Learning to link with wikipedia

By David N. Milne, Ian H. Witten. 2008

1,406 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Communication

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Communication

Citations in the corpus

DKN: Deep Knowledge-Aware Network for News Recommendation (2018, methodology): “First, to distinguish knowledge entities in news content, we utilize the technique of entity linking [31, 36] to disambiguate mentions in texts by associating them with predefined entities in a knowledge graph.”
Deep Joint Entity Disambiguation with Local Neural Attention (2017, methodology): “WLM is a well-known similarity method of (Milne and Witten, 2008).”
Fast and Accurate Annotation of Short Texts with Wikipedia Pages (2010, background): “In summary, the systems of [10, 14] seem unsuitable to annotate on-the-ﬂy short and poorly composed texts, given that they either depend on statistics that hinge on many well-focused spots [14] or they compute sophisticated scoring functions that make the whole process slow [10].”
DeepType: Multilingual Entity Linking by Neural Type System Evolution (2018, methodology): “13Note that LinkCount accuracy is stronger than the one found in (Ferragina and Scaiella 2010) or (Milne and Witten 2008) because newer Wikipedia dumps improve link coverage and reduce link distribution noisiness. curacy significantly improve over LinkCount, and achieve near perfect accuracy on all ”
Entity Linking with a Knowledge Base: Issues, Techniques, and Solutions (2015, methodology): “[58], [62], [75], [76], [78], [79], [91], [96] adopt the Wikipedia Link-basedMeasure (WLM) described in [108], [109] to calcu-”
Collective Entity Resolution with Multi-Focal Attention (2016, methodology): “Most ER systems consist of a mention model , a context model , and a coherence model (Milne and Witten, 2008; Cucerzan, 2007; Ratinov et al., 2011; Hoffart et al., 2011; Hachey et al., 2013).”
Collective Entity Disambiguation with Structured Gradient Tree Boosting (2018, background): “7 AQUAINT (Milne and Witten, 2008), MSNBC (Cucerzan, 2007), and ACE (Ratinov”
Turkers, Scholars, "Arafat" and "Peace": Cultural Communities and Algorithmic Gold Standards (2015, background): “These algorithms are both diverse and numerous, with approaches grounded in areas ranging from network analysis (e.g. [ 26]) to information theory (e.g. [ 35]) to information retrieval (e.g. [ 14, 34]) to the semantic web [13].”
Leveraging Deep Neural Networks and Knowledge Graphs for Entity Disambiguation (2015, background): “…entity relatedness as a crucial evidence, from non-collective approaches that resolve one mention at each time (Mihalcea and Csomai, 2007; Milne and Witten, 2008b; Guo et al., 2013) to collective approaches that leverage the global topical coherence for joint disambiguation (Cucerzan,…”
An open-source toolkit for mining Wikipedia (2013, methodology): “Readers are directed to [22] for more information, including comparison with related work and a broader evaluation using news articles.”
Discovering users' topics of interest on twitter: a first look (2010, background): “Other research has proceeded in this direction, leveraging Wikipedia as the knowledge base for entity disambiguation (and labeling) [11, 13, 14, 8].”
REL: An Entity Linker Standing on the Shoulders of Giants (2020, background): “This line of work has been quickly followed up by information retrieval (IR) researchers [17, 18].”
Robust Entity Linking via Random Walks (2014, methodology): “MSNBC [6], with 20 news articles from 10 diﬀerent topics (2 articles per topic) and 739 mentions in total; (2) AQUAINT, compiled by Milne and Witten [21], with 50 documents and 727 mentions from a news corpus from the Xinhua News Service, the New York Times, and the Associated Press; and (3)…”
Learning-based Multi-Sieve Co-reference Resolution with Knowledge (2012, methodology): “We are the ﬁrst to use context-sensitive disambiguation to Wikipedia, which received a lot of attention recently (Bunescu and Pasca, 2006; Cucerzan, 2007; Mihalcea and Csomai, 2007; Milne and Witten, 2008; Ratinov et al., 2011).”
Annotating and searching web tables using entities, types and relationships (2010, background): “Mining work involving social catalogs often use a relatedness measure between two types [9, 17, 14].”
From TagME to WAT: a new entity annotator (2014, background): “Milne-Witten [14] de.ned as: max(log|in(e1)|, log|in(e2)”
AIDA (2011, background): “This theme has been further pursued in Web mining, most notably, the work of [12] and [10].”
A Survey on Context Learning (2017, methodology): “Table 4 [25] shows the superiority of the ensemble approach as compared to a simple conditional probability based approach (simulating an N-gram approach) and a modification of WLM framework [22] in doing Wikification task.”