenMP mode (with multiple threads). On the other hand, the hybrid
approach runs certain primitives in sequential mode (i.e., sumAt, multiply).
The Building Blocks of QuestDB, a Time Series Databasejavier ramirez
Talk Delivered at Valencia Codes Meetup 2024-06.
Traditionally, databases have treated timestamps just as another data type. However, when performing real-time analytics, timestamps should be first class citizens and we need rich time semantics to get the most out of our data. We also need to deal with ever growing datasets while keeping performant, which is as fun as it sounds.
It is no wonder time-series databases are now more popular than ever before. Join me in this session to learn about the internal architecture and building blocks of QuestDB, an open source time-series database designed for speed. We will also review a history of some of the changes we have gone over the past two years to deal with late and unordered data, non-blocking writes, read-replicas, or faster batch ingestion.
Chatty Kathy - UNC Bootcamp Final Project Presentation - Final Version - 5.23...John Andrews
SlideShare Description for "Chatty Kathy - UNC Bootcamp Final Project Presentation"
Title: Chatty Kathy: Enhancing Physical Activity Among Older Adults
Description:
Discover how Chatty Kathy, an innovative project developed at the UNC Bootcamp, aims to tackle the challenge of low physical activity among older adults. Our AI-driven solution uses peer interaction to boost and sustain exercise levels, significantly improving health outcomes. This presentation covers our problem statement, the rationale behind Chatty Kathy, synthetic data and persona creation, model performance metrics, a visual demonstration of the project, and potential future developments. Join us for an insightful Q&A session to explore the potential of this groundbreaking project.
Project Team: Jay Requarth, Jana Avery, John Andrews, Dr. Dick Davis II, Nee Buntoum, Nam Yeongjin & Mat Nicholas
Transforming Social Big Data into Timely Decisions and Actions for Crisis Mitigation and Coordination
1. Transforming Social Big Data into Timely Decisions
and Actions for Crisis Mitigation and Coordination
Keynote @ Exploitation of Social Media for Emergency Relief and Preparedness (SMERP)
Co-located with: The Web Conference 2018 (formerly WWW)
Lyon, France. 23 April 2018
Prof. Amit Sheth
LexisNexis Ohio Eminent Scholar
Exec. Dir. - Kno.e.sis @ Wright State University
2. Kno.e.sis: Ohio Center of Excellence in Knowledge-
enabled Computing & BioHealth Innovation
NSF
★ Harassment on Social Media
★ Citizen & Physical Sensing
★ Twitris - Collective Intelligence
★ Aerial Surveillance
★ Visual Experience
★ Web Robot Traffic
NIH
★ kHealth - Asthma
★ eDrugTrends
★ eDarkTrends
★ Depression on Social Media
★ Drug Abuse Early Warning
DoD & Industry
★ Metabolomics & Proteomics
★ Medical Info Decisions
★ Human Detection on Synthetic
FMV
★ Sensor & Information
★ Material Genomics
★ Cardiology Semantic Analysis
Kno.e.sis conducts research in AI
techniques that convert physical-cyber-
social big data into smart data, enabling
building of intelligent systems for clinical,
biomedical, policy, and epidemiological
applications.
Example clinical/healthcare applications
include major diseases such as asthma,
obesity, depression, cardiology, dementia
and GI.
This is complemented by social and
development challenges such as marijuana
legalization policy, harassment on social
media, gender-based violence, and disaster
coordination.
15 faculty from 4 colleges +
~60 Funded Students
★ 35+ PhD
★ 15+ MS
★ 10 BS
Kno.e.sis’ research in World Wide Web ranks Wright State University among the top 10 organizations in the world based on 10-yr impact
[MAS: 2016]. Its total budget for currently active projects is $13+ million [2017]. World-class interdisciplinary research is complemented by
exceptional student outcomes and commercialization with local economic impact.
