2014_WWW_BTOR
0 likes699 views
Dongpo Deng attended the Linked Data on the Web 2014 workshop and WWW 2014 conference in Seoul from April 7-12. Some key highlights included: - He stayed in a reasonably priced hotel close to the metro that had small rooms and noise until 2pm. - At LDOW 2014 he attended sessions on integration, exploration, and applications of linked data that featured talks on topics like RML mappings, DBpedia exploration, and crowdsourced sensor data. - WWW 2014 had over 600 submissions across 11 areas with a 13% acceptance rate. The keynotes covered graph mining, organizing the digital world, and taming the web. - Dongpo attended talks on crowds
![• Session 1: Introduction and Keynote
• [09:00-09:10] Workshop Introduction (Christian Bizer, Tom Heath,
Sören Auer, Tim Berners-Lee)
• [09:10-10:00] Keynote: Schema.org Update (Ramanathan Guha)
• [10:00-10:30] Discussion
• [10:30-11:00] Coffee Break
• Session 2: Integration
• [11:00-11:20] RML: A Generic Language for Integrated RDF
Mappings of Heterogeneous Data
• [11:20-11:40] Knowledge Base Augmentation using Tabular Data
• [11:40-12:00] AIDA-light: High-Throughput Named-Entity
Disambiguation
• [12:00-12:20] Web-Scale Querying through Linked Data Fragments](https://image.slidesharecdn.com/2014wwwbtor-150208221300-conversion-gate02/85/2014_WWW_BTOR-9-320.jpg)
![• Session 3: Exploration
• [14:00-14:20] DBpedia Viewer – An Integrative Interface for
DBpedia leveraging the DBpedia Service Eco System
• [14:20-14:40] Linked Data Query Wizard: A Novel Interface for
Accessing SPARQL Endpoints
• [14:40-15:00] Programmable Analytics for Linked Open Data
• [15:00-15:20] Will Linked Data Benefit from Inverse Link Traversal?
• Session 4: Linked Data Applications
• [16:00-16:20] daQ, an Ontology for Dataset Quality Information
• [16:20-16:40] Publishing L2TAP Logs to Facilitate Transparency and
Accountability
• [16:40-17:00] Weaving the Web(VTT) of Data
• [17:00-17:20] Social Web Meets Sensor Web: Linked crowdsourced
observation data
• [17:20-17:40] Application of the Linked Data Visualization Model on
Real World Data from the Czech LOD Cloud](https://image.slidesharecdn.com/2014wwwbtor-150208221300-conversion-gate02/85/2014_WWW_BTOR-10-320.jpg)
2014_WWW_BTOR
- 1. LDOW & WWW 2014 Back to office report Dongpo Deng ! April 23, 2014
- 2. Itinerary • April 7 Taipei —> Seoul • April 8 Linked Data on the Web 2014 • April 9 - 11 WWW 2014 • April 12 Seoul —> Taipei
- 3. Hotel • Clean • Good facilities • Reasonable price • Kitchen and washing machine • Close to metro (10 mins on foot) • Noise till to 2 pm • Tiny room • Small desk Inn the City Serviced Residence, COEX (COEX城市服務公寓旅館) ! 424 USD for 5 nights
- 4. Food
- 5. Sightseeing
- 8. LDOW 2014
- 9. • Session 1: Introduction and Keynote • [09:00-09:10] Workshop Introduction (Christian Bizer, Tom Heath, Sören Auer, Tim Berners-Lee) • [09:10-10:00] Keynote: Schema.org Update (Ramanathan Guha) • [10:00-10:30] Discussion • [10:30-11:00] Coffee Break • Session 2: Integration • [11:00-11:20] RML: A Generic Language for Integrated RDF Mappings of Heterogeneous Data • [11:20-11:40] Knowledge Base Augmentation using Tabular Data • [11:40-12:00] AIDA-light: High-Throughput Named-Entity Disambiguation • [12:00-12:20] Web-Scale Querying through Linked Data Fragments
- 10. • Session 3: Exploration • [14:00-14:20] DBpedia Viewer – An Integrative Interface for DBpedia leveraging the DBpedia Service Eco System • [14:20-14:40] Linked Data Query Wizard: A Novel Interface for Accessing SPARQL Endpoints • [14:40-15:00] Programmable Analytics for Linked Open Data • [15:00-15:20] Will Linked Data Benefit from Inverse Link Traversal? • Session 4: Linked Data Applications • [16:00-16:20] daQ, an Ontology for Dataset Quality Information • [16:20-16:40] Publishing L2TAP Logs to Facilitate Transparency and Accountability • [16:40-17:00] Weaving the Web(VTT) of Data • [17:00-17:20] Social Web Meets Sensor Web: Linked crowdsourced observation data • [17:20-17:40] Application of the Linked Data Visualization Model on Real World Data from the Czech LOD Cloud
