id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_6gjuzkl6kbevxi5nkkalt7xhgm	Jason P.C. Chiu	Named Entity Recognition with Bidirectional LSTM-CNNs	2016	14	.pdf	application/pdf	8582	903	66	of encoding partial lexicon matches in neural networks and compare it to existing approaches. given only tokenized text and publicly available word embeddings, our system is competitive on the CoNLL-2003 dataset and surpasses the previously reported state of the art (2011b) proposed an effective neural network model that requires little feature engineering and instead learns important features from Figure 2: The convolutional neural network extracts character features from each word. NER performance, we propose a new lexicon encoding scheme and matching algorithm that can make word features into named entity tag scores. features, emb = Collobert word embeddings, caps = capitalization feature, lex = lexicon features from both SENNA Table 6: F1 score results of BLSTM and BLSTM-CNN models with various additional features; emb = Collobert capitalization features to the BLSTM-CNN models degrades performance for CoNLL and mostly Table 6 shows that on the CoNLL-2003 dataset, using features from both the SENNA lexicon and our	./cache/work_6gjuzkl6kbevxi5nkkalt7xhgm.pdf	./txt/work_6gjuzkl6kbevxi5nkkalt7xhgm.txt