2
8. Mar. 2011
Japan Earthquake
and Tsunami
http://bit.ly/gWboib
Image:http://bit.ly/fl4gEJ
http://cnet.co/jdQgME
http://bit.ly/nP1E4q
Project: Social Media Enhanced Organizational Sensemaking in Emergency Response
9. Facilitates understanding of multi-dimensional social perceptions over SMS,
Tweets, multimedia Web content, electronic news media
Twitris v1: Spatio-Temporal-Thematic
9
10. Twitris v1: Architecture
Meenakshi Nagarajan, Karthik Gomadam, Amit Sheth, Ajith Ranabahu, Raghava Mutharaju and Ashutosh Jadhav, Spatio-Temporal-
Thematic Analysis of Citizen-Sensor Data - Challenges and Experiences,’ Tenth International Conference on Web Information Systems
Engineering, 539 - 553, Oct 5-7, 2009.
10
11. But the amount of data has grown substantially
DEEP is a platform for collaborative secondary data collection, analysis and dissemination for
humanitarian crises. DEEP (thedeep.io) is a joint initiative by seven key humanitarian organizations:
UNOCHA, UNHCR, OHCHR, IDMC, JIPS, ACAPS and IFRC.
Humanitarian
agency
reports
11
14. Sample of Real-World Impact & Media Coverage
Oct 12, 2013
Google's Person Finder and
Google Crisis Response Map
for Phailin to help with crisis
information
Jun 27, 2013
Using crisis mapping to
aid Uttarakhand
Jun 24, 2013
Twitris: Taking Crisis
Mapping to the Next Level
Sep 11, 2012
Could Twitris+ Be Used
for Disaster Response?
Dec 7, 2015
Chennai floods: How social
media and crowdsourcing
helps people on ground
Sep 9, 2014
Digital soldiers emerge
heroes in Kashmir flood
rescue
And many other topics: Emoji, Religion, Gun Violence, Public Policy, Smart City, Health, Election
(currently predicting: Election2012, Brexit, Election2016, ALSenate): http://knoesis.org/amit/media/
15. Google Crisis Map for Hurricane Phalin, which used data from international
participants spearheaded by Twitris team at the Kno.e.sis center.
16
16. Important tags to
summarize Big
Data flow Related
to Oklahoma
tornado
Images and
Videos Related to
Oklahoma
tornado
Twitris: Real Time Information
ICWSM-13 Tutorial: Crisis Mapping & Citizen Sensing 17
17. Influential users are
for respective needs.
Right side shows their
interaction network on
social media.
Engaging with influencers in the self organizing communities can be
very powerful for- a.) getting important information, b.) Correcting
rumors in the network, c.) Propagating important information back into
the citizen sensors community
Influencers to engage with, for specific needs
ICWSM-13 Tutorial: Crisis Mapping & Citizen Sensing 18
18. What-Where-How-Who-Why
Coordination
Influential users to engage
with and resources for
seekers/supplies at a
location, at a timestamp
Contextual
Information for a
chosen topical
tags
ICWSM-13 Tutorial: Crisis Mapping & Citizen Sensing 19
19. Social Data is incredibly rich.
Real-time analysis of
v1: Spatio-Temporal-Thematic
v2: People-Content-Network
v3: Sentiment-Emotion-Intention
v4: Semantic filtering/knowledge
graph, IFTTT, scalability, robustness
Commercial: Cognovi Labs
TWITRIS’ technical Approach to
Understand & Analyze Social Content
20
20. 22
Twitris Technology:
Real-time, Actionable Insights from Social-media
S-E-I
Sentiment-Emotion-Intent
Extracts and assigns structured
sentiment and emotion scoring
from unstructured content to
understand motivation,
feelings, opinion and intent.
S-T-T
Spatio-Temporal-Thematic
Provides thematic context
through analysis of place
and time.
P-C-N
People-Content-Network
Analyzes influential users
and identifies who is being
listened to.
Key Differentiators:
● Comprehensive (above)
● Semantic Processing: use of public and proprietary knowledge.
● Real-time processing: used in live blogging of election debate; coordination during
disasters.
● Scalable: deployed on a large cloud (864 CPUs, 17 TB main, 435 TB disk).
21. Snapshot of Some Real-world Applications/Trials
Kno.e.sis - Ohio Center of Excellence in Knowledge-enabled Computing
Domains: Branding, Disaster Coordination, Social Movements, Election, Development, Epidemiology,..
