BERT-based conformal predictor for sentiment analysis

Lysimachos Maltoudoglou, Andreas Paisios, Harris Papadopoulos
Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications, PMLR 128:269-284, 2020.

Abstract

We deal with the Natural Language Processing (NLP) task of Sentiment Analysis (SA) on text, by applying Inductive Conformal Prediction (ICP) on a transformers based model. SA, which is the interpretation and classification of emotions, also referred to as emotional artificial intelligence, can be set up as a Text Classification (TC) problem. Transformers are deep neural network models based on the attention mechanism and make use of transfer learning by being pretrained on a large unlabeled corpus. Transformer based models have been the state of the art for dealing with various NLP tasks ever since they were proposed at the end of 2018. Our classifier consists of the BERT model for turning words into contextualized word embeddings with parameters fine-tuned on the used corpus and a fully connected output layer for performing the classification task. We examine the performance of the underlying BERT model and the proposed ICP on the Large Movie Review dataset consisting of 50000 movie reviews. The results show that the good performance of the underlying classifier is carried on to the ICP extension without any substantial accuracy loss while the provided prediction sets are tight enough to be useful in practise.

Cite this Paper

BibTeX
@InProceedings{pmlr-v128-maltoudoglou20a, title = {BERT-based conformal predictor for sentiment analysis}, author = {Maltoudoglou, Lysimachos and Paisios, Andreas and Papadopoulos, Harris}, booktitle = {Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications}, pages = {269--284}, year = {2020}, editor = {Gammerman, Alexander and Vovk, Vladimir and Luo, Zhiyuan and Smirnov, Evgueni and Cherubin, Giovanni}, volume = {128}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v128/maltoudoglou20a/maltoudoglou20a.pdf}, url = {https://proceedings.mlr.press/v128/maltoudoglou20a.html}, abstract = {We deal with the Natural Language Processing (NLP) task of Sentiment Analysis (SA) on text, by applying Inductive Conformal Prediction (ICP) on a transformers based model. SA, which is the interpretation and classification of emotions, also referred to as emotional artificial intelligence, can be set up as a Text Classification (TC) problem. Transformers are deep neural network models based on the attention mechanism and make use of transfer learning by being pretrained on a large unlabeled corpus. Transformer based models have been the state of the art for dealing with various NLP tasks ever since they were proposed at the end of 2018. Our classifier consists of the BERT model for turning words into contextualized word embeddings with parameters fine-tuned on the used corpus and a fully connected output layer for performing the classification task. We examine the performance of the underlying BERT model and the proposed ICP on the Large Movie Review dataset consisting of 50000 movie reviews. The results show that the good performance of the underlying classifier is carried on to the ICP extension without any substantial accuracy loss while the provided prediction sets are tight enough to be useful in practise.} }
Endnote
%0 Conference Paper %T BERT-based conformal predictor for sentiment analysis %A Lysimachos Maltoudoglou %A Andreas Paisios %A Harris Papadopoulos %B Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2020 %E Alexander Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Evgueni Smirnov %E Giovanni Cherubin %F pmlr-v128-maltoudoglou20a %I PMLR %P 269--284 %U https://proceedings.mlr.press/v128/maltoudoglou20a.html %V 128 %X We deal with the Natural Language Processing (NLP) task of Sentiment Analysis (SA) on text, by applying Inductive Conformal Prediction (ICP) on a transformers based model. SA, which is the interpretation and classification of emotions, also referred to as emotional artificial intelligence, can be set up as a Text Classification (TC) problem. Transformers are deep neural network models based on the attention mechanism and make use of transfer learning by being pretrained on a large unlabeled corpus. Transformer based models have been the state of the art for dealing with various NLP tasks ever since they were proposed at the end of 2018. Our classifier consists of the BERT model for turning words into contextualized word embeddings with parameters fine-tuned on the used corpus and a fully connected output layer for performing the classification task. We examine the performance of the underlying BERT model and the proposed ICP on the Large Movie Review dataset consisting of 50000 movie reviews. The results show that the good performance of the underlying classifier is carried on to the ICP extension without any substantial accuracy loss while the provided prediction sets are tight enough to be useful in practise.
APA
Maltoudoglou, L., Paisios, A. & Papadopoulos, H.. (2020). BERT-based conformal predictor for sentiment analysis. Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 128:269-284 Available from https://proceedings.mlr.press/v128/maltoudoglou20a.html.

Related Material