Searching News Articles Using an Event Knowledge Graph Leveraged by Wikidata

Rudnik, Charlotte; Ehrhart, Thibault; Ferret, Olivier; Teyssou, Denis; Troncy, Raphaël; Tannier, Xavier

doi:10.1145/3308560.3316761

Cited by 42 publications

(34 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The latter have also been used to extract events from real time news, with a particular focus on people, organisations, places and dates when semantically enriching documents [13]. Semantic labeling thanks to Wikidata and Wikipedia [5] has been proven useful in semantic search engines [21] with annotated press articles.…”

Section: Related Workmentioning

confidence: 99%

“…Wikidata supplies two different event identifiers, which point subtleties: some apply for breaking events, others for event with premisses. In this paper, we conform to the Wikidata Event Type (WET) [21] definition which accepts both. We refer to events as happenings in the real world which have spatio-temporal anchors and additional entities involved in it.…”

Section: The Extraction Of Elementary Event Informationmentioning

confidence: 99%

“…The number of events reported is not uniform over the years but tends to grow since the beginning of the 21 st century. This increase must not be interpreted as an increase of events happening in the world but as a better data quality, especially with events now better anchored in time [21]. Therefore, we decided to only focus on events happening in the last fifty years, from January 1970 to December 2019.…”

Section: Datasets Descriptionmentioning

confidence: 99%

See 2 more Smart Citations

A Comprehensive Extraction of Relevant Real-World-Event Qualifiers for Semantic Search Engines

Bernard

Suire

Faucher

et al. 2021

Linking Theory and Practice of Digital Libraries

View full text Add to dashboard Cite

In this paper, we present an efficient and accurate method to represent events from numerous public sources, such as Wikidata or more specific knowledge bases. We focus on events happening in the real world, such as festivals or assassinations. Our method merges knowledge from Wikidata and Wikipedia article summaries to gather entities involved in events, dates, types and labels. This event characterization procedure is extended by including vernacular languages. Our method is evaluated by a comparative experiment on two datasets that shows that events are represented more accurately and exhaustively with vernacular languages. This can help to extend the research that mainly exploits hub languages, or biggest language editions of Wikipedia. This method and the tool we release will for instance enhance event-centered semantic search engines, a context in which we already use it. An additional contribution of this paper is the public release of the source code of the tool, as well as the corresponding datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: The Extraction Of Elementary Event Informationmentioning

confidence: 99%

Section: Datasets Descriptionmentioning

confidence: 99%

See 1 more Smart Citation

A Comprehensive Extraction of Relevant Real-World-Event Qualifiers for Semantic Search Engines

Bernard

Suire

Faucher

et al. 2021

Linking Theory and Practice of Digital Libraries

View full text Add to dashboard Cite

show abstract

“…Previous researches mainly focus on optimizing traditional retrieval models through hand-crafted features of new articles [1,10]. However, these approaches are usually unsatisfactory, as they do not consider the essential natures of the event with event evolution.…”

Section: Introductionmentioning

confidence: 99%

An Event-Oriented Neural Ranking Model for News Retrieval

Zhao

Qian

Zang

et al. 2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

Event-oriented news retrieval (ENR) is the task of retrieving news articles related to the specific event in response to the event-oriented query. Previous approaches usually focus on optimizing traditional retrieval models through hand-crafted features from the perspective of new articles. However, these approaches often fail to work well in reality, as they do not consider the essential natures of the event, i.e., dynamics, coupling. In this paper, we propose a novel and effective event-oriented neural ranking model for news retrieval (ENRMNR). Our model exploits a deep attention mechanism to tackle the dynamics and coupling derived from event evolution. Specifically, the word-level bidirectional attention allows the model to identify which query words about the subevent are related to the news article words, and vice-versa, in order to tackle the dynamics. Moreover, the hierarchical attention at passage-level and documentlevel allows it to capture fine-grained event representations for the coupling between different events within a news article. Experimental results on real-world datasets demonstrate that ENRMNR model significantly outperforms competitive models.

show abstract

“…Furthermore, the majority of the highly-relevant background articles for that type of queries do mention the corresponding events. Consequently, we suggest that pre-identification of those "event-driven" articles (by adopting event extraction techniques, e.g., [6]) can lead to more effective retrieval of their background articles. We also observed that, high background relevance of a candidate article needs both the detailed discussion of the (sub)topics mentioned in the query article as well as the addition of newly-introduced subtopics that help the reader of the query article comprehend it.…”

mentioning

confidence: 99%

Why is That a Background Article

Essam

Elsayed

2020

Proceedings of the 29th ACM International Conference on Information &Amp; Knowledge Management

View full text Add to dashboard Cite

News background linking is the problem of finding online resources that can provide valuable context and background information to help the reader comprehend a given news article. While the problem has recently attracted several researchers, however, the notion of background relevance is not well-studied and it requires better understanding to ensure effective system performance. In this paper, we conduct a qualitative analysis on a sample of 25 query news articles and 152 of their corresponding background articles, in addition to 50 of non-background ones. The goal of the study is to shed some light on the relationship between the query articles and the background articles, and provide informative insights for developing more effective background retrieval models. For instance, our analysis shows that event-driven query articles are, on average, harder to process than others, hence they should be handled differently. It also shows that discussing subtopics in detail and adding new informative topics are both essential factors for highly-relevant background articles. Moreover, it shows that a high lexical similarity between a query article and a background one is neither sufficient nor necessary. CCS CONCEPTS • Information systems → Retrieval tasks and goals; Specialized information retrieval.

show abstract

Searching News Articles Using an Event Knowledge Graph Leveraged by Wikidata

Cited by 42 publications

References 13 publications

A Comprehensive Extraction of Relevant Real-World-Event Qualifiers for Semantic Search Engines

A Comprehensive Extraction of Relevant Real-World-Event Qualifiers for Semantic Search Engines

An Event-Oriented Neural Ranking Model for News Retrieval

Why is That a Background Article

Contact Info

Product

Resources

About