- 11. Keynote: Schema.org Update by Ramanathan Guha • Background of Schema.org development • One vocabulary understood by all search engines • It’s a vocabulary.Not the vocabulary. • Google Knowledge Graph • Pinterest Rich Pins • Mapping entities across sites
- 12. Web-Scale Querying through Linked Data Fragments
- 13. DBpedia Viewer – An Integrative Interface for DBpedia leveraging the DBpedia Service Eco System http://live.dbpedia.org/
- 14. Social Web Meets Sensor Web: From User-Generated Content to Linked Crowdsourced Observation Data Dong-Po Deng1,3, Guan–Shuo Mai2, Tyng-Ruey Chuang1, Rob Lemmens3, Kwang–Tsao Shao2 ! 1Institute of Information Science and 2Biodiversity Research Center, Academia Sinica, Taiwan 3Faculty of Geo–Information Science and Earth Observation (ITC), University of Twente, Netherlands
- 22. Submission Statistics • 11 subject areas with 22 area chairs ! • Abstracts: 905 • Submissions: 645 • Accepts: 84 (13%) • Suggested as posters: 91
- 23. Statistics per Country • Countries: 49 (submissions), 23 (accepts) Country #Paper USA 246 China 79 UK 40 Germany 38 India 25 Japan 24 Italy 20 Singapore 19 Korea 16 Country #Paper Ecuador 100% Belgium 63% Portugal 50% Israel 36% Spain 24% Denmark 22% Hong Kong 21% US 20% Finland 19%
- 24. Statistics per track Area Submitted Accepted Accept Rate Behavioural Analysis and Personalization 74 14 18.9% Content analysis 71 8 11.3% Crowd Phenamena 43 8 18.6% Internet, Economics & Monetization 35 8 22.9% Security, Privacy, Trust, and Abuse 55 7 12.7% Semantic Web 51 6 11.8% Social Networks and Graph Analysis 126 13 10.3% Software Infrastructure, Performance, Scalability, and Availablity 23 3 13.0% User Interface, Human Factors, and Smart Devices 33 3 9.1% Web mining 91 8 8.8% Web Search 43 6 14.0% Total 645 84 13.0%
- 25. Review Process • Reviewers: • 3 program chairs • 22 Track chairs • 422 PC members • 265 external reviewers • All paper got 3+ reviews • Ravi Kumar’s gradin scheme: +6, +3 -2, -4, -6 • Two-day TPC meeting at Google
- 26. Best paper awards candidates
- 27. A glance of WWW program • 2 workshop/tutorial days + 3 days main conference • 19 workshops • WI&C2014, Vertical Search Relevance, SOCM2014, SIPLEX2014, WebQuality 2014, WS-REST2014, #Microposts2014, SNOW II, DEOS’14, Big Graph Mining, MSM 2014, 4th Temporal Web Analytics, LDOW2014, SRS2014, WOW2014, 2nd PHDA 2014, COOL’2014, WebET 2014, LSNA, BigScholar 2014, • 6+1 tracks • Research, Industry, Developers, W3C, Web Science, PhD Symposium, Demo/Poster • 3 keynotes • Large Graph Mining: Patterns, Cascades, Fraud Detection, and Algorithms (Christos Faloutsos) • Taming the Web (Jong-Deok Choi) • Organizing the Digital World to Empower People to Do More, Know More, and Be More (Qi Lu) • 4 panels
- 29. Some notes • Patterns in graphs • Patterns • Anomaly/ fraud detection • Graph understanding • Time-evolving graph; tensor • Cascades and immunization
- 30. Keynote #3 • Dr. Qi Lu (Microsoft) • Organizing the Digital World to Empower People to Do More, Know More, and Be More
- 47. Plenary Penal WWW: The next 25 years
- 48. • TimBL • toward decentralized system • web science is social thing • social piece will be reborn • Mary Ellen Zurko • mobilized entities for creating context • connect everything in next 25 years web Photo from @jure
- 49. • Yong Hak Kim • how web change our society in past 25 years--> democracy • evolution of education and age • Social divided
- 50. • Unna Huh • E-learning • with having big data advances tech., we can make better social welfare. • One million data leaking scandal private data security • Prabhakar Raghavan • more social sciences needed when we look at future of the web(instead of leaving it to comp science)
