Filters








594 Hits in 5.6 sec

Towards Autoencoding Variational Inference for Aspect-based Opinion Summary [article]

Tai Hoang, Huy Le, Tho Quan
2019 arXiv   pre-print
Ultimately, we present the Autoencoding Variational Inference for Joint Sentiment/Topic (AVIJST) model.  ...  Firstly, we introduce the Autoencoding Variational Inference for Aspect Discovery (AVIAD) model, which extends the previous work of Autoencoding Variational Inference for Topic Models (AVITM) to embed  ...  Furthermore, the model II semi supervised variational autoencoder (SSVAEII-MLP and SSVAEII-CNN) which was first proposed for semi-supervised problem in Kingma, Rezende, Mohamed, and Welling (2014) is  ... 
arXiv:1902.02507v2 fatcat:arwqphelbnfzpba26vpcc4dfcq

Improved Variational Autoencoders for Text Modeling using Dilated Convolutions [article]

Zichao Yang, Zhiting Hu, Ruslan Salakhutdinov, Taylor Berg-Kirkpatrick
2017 arXiv   pre-print
Further, we conduct an in-depth investigation of the use of VAE (with our new decoding architecture) for semi-supervised and unsupervised labeling tasks, demonstrating gains over several strong baselines  ...  We show that with the right decoder, VAE can outperform LSTM language models.  ...  & Le, 2015) ) on both text categorization and sentiment analysis.  ... 
arXiv:1702.08139v2 fatcat:liugemfo5jeblczitt5xpy2fxm

Tibetan Sentiment Classification Method Based on Semi-Supervised Recursive Autoencoders

Xiaodong Yan, Wei Song, Xiaobing Zhao, Anti Wang
2019 Computers Materials & Continua  
We apply the semi-supervised recursive autoencoders (RAE) model for the sentiment classification task of Tibetan short text, and we obtain a better classification effect.  ...  The input of the semi-supervised RAE model is the word vector.  ...  Based on the above related work, this paper applies the semi-supervised recursive autoencoders RAE model to sentiment classification task of Tibetan short text.  ... 
doi:10.32604/cmc.2019.05157 fatcat:7jzusdxtsncsheikdjqgcxg4cy

Deep Variational Semi-Supervised Novelty Detection [article]

Tal Daniel, Thanard Kurutach, Aviv Tamar
2021 arXiv   pre-print
In semi-supervised AD (SSAD), the data also includes a small sample of labeled anomalies. In this work, we propose two variational methods for training VAEs for SSAD.  ...  A recent and promising approach to AD relies on deep generative models, such as variational autoencoders (VAEs), for unsupervised learning of the normal data distribution.  ...  Semi-Supervised Deep Generative Models (SS-DGM) [24] proposed a deep variational generative approach to semi-supervised learning.  ... 
arXiv:1911.04971v3 fatcat:m2ymdjeuuvaodjnvudkr3bveqy

Sentiment Analysis of Social Media via Multimodal Feature Fusion

Kang Zhang, Yushui Geng, Jing Zhao, Jianxin Liu, Wenxiao Li
2020 Symmetry  
Most of the information posted by users on social media has obvious sentimental aspects, and multimodal sentiment analysis has become an important research field.  ...  Previous studies on multimodal sentiment analysis have primarily focused on extracting text and image features separately and then combining them for sentiment classification.  ...  denoising autoencoder, and classified textual and image-fusion features in an unsupervised and semi-supervised manner.  ... 
doi:10.3390/sym12122010 fatcat:r4xhs7pblneyvh67rw2ydfsddi

Deep Learning Models in Software Requirements Engineering [article]

Maria Naumcheva
2021 arXiv   pre-print
In this article we have accomplished the first step of the research on this topic: we have applied the vanilla sentence autoencoder to the sentence generation task and evaluated its performance.  ...  Semi-supervised Sequential Variational Autoencoder (SS-VAE), suggested by Xu et al. in [32] , targets the text classification task.  ...  Deep learning can be supervised, semi-supervised, or unsupervised. In supervised learning the data in the training set has labels, such as class name or target numeric value.  ... 
arXiv:2105.07771v1 fatcat:fxuknpdgmjahlcogbg3t3wy3f4

Table of Contents [EDICS]

2020 IEEE/ACM Transactions on Audio Speech and Language Processing  
Richard 2638 Semi-Supervised Neural Chord Estimation Based on a Variational Autoencoder With Latent Chord Labels and Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  ...  Lee 876 Speech Analysis Semi-Supervised Speech Emotion Recognition With Ladder Networks . . . . . . . . . . . . . . . . . S. Parthasarathy and C.  ... 
doi:10.1109/taslp.2020.3046150 fatcat:easrxuwl6zdppejsrf4bskxfw4

A multimodal feature learning approach for sentiment analysis of social network multimedia

