Tweeki: Linking Named Entities on Twitter to a Knowledge Graph

Abstract

To identify what entities are being talked about in tweets, we need to automatically link named entities that appear in tweets to structured KBs like WikiData. Existing approaches often struggle with such short, noisy texts, or their complex design and reliance on supervision make them brittle, difficult to use and maintain, and lose significance over time. Further, there is a lack of a large, linked corpus of tweets to aid researchers, along with lack of gold dataset to evaluate the accuracy of entity linking. In this paper, we introduce (1) Tweeki, an unsupervised, modular entity linking system for Twitter, (2) Tweeki-Data, a large, automatically-annotated corpus of Tweets linked to entities in WikiData, and (3) TweekiGold, a gold dataset for entity linking evaluation. Through comprehensive analysis, we show that Tweeki is comparable to the performance of recent state-of-the-art entity linkers models, the dataset is of high quality, and a use case of how the dataset can be used to improve downstream tasks in social media analysis (geolocation prediction).

Survey on English Entity Linking on Wikidata: Datasets and approaches

“…Or some NIL entities that do not exist in other KGs could exist in Wikidata. Eleven datasets [16,23,24,27,29,33,46,56,69,80] were found for which Wikidata identifiers were available from the start. In the following the datasets are separated by their domain.…”

“…TweekiData and TweekiGold [46] are an automatically annotated corpus and a manually annotated dataset for EL over tweets. TweekiData was created by using other existing tweet-based datasets and linking them to Wikidata data via the Tweeki EL.…”

“…Kensho Derived Wikimedia Dataset [56] 1 [ 86] CLEF HIPE 2020 [29] 3 [ 15,65,89] Mewsli-9 [16] 1 [ 16] TweekiData [46] 1 [ 46] TweekiGold [46] 1 [ 46] Table 9 Ambiguity of mentions (existence of a match does not correspond to a correct match), NYT2018 dataset was not available and LC-QuAD Wikimedia Dataset and the Mewsli-9 training dataset have the largest percentage of exact matches for labels. While TweekiGold is quite ambiguous, deciding on the most prominent entity appears to produce good EL results.…”

“…Some of those approaches solve the ER and EL jointly as an end-to-end task. Besides that, there exist two rule-based approaches [46,100] and two based on graph optimization [60,69].…”

“…Tweeki [46] is an approach focusing on unsupervised EL over tweets. The ER is done by a pre-existing Entity Recognizer [40] which also tags the mentions.…”