22. • Coordination during disasters (QCRI, Microsoft Research NYC, CrisisNET, UN)
• Harassment on social media (WSU cognitive scientists)
• Prescription drug and opioid abuse, Cannabis & Synthetic Cannabinoid
epidemiology (Center for Interventions, Treatment and Addictions Research, ….)
• Depressive disorders (Weill Cornell Med)
• Gender-based violence (UNFPA), Zika Spread
• and extensive applications in personalized digital health, public health
(Dayton Children’s Hospital, Wright St Physicians, …)
Highly multidisciplinary team efforts, often with significant partners, with real
world data, intended to achieve real-world impact
Some of the significant human, social & economic
development applications we work on at Kno.e.sis
24
23. • Named Entity Recognition, Implicit Entity
• Relationship Extraction (E.g., ADR)
• Language usage in Social Media
• Exploration of People, Content and Network dynamics
• Sentiment, Emotion, Intent extraction; Opinion mining
• Trust
• Integrated exploitation of Multimodal data (text, photo-satellite images),
sensor/IoT-web-social data and knowledge (PCS applications)
All embodied in Twitris technology, commercialized as Cognovi Labs
Some of the topics on Online Social Media at Knoesis
25
24. Why People-Content-Network + Spatial-Temporal-
Thematic metadata?
(Example of Understanding Crisis Data)
, Offer help, etc.
26
25. Content Analysis: Typical Sub-tasks
• Recognize key entities mentioned in content
– Information Extraction (entity recognition, anaphora resolution, entity
classification..)
– Discovery of Semantic Associations between entities
• Topic Classification, Aboutness of content
– What is the content about?
• Intention Analysis
– Why did they share this content?
27
26. 28
Mining Actionable Information to Support Disaster
Coordination
ACTIONABLE: Timely delivery of
right resources and information to
the right people at right location!
29. Example of coordination during #Oklahoma-tornado response based on
automatically matching need-offer pairs of community members.
More: Purohit et al. Emergency-relief coordination on social media: Automatically matching resource requests and offers. First Monday,
2014.
Community Level: Oklahoma Tornado
31
30. More: Purohit et al. Empowering Crisis Response-Led Citizen Communities: Lessons Learned from JKFloodRelief.org Initiative. IGI Global
book, 2016.
Rescue and Evacuation Stream Map during the historic Jammu & Kashmir Floods in Sep/2014.
Twitris supported the scalable relief effort of JKFloodRelief.org initiative.
Regional Level: Kashmir Floods
32
31. Regional Level: Kashmir Floods
Moriam Nessa
Sep 8th, 11:09am
I do not know if anyone will be reading this
message and if this will be of any help. But I have
my sister who is stranded in Srinagar. She is 9
months pregnant and they need help. We have
been trying to get through the help lines but nothing
is working. Somebody please help ...
ADGPI - Indian Army
Sep 8th, 3:24pm
Jawahar Nagar is heavily flooded. Rescue teams
will be going there. So dont worry they will be all
right. Indian Army is there
Moriam Nessa
Sep 8th, 7:28pm
Hey, Thank you so much for your effort in rescue
operations. But I’m writing to you again about
update of the situation in Jawahar Nagar. My
concern is that a young girl there is pregnant.
Moriam Nessa
Sep 9th, 8:27am
Thank you so much again for your relief work. My
sister has been rescued. All thanks to you and your
team.
33
32. Finding Actionable Information
NUGGETS for Responders
For responders, most important information to manage coordination
dependencies is to know: WHO-WHAT-WHERE-WHEN
● Scarcity of resources → Demand
● Availability of resources → Supply
34
33. Really sparse Signal to Noise:
2M tweets during the first 48 hrs. of #Oklahoma-tornado-2013
➔ 1.3% as the precise resource donation requests to help
➔ 0.02% as the precise resource donation offers to help
• Anyone know how to get involved to
help the tornado victims in
Oklahoma??#tornado #oklahomacity
(OFFER)
• I want to donate to the Oklahoma cause
shoes clothes even food if I can
(OFFER)
• Text REDCROSS to 909-99 to donate to
those impacted by the Moore tornado!
http://t.co/oQMljkicPs (REQUEST)
• Please donate to Oklahoma disaster
relief efforts.: http://t.co/crRvLAaHtk
(REQUEST)
Blog by our colleague Patrick Meier on this analysis: http://irevolution.net/2013/05/29/analyzing-tweets-tornado/
Demand-Supply Identification:
Oklahoma Tornado
35
34. Who are the people to
engage with in the evolving
ad-hoc social community?