Claudio Baecchi, Tiberio Uricchio, Marco Bertini, Alberto Del Bimbo
2015 Multimedia tools and applications  
In this paper we investigate the use of a multimodal feature learning approach, using neural network based models such as Skip-gram and Denoising Autoencoders, to address sentiment analysis of micro-blogging  ...  , and have shown very good performances when dealing with syntactic and semantic word similarities; ii) unsupervised learning, with neural networks, of robust visual features, that are recoverable from  ...  Note that polarity supervision is limited and possibly weak, thus a robust semi-supervised setting is preferred: on the one hand, a model of sentiment polarity can use the limited supervision available  ... 
doi:10.1007/s11042-015-2646-x fatcat:nbaw3j4opnagfiqmru5lpyxuda

Semi-supervised Structured Prediction with Neural CRF Autoencoder

Xiao Zhang, Yong Jiang, Hao Peng, Kewei Tu, Dan Goldwasser
2017 Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing  
In this paper we propose an end-toend neural CRF autoencoder (NCRF-AE) model for semi-supervised learning of sequential structured prediction problems.  ...  Our experimental results over the Part-of-Speech (POS) tagging task on eight different languages, show that the NCRF-AE model can outperform competitive systems in both supervised and semi-supervised scenarios  ...  Semi-supervised Learning In the semi-supervised settings we compared our models with other semi-supervised structured prediction models.  ... 
doi:10.18653/v1/d17-1179 dblp:conf/emnlp/ZhangJPTG17 fatcat:74zzfoqkyjginic6dt732onxee

Cross Lingual Sentiment Analysis using Modified BRAE

Sarthak Jain, Shashank Batra
2015 Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing  
In this paper, we use the Recursive Autoencoder architecture to develop a Cross Lingual Sentiment Analysis (CLSA) tool using sentence aligned corpora between a pair of resource rich (English) and resource  ...  It is shown that our approach significantly outperforms state of the art systems for Sentiment Analysis, especially when labeled data is scarce.  ...  In Fig. 3 , we show the variation in accuracy of the classifiers with amount of sentiment labeled Training data used.  ... 
doi:10.18653/v1/d15-1016 dblp:conf/emnlp/JainB15 fatcat:twu6u4svpffzpgfubyt5jsieiy

Comparative Study on Generative Adversarial Networks [article]

Saifuddin Hitawala
2018 arXiv   pre-print
We study the original model proposed by Goodfellow et al. as well as modifications over the original model and provide a comparative analysis of these models.  ...  The authors evaluated the performance of adversarial autoencoders on MNIST and Toronto Face datasets using loglikelihood analysis in supervised, semi-supervised and unsupervised settings.  ...  The authors also show how adversarial autoencoders can be used for dimensionality reduction.  ... 
arXiv:1801.04271v1 fatcat:g2quw4bnfzdgpjjvt3kgiwyx44

Neural Structural Correspondence Learning for Domain Adaptation [article]

Yftah Ziser, Roi Reichart
2017 arXiv   pre-print
On the task of cross-domain product sentiment classification (Blitzer et al., 2007), consisting of 12 domain pairs, our model outperforms both the SCL and the marginalized stacked denoising autoencoder  ...  ., 2006)) and autoencoder neural networks.  ...  There is a recent interest in models based on variational autoencoders (Kingma and Welling, 2014; Rezende et al., 2014) , for example the variational fair autoencoder model (Louizos et al., 2016) , for  ... 
arXiv:1610.01588v3 fatcat:ydxjqkd4pnbtzmq5yndiunstiq

Active Learning via Membership Query Synthesis for Semi-Supervised Sentence Classification

Raphael Schumann, Ines Rehbein
2019 Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)  
We present the first successful attempt to use Membership Query Synthesis for generating AL queries for natural language processing, using Variational Autoencoders for query generation.  ...  Training Variational Autoencoder vastava et al., 2014) .  ...  Like other autoencoders, VAEs learn a mapping q θ (z|x) from high dimensional input x to a low dimensional latent variable z.  ... 
doi:10.18653/v1/k19-1044 dblp:conf/conll/SchumannR19 fatcat:hte7yubcijaktbfebga24satqu

Neural Structural Correspondence Learning for Domain Adaptation

Yftah Ziser, Roi Reichart
2017 Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017)  
We experiment with the task of cross-domain sentiment classification on 16 domain pairs and show substantial improvements over strong baselines. 1  ...  ., 2006)) and autoencoder neural networks (NNs).  ...  There is a recent interest in models based on variational autoencoders (Kingma and Welling, 2014; Rezende et al., 2014) , for example the variational fair autoencoder model (Louizos et al., 2016) , for  ... 
doi:10.18653/v1/k17-1040 dblp:conf/conll/ZiserR17 fatcat:hvkpr7vulvhibcjbshc5toeqny

Deep Learning for Sentiment Analysis : A Survey [article]

Lei Zhang, Shuai Wang, Bing Liu
2018 arXiv   pre-print
Along with the success of deep learning in many other application domains, deep learning is also popularly used in sentiment analysis in recent years.  ...  This paper first gives an overview of deep learning and then provides a comprehensive survey of its current applications in sentiment analysis.  ...  Ltd with a research gift.  ... 
arXiv:1801.07883v2 fatcat:nplicfgaozb6fbfx4eyts4zt7e
« Previous Showing results 1 — 15 out of 594 results