First International Workshop on Professional Search

Verberne, Suzan; He, Jiyin; Kruschwitz, Udo; Wiggers, Gineke; Larsen, Birger; Russell-Rose, Tony; Vries, Arjen P. de

doi:10.1145/3308774.3308799

Cited by 16 publications

(14 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the way we construct our dataset (Section 3.1), we only have one relevant news article per query and thus MRR is equivalent to MAP. We use a cut-off of 20 at recall since we expect writers to be willing to navigate the ranked list to lower positions [23,38,45]. We report on statistical significance with a paired two-tailed t-test.…”

Section: Experimental Setup 51 Evaluation Metricsmentioning

confidence: 99%

News Article Retrieval in Context for Event-centric Narrative Creation

Voskarides¹,

Meij²,

Sauer

et al. 2021

Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval

View full text Add to dashboard Cite

Writers such as journalists often use automatic tools to find relevant content to include in their narratives. In this paper, we focus on supporting writers in the news domain to develop event-centric narratives. Given an incomplete narrative that specifies a main event and a context, we aim to retrieve news articles that discuss relevant events that would enable the continuation of the narrative. We formally define this task and propose a retrieval dataset construction procedure that relies on existing news articles to simulate incomplete narratives and relevant articles. Experiments on two datasets derived from this procedure show that state-of-the-art lexical and semantic rankers are not sufficient for this task. We show that combining those with a ranker that ranks articles by reverse chronological order outperforms those rankers alone. We also perform an in-depth quantitative and qualitative analysis of the results that sheds light on the characteristics of this task. CCS CONCEPTS• Information systems → Content ranking.

show abstract

Section: Experimental Setup 51 Evaluation Metricsmentioning

confidence: 99%

News Article Retrieval in Context for Event-centric Narrative Creation

Voskarides¹,

Meij²,

Sauer

et al. 2021

Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval

View full text Add to dashboard Cite

show abstract

“…Recent research, in particular the development of the Complex Searcher Model (Maxwell, 2019) and on professional search (Verberne et al, 2019) suggests that the search process is largely one of information foraging and this has implications for ranking models and user interface design.…”

Section: Seeking Searching and Findingmentioning

confidence: 99%

Information flows – Or does it? The complexity of enterprise information management

White¹

2020

Business Information Review

View full text Add to dashboard Cite

Although there has been a degree of acceptance that information is a core business asset almost no research has been carried out into the factors that affect the way that information flows around the enterprise as employees at all levels seek to make use of it to complete tasks and make decisions. These decisions can have a significant impact on business performance and on the career development of employees. The challenge lies in finding ways to track the flow of information, especially where workarounds have been developed by employees to overcome their frustration with ‘approved’ processes. Over the last few years computational ethnography, initially applied to assessing issues relating to culture and communication. is becoming increasing widely used to track the flow of documents. In the wake of Covid19 organisations are going to have to operate in a rapidly changing business environment. Unless the flow of information is better understood short-term changes may be implemented which turn out to be of limited benefit or even undermine the rate of achievement of new business objectives.

show abstract

“…Query by document (QBD) [38,37], is a widely-used practice across professional, domain-specific retrieval tasks [33,35] such as scientific literature retrieval [25,9], legal case law retrieval [2,3,30,34], and patent prior art retrieval [13,28]. In these tasks, the user's information need is based on a seed document of the same type as the documents in the collection.…”

Section: Introductionmentioning

confidence: 99%

Improving BERT-based Query-by-Document Retrieval with Multi-Task Optimization

Abolghasemi¹,

Verberne²,

Azzopardi³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Query-by-document (QBD) retrieval is an Information Retrieval task in which a seed document acts as the query and the goal is to retrieve related documents -it is particular common in professional search tasks. In this work we improve the retrieval effectiveness of the BERT re-ranker, proposing an extension to its fine-tuning step to better exploit the context of queries. To this end, we use an additional document-level representation learning objective besides the ranking objective when fine-tuning the BERT re-ranker. Our experiments on two QBD retrieval benchmarks show that the proposed multi-task optimization significantly improves the ranking effectiveness without changing the BERT re-ranker or using additional training samples. In future work, the generalizability of our approach to other retrieval tasks should be further investigated.

show abstract

First International Workshop on Professional Search

Cited by 16 publications

References 10 publications

News Article Retrieval in Context for Event-centric Narrative Creation

News Article Retrieval in Context for Event-centric Narrative Creation

Information flows – Or does it? The complexity of enterprise information management

Improving BERT-based Query-by-Document Retrieval with Multi-Task Optimization

Contact Info

Product

Resources

About