- 51. • James Hendler • Take wikipedia as an example. It’s needed to think of • 1.which tech. we should keep going • 2. web services easy to develop ? • 3. social machine human creativity social web activity • make world to be a better place • education uses web tech. help and affect young generation • the internet is connected more than we thought
- 52. • Questions from a audience (Facebook) • opposite to connectivity, the next should be though of setting up the cyber border • Mary: Keep web open • TimBL: • what AI give the social status • Fight to keep the web neutral, open, decentralised - not just on technology/cool stuff - it's a social system says • encouraging us to make the web a decentralised system again, keep binding on open standards • web is needed to re-decentralized
- 53. Talks I attended • 4/9 14:00 - 15:30 Research Track Crowdsourcing 1 • Quizz: Targeted Crowdsourcing with a Billion (Potential) Users • Community-Based Bayesian Aggregation Models for Crowdsourcing • The Wisdom of Minority: Discovering and Targeting the Right Group of Workers for Crowdsourcing • 4/9 16:00 17:30 Industry Track • Analyzing Behavioral Data for Improving Search Experience at Yandex • Twitter’s Recommendation System: Algorithms, Architectures, and Applications
- 54. • 4/10 11:00 - 12:30 Content analysis 1 - Entities • Discovering Emerging Entities with Ambiguous Names • Effective Named Entity Recognition for Idiosyncratic Web Collections • Deduplicating a Places Database • 4/10 14:00 - 15:20 Web Science 2 Communities and Network 1 • The Semantic Evolution of Online Communities • User Churn in Focused Q&A Sites: Characterizations and Prediction • Evolution of Reddit: From the Front Page of the Internet to a Self- referential Community? • A Study of the Online Profile of Enterprise Users in Professional Social Network
- 55. • 4/10 16:00 - 17:20 Web Science 3 Communities and Network 2 • Analysis of Local Online Review Systems as Digital Word-of-Mouth • Modelling Patient Engagement in Peer-to-Peer Healthcare • Songrium: A Music Browsing Assistance Service with Interactive • Web Science Meets Big Smog: Using Web Data to Analyze China’s Smog Disaster • 4/11 11:00 - 12:30 • Web Science 4: Structure on the Web • Graph Structure in the Web — Revisited. • Reachable Subwebs for Traversal-Based Query Execution • Information Network or Social Network? The Structure of the Twitter Follow Graph • Mining Triadic Closure Patterns in Social Networks
- 60. Given two place names n1 and n2, and an approximate geographical location containing both of them, determine if the two names can refer to the same entity.
- 62. Methods • A novel generative language model for places that identifies the core of a place name. E.g., it identifies that in “Fresca’s Peruvian Restaurant”, Fresca’s is the core and Peruvian Restaurant is the description. • A language model that incorporates the spatial context, e.g., the presence of nearby landmarks, parks, malls, airports, major streets and the current city. • The spatial model helps us with distinguishing “Newpark Mall Gap Outlet” from “Newpark Mall Sears Outlet” and matching/non- matching Central Park and Central Park Cafe. • A place name similarity measure that incorporates both core-word detection and spatial context to substantially improve precision and recall for place name matching.
- 68. Demo http://freebase-easy.cs.uni-freiburg.de/ Easy access to the freebase dataset
- 69. SemFacet: Semantic Faceted Search over Yago http://www.cs.ox.ac.uk/isg/tools/SemFacet/ http://semantic-facets.cs.ox.ac.uk:8080/QueryNavigation/
- 70. Exploring the Web of Spatial Data with Facete http://cstadler.aksw.org/facete2-geoknow-demo-y1/ https://github.com/GeoKnow/Facete2
- 71. Business cards • Christian Bizer • Tom Heath • ASKW • Claus Stadler • Leipzig Univ. • Dimitris Kontokostas • AIST • 後藤真孝 (Masataka Goto) • 濱崎雅弘 (Hamsaki Masahiro) • 台⼤大資⼯工 : 李龍豪、︑⿈黃信輔、︑許裕明 • 中研院 : 陳昇瑋