Lessons learnt from the Named Entity rEcognition and Linking (NEEL) challenge series

Abstract

Abstract. The large number of tweets generated daily is providing decision makers with means to obtain insights into recent events around the globe in near real-time. The main barrier for extracting such insights is the impossibility of manual inspection of a diverse and dynamic amount of information. This problem has attracted the attention of industry and research communities, resulting in algorithms for the automatic extraction of semantics in tweets and linking them to machine readable resources. While a tweet is shallowly comparable to any other textual content, it hides a complex and challenging structure that requires domainspecific computational approaches for mining semantics from it. The NEEL challenge series, established in 2013, has contributed to the collection of emerging trends in the field and definition of standardised benchmark corpora for entity recognition and linking in tweets, ensuring high quality labelled data that facilitates comparisons between different approaches. This article reports the findings and lessons learnt through an analysis of specific characteristics of the created corpora, limitations, lessons learnt from the different participants and pointers for furthering the field of entity recognition and linking in tweets.

Unsupervised Context Retrieval for Long-tail Entities

“…en, in our se ing, an unsupervised approach appears more suitable, since structural document features they use, like informativeness or timeliness, are no longer able to be captured. Supervised learning models for EL highly depend on ne-tuned training data from KBs during the candidate selection stage [16]. Related work also nds that EL datasets are skewed towards popular and frequent entities of general-purpose KBs, and emphasizes the need for datasets that focus on the long tail [20].…”

Mining and Leveraging Background Knowledge for Improving Named Entity Linking

“…Derczynski et al [4] analyze the NEL pipelines used for short texts (e.g., tweets, microblogs) and offer solutions for better pre-processing (e.g., language identification, POS tagging, normalization). Rizzo et al [20] summarize the lessons learned during the several editions of the NEEL Challenge with a focus on changes to the annotation methodology, corpus analysis, emerging trends in the design and evaluation of NEL system. The work also includes a long analysis of the evaluation measures (e.g.…”

Improving Language-Dependent Named Entity Detection

“…However, improvements in a certain language usually come at the expense of ease of adaptation to new languages. In addition, the established NER/NEL challenges and tasks of the scientific community like the OKE challenges [8], the NEEL challenge series [9], or the ERD challenges [10] are in the English language and therefore language-dependent improvements are often not in the focus of the research.…”