Sensing City Potential through Social Data @ ICMU2014 Panel

67 %
33 %
Information about Sensing City Potential through Social Data @ ICMU2014 Panel

Published on February 17, 2014

Author: YutakaArakawa



Introduction of my researches in social sensing. Especially, the following two topics are explained.

1. Sightseeing spots retrieving
2. Photo spots recommendation

Urban Computing: Sensing City Potential through Social Data Yutaka Arakawa, ph.d Associate Professor Nara Institute of Science and Technology, JAPAN

A human works as a sensor • Traditional Sensors • Mote, Micaz, Zigbee, Arduino, etc. • Numerical info. Ex:Temperature, Humidity • Human Sensors • Twitter, Facebook, Foursquare, etc. (User generated contents) • Textual info. Ex: Delicious, Hot, Noisy Crowdsourcing Delicious! I’m in Singapore So Exciting! A Train is delayed.

Learn from 3.11 • 位置登録実績MAP (コロプラ) Before 3.11 Safety status of 3G networks Access log of Location-based social “GAME” After 3.11 3

Research Target What kind of information can be found from social data? City Potential • • • • • Popular Sightseeing Spots Popular Photo Spots Popular Events Movement of people etc...

Our research results 1. Sightseeing spots retrieving 1. Photo spots recommendation

Our research results 1. Sightseeing spots retrieving • Where is the most attractive spot in the city? • Not only the location but also its name. • Social City Maps 1. Photo spots recommendation

Motivation: City Map Only STAR no text, no figure Latest map Old traditional map 3D figure, and title

Finding spots from social data  The number of photo || Famousness Clustering Flickr provides good data. Twitter is noisy. Crandall, D., Backstrom, L., Huttenlocher, D. and Kleinberg, J.: “Mapping the world’s photos”, ACM WWW2009, pp. 761– 770.

Our research issues What attracts the people? Analyze other social data, such as Foursquare, Facebook, etc. Is is really good spots? Develop an android application as an evaluation platform

Our developed application A platform for publishing and evaluating UGM (User Generated Maps)

Other research issues Clustering Naming Output by sklearn Source by place API as a kml file Facebook Facebook Mean Shift Clustering (Flat Kernel) Flickr Foursquare Yahoo Research topic 2 Twitter Foursquare Research topic 7  how to estimate proper BW?  how to change BW dynamically? Wikilocation  how to evaluate ? Google Research topic 1  how to eliminate a noise? Research topic 6 Research topic 3  how to select only famous spots?  how to quantify famousness? Twitter  how to select a representative image or additional information for the cluster ? Research topic 4 Research topic 5  How to take user’s preference into consideration?  how to decide a proper name of each cluster?

Our research results 1. Sightseeing spots retrieving 1. Photo spots recommendation • City potential is represented by photos. • Place and setting recommendation for amateur user.

Photo spot recommendation Camera setting? (ISO, exposure,etc) From where? Which season? What time? How about a weather?

My experience @Singapore From where? Which season? What time? How about a weather? Camera setting? (ISO, exposure,etc)

Architecture of our system


Screen shots Various Conditions can be set. Photos are selected from Flickr.

Screen shots Camera setting info. Navigation

Summary Social Network Services Open data (Government) Find city potential (Our researches) Ubiquitous city

Thank you! This work is partially supported by SCOPE(Strategic Information and Communications R&D Promotion Program).

Add a comment

Related presentations

Presentación que realice en el Evento Nacional de Gobierno Abierto, realizado los ...

In this presentation we will describe our experience developing with a highly dyna...

Presentation to the LITA Forum 7th November 2014 Albuquerque, NM

Un recorrido por los cambios que nos generará el wearabletech en el futuro

Um paralelo entre as novidades & mercado em Wearable Computing e Tecnologias Assis...

Microsoft finally joins the smartwatch and fitness tracker game by introducing the...

Related pages

City Model - City Sensing - Documents

City Model, City Sensing ... Temi City user Tecnico Decisore ... Sensing City Potential through Social Data @ ICMU2014 Panel.
Read more

CU Department of Geography: Academic Departments ...

... Urban Social Problems; Biogeography, Remote Sensing; ... Sensing; City and ... problems through maps, images, and data. ...
Read more

MIT Senseable City Lab

Urban imagination and social innovation through design & science. ... through the Sense able City Lab Consortium can be found here. ... Data Visualization ...
Read more

Center for International Earth Science Information Network

Thematic Guide to Social Science Applications of Remote Sensing. ... International Panel on Climate Change Data ... of Remote Sensing. Data ...
Read more


Mapping, Remote Sensing, and Geospatial Data; Minerals; ... News Dive into the world ... News Releases. Browse through a
Read more

International Conference on Big Data for Official ...

... the second global conference on Big Data for official ... Panel Discussion: “Social Media data ... Panel Discussion: “Remote Sensing Data ...
Read more

International Society for Photogrammetry and Remote Sensing

information packages or data sets; ... designed and used specifically for computer assisted teaching in photogrammetry, remote sensing or spatial ...
Read more

Solar Power Energy Systems | Solar Panels | Pure Energies

Solar Panels; Types of Solar Panels; ... PURE Energies Solar Advisors will help you through the whole process.
Read more