A modular real-time clustering pipeline detects events from Twitter streams of millions of entities per minute and introduces new clustering quality metrics evaluated on a Firehose-derived dataset.
Event Detection with Neural Networks: A Rigorous Empirical Evaluation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Detecting events and classifying them into predefined types is an important step in knowledge extraction from natural language texts. While the neural network models have generally led the state-of-the-art, the differences in performance between different architectures have not been rigorously studied. In this paper we present a novel GRU-based model that combines syntactic information along with temporal structure through an attention mechanism. We show that it is competitive with other neural network architectures through empirical evaluations under different random initializations and training-validation-test splits of ACE2005 dataset.
fields
cs.SI 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Real-time Event Detection on Social Data Streams
A modular real-time clustering pipeline detects events from Twitter streams of millions of entities per minute and introduces new clustering quality metrics evaluated on a Firehose-derived dataset.