Earthquake Knowledge Graph in Japan

NODES 2024 (Dev Conference by Neo4j)

https://neo4j.com/nodes2024/agenda/earthquake-knowledge-graph-in-japan/

Session Track: Graphs

Session Time: 2:30 pm - 2:45 pm November 7

Session description

Japan experiences frequent earthquakes, including a significant one in Noto Peninsula (M7.6) in January 2024. This session will explore structuring and representing earthquake knowledge using Neo4j. The speaker will cover the evolution of theories, key references, preparedness knowledge graphs, and prediction techniques. Attendees will gain valuable insights into constructing comprehensive knowledge systems and enhancing earthquake preparedness. This session is ideal for those interested in knowledge construction and disaster management.

Transcript

1. NODES 2024, Nov 7 2024 14:30-14:45 Earthquake Knowledge Graph in Japan Koji Annoura ҊӜߒೋ koji@annoura.com $PQZSJHIU
   ˜
   ͍Β͢ͱ΍
2. Who am I ❖ Koji Annoura ❖ https://annoura.com/ ❖ I live in Fukuoka Japan ❖ I love ❖ Neo4j Graph Database ❖ Apache Hop ❖ LibreOf fi ce
3. Agenda ❖ Earthquake ❖ Public Data ❖ Knowledge Graph ❖ Research Papers ❖ Earthquake ❖ Next plan ❖ Conclusion $PQZSJHIU
   ˜
   ͍Β͢ͱ΍
4. Earthquake
5. Once upon a time ❖ The Japanese believed this ❖ A giant cat fi sh lived underground and caused earthquakes. $PQZSJHIU
   ˜
   ͍Β͢ͱ΍
6. Catfish club ❖ I was a member ❖ Namazu-no-Kai (Cat fi sh club) is a volunteer group that aims to connect with earthquakes by measuring things like well water levels and hot spring temperatures ❖ Started in May 1976 ❖ https://gbank.gsj.jp/ld/resource/ geolis/200004082.html $PQZSJHIU
   ˜
   ͍Β͢ͱ΍
7. Water level ❖ Water levels ❖ water level gauge to a small well ❖ Before a major earthquake ❖ The water level in wells rises ❖ The temperature and volume of hot springs increase Source: Namazu-no-Kai (Cat fi sh club), https://www.onken.odawara.kanagawa.jp/modules/Information/index.php
8. What do I/you want to know? ❖ How are the references in the paper ❖ The distance to the epicenter? ❖ When will the next earthquake happen? ❖ How strong will the earthquake be? ❖ How do earthquakes work? ❖ How is earthquake prediction going? $PQZSJHIU
   ˜
   ͍Β͢ͱ΍
9. Japan Meteorological Agency ❖ Earthquake and related information is available in multiple languages ❖ Home ❖ Weather ❖ Weather warning / advisory ❖ Weather forecast ❖ Tropical Cyclone Information ❖ Real-time Risk Map ❖ Real-time Risk Map: Landslide ❖ Real-time Risk Map: Inundation ❖ Real-time Risk Map: Flood ❖ Earthquake-related ❖ Tsunami warning / advisory ❖ Earthquake information ❖ Volcanic warning / forecast Source: Japan Meteorological Agency, https://www.data.jma.go.jp/multi/index.html?lang=en
10. Earthquake information ❖ Published 3 seconds after detection ❖ Information ❖ Observed at ❖ Place name of epicenter ❖ Magnitude ❖ Maximum seismic intensity ❖ Date and time of issuance Source: Japan Meteorological Agency, https://www.data.jma.go.jp/multi/quake/index.html?lang=en
11. Detail information ❖ Issued on 2024/11/06 06:45 ❖ Observed at: 2024/11/06 06:42 ❖ Latitude: 36.5N ❖ Longitude: 136.6E ❖ Magnitude: 3.8 ❖ Epicenter depth: 10km ❖ Place: Kaga, Ishikawa Prefecture Source: Japan Meteorological Agency, https://www.data.jma.go.jp/multi/quake/quake_detail.html?eventID=20241106064548&lang=en
12. Explanation of the seismic intensity ❖ Seismic intensity 4 ❖ Almost everyone is surprised. Suspended objects such as electric lights may swing violently, and unstable objects may overturn ❖ but most of Japanese are not surprised ❖ It’s an everyday occurrence Source: Japan Meteorological Agency, https://www.data.jma.go.jp/multi/quake/quake_advisory.html?lang=en
