Improving Supervised Classification Using Information Extraction

Du, Mian; Pierce, Matthew; Pivovarova, Lidia; Yangarber, Roman

doi:10.1007/978-3-319-19581-0_1

Cited by 3 publications

(4 citation statements)

References 23 publications

(24 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the reported state-of-the-art results on Sector Classification (Table 2), our best model yields a 10% gain in µ-F1, (Cisse et al, 2013), and a 6% gain in M-F1 (Du et al, 2015). The best µ-F1 and M-F1 results are obtained by the same model.…”

Section: Algorithm (Prior)supporting

confidence: 51%

“…To the best of our knowledge, our previous work (Du et al, 2015) was the only study of the utility of NEs for RCV1 classification. We demonstrated that using a combination of keyword-based and NE-based classifiers works better than either classifier alone.…”

Section: Data and Prior Workmentioning

confidence: 99%

“…M-F1 µ-F1 SVM (Lewis et al, 2004) 29.7 51.3 SVM (Zhuang et al, 2005) 30.1 52.0 Naive Bayes (Puurula, 2012) -70.5 Bloom Filters (Cisse et al, 2013) 47.8 72.4 SVM + NEs (Du et al, 2015) 57 training instances, which is not sufficient for learning word embeddings.…”

Section: Algorithm (Prior)mentioning

confidence: 99%

See 2 more Smart Citations

Comparison of Representations of Named Entities for Document Classification

Pivovarova

Yangarber

2018

Proceedings of the Third Workshop on Representation Learning for NLP

Self Cite

View full text Add to dashboard Cite

We explore representations for multitoken names in the context of the Reuters topic and sector classification tasks (RCV1). We find that: the best way to treat names is to split them into tokens and use each token as a separate feature; NEs have more impact on sector classification than on topic classification; replacing all NEs with special entity-type tokens is not an effective strategy; representing tokens by different embeddings for proper names vs. common nouns does not improve results. We highlight the improvements over state-of-the-art results that our CNN models yield.

show abstract

Section: Algorithm (Prior)supporting

confidence: 51%

Section: Data and Prior Workmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of Representations of Named Entities for Document Classification

Pivovarova

Yangarber

2018

Proceedings of the Third Workshop on Representation Learning for NLP

Self Cite

View full text Add to dashboard Cite

show abstract

“…Additionally, because subword tokenization is not used, their motivation differs from ours that investigates the effectiveness of NEs on subword-based neural networks. Du et al (2015) investigated the use of NEs in non-neural network models on document classification. While they reported use of NEs improve the accuracy of document classification, the contribution to subword-based neural network models was not investigated.…”

Section: Related Workmentioning

confidence: 99%

Improving Neural Language Processing with Named Entities

Matsushita¹,

Makino

Iwakura

2021

Proceedings of the Conference Recent Advances in Natural Language Processing - Deep Learning for Natural Language Processing Me

View full text Add to dashboard Cite

Pretraining-based neural network models have demonstrated state-of-the-art (SOTA) performances on natural language processing (NLP) tasks. The most frequently used sentence representation for neural-based NLP methods is a sequence of subwords that is different from the sentence representation of non-neural methods that are created using basic NLP technologies, such as part-of-speech (POS) tagging, named entity (NE) recognition, and parsing. Most neural-based NLP models receive only vectors encoded from a sequence of subwords obtained from an input text. However, basic NLP information, such as POS tags, NEs, parsing results, etc, cannot be obtained explicitly from only the large unlabeled text used in pretraining-based models. This paper explores use of NEs on two Japanese tasks; document classification and headline generation using Transformer-based models, to reveal the effectiveness of basic NLP information. The experimental results with eight basic NEs and approximately 200 extended NEs show that NEs improve accuracy although a large pretrainingbased model trained using 70 GB text data was used.

show abstract

Tracking Interactions Across Business News, Social Media, and Stock Fluctuations

Karkulahti

Pivovarova

et al. 2016

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

In this paper we study the interactions between how companies are mentioned in news, their presence on social media, and daily fluctuation in their stock prices. Our experiments demonstrate that for some entities these time series can be correlated in interesting ways, though for others the correspondences are more opaque. In this study, social media presence is measured by counting Wikipedia page hits. This work is done in a context of building a system for aggregating and analyzing news text, which aims to help the user track business trends; we show results obtainable by the system.

show abstract

Improving Supervised Classification Using Information Extraction

Cited by 3 publications

References 23 publications

Comparison of Representations of Named Entities for Document Classification

Comparison of Representations of Named Entities for Document Classification

Improving Neural Language Processing with Named Entities

Tracking Interactions Across Business News, Social Media, and Stock Fluctuations

Contact Info

Product

Resources

About