Narrative Graph: Telling Evolving Stories Based on Event-centric Temporal Knowledge Graph

Abstract

As the main channel for people to obtain information and express their opinions, online media generate a huge amount of unstructured news documents every day and make it difficult for people to perceive major societal events and grasp the evolution of events. Previous studies on storyline generation are generally based on document clustering without considering event arguments and relations between events. Event-centric knowledge graph has been used to facilitate the construction of news documents to form structured event representation. Although some studies have attempted to construct timelines based on event-centric knowledge graphs, it is difficult for timelines to depict the complex structures of event evolution. In this paper, we try to represent news documents as an event-centric knowledge graph, and compress the whole knowledge graph into salient complex events in temporal order to generate storylines named narrative graph. We first collect news documents from news platforms, construct an event ontology, and build an event-centric knowledge graph with temporal relations. Graph neural network is used to detect events, while BERT fine-tuning is leveraged to identify temporal relations between events. Then, a novel generation framework of narrative graph with constraints of coherence and coverage is proposed. In addition, a case study is implemented to demonstrate how to utilize narrative graph to analyze real-world event. The experiment results show that our approach significantly outperforms the baseline approaches.

Keywords

Storyline / event-centric knowledge graph / event evolution / community detection

References

[1] AllanJTopic detection and tracking: Event-based information organization, 2012, Singapore, Springer Science & Business Media

[2] BlondelV D, GuillaumeJ L, LambiotteR, LefebvreE. Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 2008, 2008(10): P10008

[3] CarlsonA, BetteridgeJ, KisielB, SettlesB, HruschkaE R, et al.. Toward an architecture for never-ending language learning. Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence, 2010

[4] CelikkaleB, ErdoganG, ErdemA, ErdemE. Generating visual story graphs with application to photo album summarization. Signal Processing: Image Communication, 2021, 90: 116033

[5] DehghaniN, AsadpourM. SGSG: Semantic graphbased storyline generation in Twitter. Journal of Information Science, 2019, 45(3): 304-321

[6] DingX, LiZ, LiuT, LiaoKELG: An event logic graph, 2019

[7] DongX, GabrilovichE H ^ H ^, LaoN, et al.. Knowledge vault: A web-scale approach to probabilistic knowledge fusion. Proceedings of the 20th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining, 2014

[8] DongT, LiangC H ^. Social media and internet public events. Telematics and Informatics, 2017, 34(3): 726-739

[9] Dsouza A, Tempelmeier N, Yu R, Gottschalk S, Demidova E (2021). Worldkg: A world-scale geographic knowledge graph. Proceedings of the 30th ACM International Conference on Information and Knowledge Management. Gold Coast, Australia, November 1-5, 2021.

[10] GlavašG, ŠnajderJ. Event graphs for information retrieval and multi-document summarization. Expert Systems with Applications, 2012, 41(15): 6904-6916

[11] Gottschalk S, Demidova E (2018). EventKG:Amultilingual event-centric temporal knowledge graph. European SemanticWeb Conference. Heraklion, Greece, June 3-7, 2018.

[12] GottschalkS, DemidovaE. EventKG - The hub of event knowledge on the web-and biographical timeline generation. Semantic Web, 2019, 10(6): 1039-1070

[13] Gottschalk S, Demidova E (2019b). Happening: Happen, predict, infer-event series completion in a knowledge graph. International Semantic Web Conference. Auckland, New Zealand, October 26-30, 2019.

[14] Gottschalk S, Demidova E (2020). EventKG+ BT: Generation of interactive biography timelines from a knowledge graph. European Semantic Web Conference. Heraklion, Greece, June 2-4, 2020.

[15] Gottschalk S, Kacupaj E, Abdollahi S, AlvesD, Amaral G, et al. (2021).OEKG: The open event knowledge graph. Proceedings of the 2nd InternationalWorkshop on Cross-lingual Event-centric Open Analytics co-located with the 30th The Web Conference. Ljubljana, Slovenia, April 12, 2021.

[16] Keith B F, Mitra T (2021). Narrative maps: An algorithmic approach to represent and extract information narratives. Proceedings of the ACM on Human-Computer Interaction. Online Virtual Conference, May 8-13, 2021.

[17] LehmannJ, IseleR, JakobM, JentzschA, KontokostasD, et al.. DBpedia-a large-scale, multilingual knowledge base extracted fromWikipedia. SemanticWeb, 2015, 6(2): 167-195

[18] Li Z, Zhao S, Ding X, Liu T, et al. (2017). EEG: Knowledge base for event evolutionary principles and patterns. Chinese National Conference on Social Media Processing. Beijing, China, September 14-17, 2017.

[19] Li Z, Ding X, Liu T (2018). Constructing narrative event evolutionary graph for script event prediction. Proceedings of the 27th International Joint Conference on Artificial Intelligence. Stockholm, Sweden, July 13-19, 2018.