13. Explanation of the seismic intensity ❖ From 2011 to 2024 ❖ seismic intensity ❖ 4 or more, 1244 ❖ 5 or more, 249 ❖ 6 or more, 38 ❖ 7, 5 Source: Japan Meteorological Agency, https://www.data.jma.go.jp/multi/quake/quake_advisory.html?lang=en
14. Public Data
15. Public earthquake data in Japan ❖ Japan Meteorological Agency (JMA): ❖ https://www.jma.go.jp/jma/indexe.html ❖ National Research Institute for Earth Science and Disaster Resilience (NIED): ❖ https://www.bosai.go.jp/ ❖ Geospatial Information Authority of Japan (GSI): ❖ https://www.gsi.go.jp/ ❖ Director General for Disaster Management ❖ https://www.bousai.go.jp/index-e.html Source: Japan Meteorological Agency, https://www.data.jma.go.jp/multi/index.html?lang=en
16. Tellus ❖ Tellus is a satellite data platform ❖ Satellite data ❖ Terrestrial data ❖ Provides the tool APIs Source: Tellus Inc, https://www.tellusxdp.com/en-us/
17. TelluSAR Source: Tellus Inc, https://www.tellusxdp.com/ja/catalog/tool/tellusar.html ❖ TelluSAR ❖ SAR (Synthetic Aperture Radar) data, it is possible to detect surface changes across large areas at the scale of several centimeters ❖ Provider: Tellus ❖ API, QGIS Plugin ❖ Supported Satellites ❖ PALSAR, PALSAR-2, ASNARO-2 ❖ Price: Free
18. Knowledge Graph of Research Papers
19. Knowledge Graph of Research Papers ❖ Example of a research paper ❖ IOP Conference ❖ Knowledge Graph Source: IOP Publishing Ltd, https://iopscience.iop.org/article/10.1088/1755-1315/428/1/012080/pdf
20. Knowledge Graph of Research Papers ❖ 34 References ❖ What is the title? ❖ Who is the author? ❖ Published date? ❖ Easy to read? ❖ No! ❖ So, I converted it to Graph Source: IOP Publishing Ltd, https://iopscience.iop.org/article/10.1088/1755-1315/428/1/012080/pdf
21. Knowledge Graph of Research Papers ❖ References ❖ Nodes ❖ Paper (Red) ❖ Jounal (Blue) ❖ URL (Orange) ❖ Relationship ❖ IS_REFERENCING ❖ IS_AUTHOR
22. Knowledge Graph of Earthquake
23. News-based Earthquake ❖ Example of a research paper ❖ IOP Conference ❖ Knowledge Graph Source: IOP Publishing Ltd, https://iopscience.iop.org/article/10.1088/1755-1315/428/1/012080/pdf
24. News-based Earthquake ❖ Example of a research paper ❖ IOP Conference ❖ Knowledge Graph Source: IOP Publishing Ltd, https://iopscience.iop.org/article/10.1088/1755-1315/428/1/012080/pdf
25. Seismic wave ❖ has waves ❖ Primary wave ❖ 7km/s, Fast ❖ Secondary wave ❖ 4km/s, Slow ❖ Place? (Epicenter) ❖ About (S - P) x 8 ❖ Ex ❖ S - P = 2 ❖ 2 x 8 = 16km Source: Japan Meteorological Agency, https://www.data.jma.go.jp/eew/data/nc/shikumi/shikumi.html
26. Seismic wave ❖ References ❖ Nodes ❖ Wave ❖ Time ❖ Distance ❖ Relationship ❖ IS_TYPE_OF ❖ HAS_TIME
27. Seismic wave ❖ References ❖ Nodes ❖ Wave ❖ Time ❖ Distance ❖ Relationship ❖ IS_TYPE_OF ❖ HAS_TIME
28. You can calculate the distance ❖ Place? (Epicenter) ❖ Time ❖ P is Primary time ❖ S is Secondery time ❖ Formula ❖ (S - P) x 8 ❖ Example ❖ S - P = 2 sec ❖ 2 x 8 = 16km
29. Next plan
30. Misinformation ❖ Noto Peninsula Earthquake ❖ After the Earthquake ❖ A large amount of information is posted in a short time. ❖ False Information ❖ Misleading Information ❖ Fraud ❖ Conspiracy Theory ❖ Rumor ❖ Pseudoscience $PQZSJHIU
    ˜
    ͍Β͢ͱ΍
31. Category-specific misinformation Source: Ministry of Internal Affairs and Communications, https://www.soumu.go.jp/johotsusintokei/whitepaper/ja/r06/pdf/n122000c.pdf
32. GraphRAG ❖ adding GraphRAG over X(Twitter) browsing to see if it solves the issue? $PQZSJHIU
    ˜
    ͍Β͢ͱ΍
33. Conclusion ❖ Summary ❖ There is a lot of open data related to earthquakes in Japan ❖ Creating a knowledge graph is simple and interesting ❖ It can be built easily even without using RDF ❖ Graphs are effective for ensuring reliable relationships Source: Neo4j, https://neo4j.com/product/neo4j-graph-database/
34. Questions ❖ Name ❖ Koji Annoura ❖ Linkedin ❖ kojiannoura ❖ X(Twitter) ❖ @kojiannoura $PQZSJHIU
    ˜
    ͍Β͢ͱ΍