Text classification and named entities for new event detection

By Giridhar Kumaran, James Allan. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2004

424 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/0229c0eb39efed90db6691469daf0bb7244cf649

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 Information Systems

Citations in the corpus

Evolutionary timeline summarization: a balanced optimization framework via iterative substitution (2011, background): “6 topics belong to different categories of Rule of Interpretation (ROI) [8].”
Learning similarity metrics for event identification in social media (2010, methodology): “idf weight vector and use the cosine similarity metric, as defined in [26], as the feature similarity metric.”
Event Detection and Tracking in Social Streams (2009, methodology): “While most of event detection models use similar algorithms, many variations of the document representation, distance or similarity metrics, and clustering algorithm are proposed in the literature (Kumaran & Allan 2004; Brants, Chen, & Farahat 2003; Yang et al. 2002; Lam et al. 2001).”
Beyond Trending Topics: Real-World Event Identification on Twitter (2011, methodology): “We represent each message as a tf-idf weight vector of its textual content, and use the cosine similarity metric, as defined by Kumaran and Allan [12], as the clustering similarity function σ. Based on our experiments on training data, we perform traditional text processing steps such as stop-word e”
Feature Generation for Text Categorization Using World Knowledge (2005, neutral): “In particular, representations based on phrases [Dumais et al., 1998; Fuernkranz et al., 2000], named entities [Kumaran and Allan, 2004], and term clustering [Lewis and Croft, 1990; Bekkerman, 2003] have been explored.”
UMass at TREC 2004: Novelty and HARD (2004, background): “We were interested in exploring the potential advantage of using different types of words as feedback candidates [16].”
Open domain event extraction from twitter (2012, background): “Also relevant is work on identifying events [23, 10, 6], and extracting timelines [30] from news articles.”