[20] Li M, Ma T, Yu M, Wu L, Gao T, et al. (2021). Timeline summarization based on event graph compression via time-aware optimal transport. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. Online and Punta Cana, Dominican Republic, November 7-11, 2021.

[21] LiuB, HanF X, NiuD, KongL, LaiK, et al.. Story forest: Extracting events and telling stories from breaking news. ACMTransactions on Knowledge Discovery from Data, 2020, 14(3): 1-28

[22] LvS, HuangL, ZangL, ZhouW, HanJ, et al.. Yet another approach to understanding news event evolution. World Wide Web, 2020, 23(4): 2449-2470

[23] Mahdisoltani F, Biega J, Suchanek F (2015). YAGO3: A knowledge base from multilingualWikipedias. Proceedings of CIDR 2015. California, USA, January 4-7, 2015.

[24] MeiQ, Zhai C (2005). Discovering evolutionary theme patterns from text: An exploration of temporal text mining. Proceedings of the 11th ACM SIGKDD International Conference on Knowledge Discovery in Data Mining. Illinois, USA, August 21-24, 2005.

[25] Nallapati R, FengA, Peng F, Allan J (2004). Event threading within news topics. Proceedings of the 13th ACM International Conference on Information and Knowledge Management. Washington, DC, USA, November 8-13, 2004.

[26] Norambuena B K, Mitra T, North C (2023). A survey on event-based news narrative extraction. arXiv Preprint arXiv: 2302.08351.

[27] RospocherM, ErpV M, VossenP, FokkensA, AldabeI, et al.. Building event-centric knowledge graphs from news. Journal of Web Semantics, 2016, 37: 132-151

[28] SakorA, JozashooriS, NiazmandE, RivasA, et al.. Knowledge4COVID-19: A semantic-based approach for constructing a COVID-19 related knowledge graph from various sources and analyzing treatments' toxicities. Journal of Web Semantics, 2023, 75: 100760

[29] Shahaf D, Yang J, Suen C, Jacobs J, Wang H, et al. (2013). Information cartography: Creating zoomable, largescale maps of information. Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Illinois, USA, August 11-14, 2013.

[30] Van HageW R, MalaiséV, SegersR, HollinkL, SchreiberG. Design and use of the Simple Event Model (SEM). Journal of Web Semantics, 2011, 9(2): 128-136

[31] WangZ, ShouL C ^ C ^, MehrotraS. On summarization and timeline generation for evolutionary tweet streams. IEEE Transactions on Knowledge and Data Engineering, 2015, 27(5): 1301-1315

[32] Wang Q, Li M, Wang X, Parulian N, Han G, et al. (2021). COVID-19 Literature knowledge graph construction and drug repurposing report generation. Proceedings of the 2021 Conference of theNorth American Chapter of the Association for Computational Linguistics: Human Language Technologies: Demonstrations. Online June 6-11, 2021.

[33] Wu W, Li H, Wang H, Zhu K Q (2012). Probase: A probabilistic taxonomy for text understanding. Proceedings of the 2012 ACMSIGMOD International Conference on Management of Data. Scottsdale, AZ, USA, May 20-24, 2012.

[34] Wu J, Zhu X, Zhang C, Hu Z (2020). Event-centric tourism knowledge graph-a case study of Hainan. International Conference on Knowledge Science, Engineering and Management. Hangzhou, China, August 28-30, 2020.

[35] Xu N, Tang X J (2018). Generating risk maps for evolution analysis of societal risk events. Proceedings of KSS 2018. Tokyo, Japan, November 25-27, 2018.

[36] XuN T ^ J. Evolution analysis of societal risk events by risk maps. Journal of Systems Science and Systems Engineering, 2020, 29(4): 454-467

[37] XuanJ, LuoX, LuJ Z ^. Web event evolution trend prediction based on its computational social context. World Wide Web, 2020, 23(3): 1861-1886

[38] Yan R, Wan X, Otterbacher J, Kong L, Li X, et al. (2011). Evolutionary timeline summarization: A balanced optimization framework via iterative substitution. Proceedings of the 34th International ACMSIGIR Conference on Research and Development in Information Retrieval. Beijing, China, July 24-29, 2011.

[39] Yan Z H, Tang X J (2019), Understanding shifts of public opinions on emergencies through social media. Proceedings of KSS 2019. Da Nang, Vietnam, November 29 - December 1, 2019.

[40] YangC C, ShiX W ^ P. Discovering event evolution graphs from news corpora. IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 2009, 39(4): 201-211

[41] ZhanM, LiangH D ^, DongY. Uncertain opinion evolution with bounded confidence effects in social networks. Journal of Systems Science and Systems Engineering, 2019, 28(4): 494-509

[42] Zhang H, Liu X, Pan H J, Song Y Q, Leung C W K (2020). ASER: A large-scale eventuality knowledge graph. Proceedings of the Web Conference 2020. Taipei, April 20-24, 2020.

RIGHTS & PERMISSIONS

Systems Engineering Society of China and Springer-Verlag GmbH Germany