Which needs are of utmost
importance?
Actionable information improves decision making process.
Who are the resource
seekers and suppliers?
Questions for social media tool
to support Disaster Response Coordination
Where can I go for volunteering at my
location?
How and Where
can one donate?
ICWSM-13 Tutorial: Crisis Mapping & Citizen Sensing 36
35. Data used for this simulation was based on repurposing of 2013 Boston Bombing dataset.
More: Hampton et al. Constructing Synthetic Social Media Stimuli for an Emergency Preparedness Functional Exercise. ISCRAM-2017.
Designing Real-time Coordination Tools:
Dayton Regional EM Exercise
37
Snapshot
Twitris-based
simulation tool for
filtering social system
Used in a
functional exercise of
emergency response
organizations
City: Dayton
Date: 5/28/14
36. Legends for
Different needs
#OKC
Clicking on a tag
brings contextual
information– relevant
tweets, news/blogs,
and Wikipedia
articles
Incoming Tweets with
need types to give
quick idea of what is
needed and where
currently #OKC
Designing Real-time Coordination Tools:
Oklahoma Tornado
38
38. Real Example: use case of images
Social media is indispensable for
supporting relatively simple but
extremely important user needs
during crisis. This is an illistration.
40
39. A drilled down view of a photo
from a neighborhood of user's
interest.
Powerful metadata extraction
capabilities of Twitris can be
combined
➔ to give more information
on the photo.
Visualization of Images
41
40. Tweets Images Percentages
All 229,384 78,207 34%
India 105,200 35,609 34%
Tamil Nadu 34,387 10,622 31%
Chennai 21,260 6,195 29%
Geo-Tagged Tweets 55%*
* Tweets with longitude and latitude values of Twitter user-- this is very approximate as most of them is
based on location (.e.g., Chennai) provided in the profile of the poster, unless tweet itself has a geocode--
which would be a lot more precise (Twitter tweets have only 1% that are geotagged).
Some Statistics of Images
42
41. Plotting Images on the Map
* Twitris geotagged tweets using the user profile information when the geo-location
information is missing mapping to the center of the city. That’s why many tweets have
the same location.
43
44. ➔ Two annotators from our team (one of them is a Chennai local)
➔ They relied on:
◆ Their knowledge of the area.
◆ The textual and metadata content about the images.
◆ Their friends and other sources of help and information
◆ Directly contact the authors on Twitter.
Crowdsourcing Location Features
46
45. Answers:
And most of the time they don’t answer
or they answer saying there is no
water anymore there.
47
46. 1. Crowdsourcing is very hard due to human training, coordination, and time
difference between team members.
1. 52% of the images were localized by our annotations to a neighborhood
level.
1. Our location extraction tool (LNEx) was able to localize 13% of images
(relying on textual content only) by geoparsing fine-grained location
mentions (i.e., finding the geo-coordinates)
Lessons Learned
48
48. ➔ Disaster Management (Response and Recovery)
◆ Road Closures or Evacuations
◆ Disaster Relief (shelters, food, and donations)
➔ Provide a system for Disaster Assistance Centers
Road Closures
Help Needed
Injuries Reported
...
Automatic Extraction of Location Mentions
50
50. ▪ Previous works categorize location names
based on their types (e.g., building, street)
[Matsuda et al., 2015; Gelernter and Balaji, 2013]
▪ We categorize the location names based on
their geo-coordinates and meaning into:
□ In-area Location Names (inLoc)
□ Location names that are inside the area
of interest.
□ Out-area Location Names (outLoc)
□ Location names that are outside the area
of interest.
□ Ambiguous Location Names (ambLoc)
□ Ambiguous in nature, need more context
or background data for disambiguation
my house
?????
boundingbox.klokantech.com
[#] Matsuda et al. 2015. Annotating geographical entities on microblog text. In The 9th Linguistic Annotation Workshop.
[#] Gelernter et al. 2013. Automatic gazetteer enrichment with user-geocoded data. SIGSPATIAL Workshop.
Location Names Categorization
52
51. Contractions
Abbreviations
Ambiguity
Nicknames
Referring to “Balalok Matric Higher Secondary School” as
“Balalok School”
Referring to “Wright State University” as “WSU”
My backyard, My house, Buffalo
HTown
Mentions in Hashtags #LouisianaFlood
Misspellings sou th kr koil street
Word Shape Problems west mambalam
Ungrammatical Writing Oxford school.west mambalam
53
Challenges of Location Extraction
52. ➔ Cities
➔ Countries
➔ Street names
➔ Neighborhoods
➔ Points of interest
➔ Building names
➔ Organizations
➔ Districts
➔ States
Bounding Box
boundingbox.klokantech.com
54
Building Region-Specific Gazetteers
53. ▪ Regarded as the Wikipedia of maps.
▪ Contains more fine-grained locations than any other resource.
▪ More accurate geo-coordinates in comparison with Geonames [#]
▪ and, it has a strong volunteer foundation (such as hotosm.org) which maps
thousands of locations during a disaster.
55
OpenStreetMap: Our Choice
[#] Gelernter et al. 2013. Automatic gazetteer enrichment with user-geocoded data. SIGSPATIAL Workshop.
54. ★ Capturing different forms of a toponym (to improve recall)
○ “Balalok Matric Higher Secondary School” → “Balalok School”
★ Recording alternative names (to improve recall)
○ “Anna Salai (Mount Road)” → “Anna Salai” and “Mount Road”
★ Filtering toponym names (to improve recall)
○ Break records: “Tamilnadu Housing Board Road , Ayapakkam”
★ Filtering out very noisy toponyms (to improve precision)
https://foursquare.com/
Ball
Town in Louisiana
Clinton
Town in Louisiana
Jackson
City in Mississippi
56
Gazetteers Preprocessing
58. Our Chain of Possibility Disaster Response Tool
(scenario of supply demand match during flood crisis)
60
External link to the video: https://youtu.be/01vdzYmS-ck
59. Medical Help
Number of people
needing help
Type of help
needed
Example Smartphone App for Rescue/Response
61
Food/Shelter
61. Special
Thanks
Hazards SEES Current Team Members
Hussein S. Al-Olimat
(Graduate Student)
Valerie Shalin
(Faculty - WSU)
Collaborators
Hazards SEES project
is funded by NSF
Award#:
EAR 1520870
Dr. Krishnaprasad
Thirunarayan
(Faculty - WSU)
Hemant Purohit
(Faculty - GMU)
Shruti Kar
(Graduate Student)
Manas Gaur
(Graduate Student)
Amelie Gyrard
(Post-Doctoral
Researcher)
Hazards SEES Past
Members:
Sarasi Lalithsena, Pavan Kapanipathi,
Siva Kumar, Saeedeh Shekarpour, and
Tanvi Banerjee
Hussein Al-Olimat SoonJye Kho
(Graduate Student) (Graduate Student)
Srinivasan
Parthasarathy
(Faculty -OSU)
63
62. Significant components of this talk is from the tutorials given by my team at
• WWW2011: “Citizen Sensor Data Mining, Social Media Analytics and Development Centric Web
Applications,” with Meena Nagarajan and Selvam Velmurugan.
• ICWSM2013: “Crisis Mapping, Citizen Sensing and Social Media Analytics: Leveraging Citizen Roles
for Crisis Response,” with Carlos Castillo, Patrick Meier, and Hemant Purohit.
• ISCRAM 2018: “Location Extraction and Georeferencing in Social Media: Challenges, Techniques, and
Applications” with Hussein Al-Olimat, Amir Yazdavar, and T.K. Prasad.
Contributors to Twitris and/or Semantic Social Web Research @ Kno.e.sis: L. Chen, H. Purohit, W. Wang
with: P. Anantharam, A. Jadhav, P. Kapanipathi, Dr. T.K. Prasad, and
alumni: K. Gomadam, M. Nagarajan, A. Ranabahu
Funding: NSF, AFRL, NIH.
Acknowledgements
64