478 Hits in 4.0 sec

Blending Sentence Optimization Weights of Unsupervised Approaches for Extractive Speech Summarization

Noraini Seman, Nursuriati Jamil
2015 Procedia Computer Science  
This paper evaluates the performance of two unsupervised approaches, Maximum Marginal Relevance (MMR) and concept-based global optimization framework for speech summarization.  ...  We propose improved methods by blending each unsupervised approach at sentence level. Sentence level information is leveraged to improve the linguistic quality of selected summaries.  ...  Blending Sentence Optimization Weights of Unsupervised Approaches for Extractive Speech Summarization Noraini Seman and Nursuriati Jamil Sentence Weights in Global Optimization Framework The MMR method  ... 
doi:10.1016/j.procs.2015.05.330 fatcat:2w6cl5liszejplja3z5rc5vwce

Re-ranking Summaries Based on Cross-Document Information Extraction [chapter]

Heng Ji, Juan Liu, Benoit Favre, Dan Gillick, Dilek Hakkani-Tur
2010 Lecture Notes in Computer Science  
This paper describes a novel approach of improving multi-document summarization based on cross-document information extraction (IE).  ...  We describe a method to automatically incorporate IE results into sentence ranking.  ...  Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation here on.  ... 
doi:10.1007/978-3-642-17187-1_42 fatcat:ccwrcmavtvf2tfpc255pl47uvq

Sentiment Analysis Approaches for Social Media Monitoring

Bisma Shah, Farheen Siddiqui
2018 Indian Journal of Computer Science and Engineering  
The current paper aims at studying and providing the comparison of different methods of sentiment analysis used for extracting the polarity (positive, negative or neutral) of social media dataset.  ...  Social media has become an indispensable part of social life. It influences the beliefs, values, and attitudes of people, as well as their intentions and behaviours.  ...  An example of unsupervised learning approach is given as: Opinion Digger Opinion Digger is a Sentence level Sentiment classification approach.  ... 
doi:10.21817/indjcse/2018/v9i1/180901100 fatcat:gb4a3qgezzcotkae6g7sw4bawy

Supervised and Unsupervised Approaches for Controlling Narrow Lexical Focus in Sequence-to-Sequence Speech Synthesis [article]

Slava Shechtman, Raul Fernandez, David Haws
2021 arXiv   pre-print
Although Sequence-to-Sequence (S2S) architectures have become state-of-the-art in speech synthesis, capable of generating outputs that approach the perceptual quality of natural samples, they are limited  ...  and we explore how the different approaches compare when synthesizing in a target voice with or without labeled data.  ...  the sentence-level targets, each phone-level observation in a 10-phone sentence receives a weight of 0.1; a similar approach is applied to the word-level targets).  ... 
arXiv:2101.09940v1 fatcat:h7kavilufjhqjctsmdwhoyp23m

Spoken document understanding and organization

Lin-shan Lee, B. Chen
2005 IEEE Signal Processing Magazine  
The user's instructions can be entered not only by text but possibly through speech as well since speech is a convenient user interface for a variety of user terminals, especially for small handheld devices  ...  Apparently, the most attractive form of the network content will be in multimedia, including speech information.  ...  The unigram and bigram probabilities, as well as the weighting parameters, m 1 , . . . , m 4 , can be further optimized.  ... 
doi:10.1109/msp.2005.1511823 fatcat:3agy74qwjfcunf5xdroqqjzoge

Open-Domain Multi-Document Summarization via Information Extraction: Challenges and Prospects [chapter]

Heng Ji, Benoit Favre, Wen-Pin Lin, Dan Gillick, Dilek Hakkani-Tur, Ralph Grishman
2012 Multi-source, Multilingual Information Extraction and Summarization  
This paper explores novel approaches to taking advantage of cross-document IE for multi-document summarization.  ...  Information Extraction (IE) and Summarization share the same goal of extracting and presenting the relevant information of a document.  ...  Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation hereon.  ... 
doi:10.1007/978-3-642-28569-1_9 dblp:series/tanlp/JiFLGHG13 fatcat:nlu6ve57o5blzblek4mrfwl6la

FADOHS: Framework for Detection and Integration of Unstructured Data of Hate Speech on Facebook Using Sentiment and Emotion Analysis

Axel Rodriguez, Yi-Ling Chen, Carlos Argueta
2022 IEEE Access  
The proposed framework for automatic detection of hate speech (FADOHS) surpasses the most recent methods identified in A Survey on Automatic Detection of Hate Speech in Text in terms of precision, recall  ...  Posts suspected of containing dehumanizing words are preprocessed before being fed to a K-means clustering algorithm, one of the simplest and most popular unsupervised machine learning methods.  ...  ACKNOWLEDGMENT This work was supported in part by the Ministry of Science and Technology in Taiwan through grants No.  ... 
doi:10.1109/access.2022.3151098 fatcat:l56aoqrf65g57hw7c3lcynrb5e


2015 Procedia Computer Science  
of Unsupervised Approaches for Extractive Speech Summarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 612  ...  myocardium hyper-trabeculation degree . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602 Blending Sentence Optimization Weights  ... 
doi:10.1016/s1877-0509(15)01317-4 fatcat:gk3jnjxu3zdwrjjdavlgknvkg4

Sentiment-Aspect Extraction based on Restricted Boltzmann Machines

Linlin Wang, Kang Liu, Zhu Cao, Jun Zhao, Gerard de Melo
2015 Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)  
Aspect extraction and sentiment analysis of reviews are both important tasks in opinion mining.  ...  We propose a novel sentiment and aspect extraction model based on Restricted Boltzmann Machines to jointly address these two tasks in an unsupervised setting.  ...  Model Figure 3 : 3 Prior Feature Extraction The_DT delicious_JJ dishes_NN in_IN the_DT restaurant_NN taste_VBZ great_JJ Sentence Part of Speech Tagging φ2 Aspect φ3 φ1 Aspect_i Sentiment_i  ... 
doi:10.3115/v1/p15-1060 dblp:conf/acl/WangLC0M15 fatcat:k5udmgkct5c35p33bvao72favq

Automatic Comparison of Products based on Opinion Features using Synonym and Jaccard Similarity

Agus Zainal Arifin, Dinial Utami Nurul Qomariah, Muhammad Riduwan, Addien Haniefardy, Yufis Azhar, Rizka Wakhidatus Sholikah, Dini Adni Navastara
2018 2018 Third International Conference on Informatics and Computing (ICIC)  
While LFS is weight of features in each opinion sentence. Equation (1) - (6) shows the calculation process of GFS and LFS. = + (1) For the new extracted features, GFS is initialized with 0.  ...  The process to weighting CSF consists of 2 steps, identifies comparative sentence and computes the weight of CSF.  ... 
doi:10.1109/iac.2018.8780458 fatcat:saqmavxhzrhiho6llhfguqvbum


Dominic W. Massaro, Miguel Á Carreira-Perpiñán, David J. Merrill, Cass Sterling, Stephanie Bigler, Elise Piazza, Marcus Perlman
2008 Proceedings of the 10th international conference on Multimodal interfaces - IMCI '08  
/H2/ Acoustic Feature Analysis /H3/ Evaluating Different Approaches to the Extraction of Features.  ...  In this case, "best" means optimal for classification, and the projection may differ from the PCA projection that we would obtain in an unsupervised training.  ... 
doi:10.1145/1452392.1452432 dblp:conf/icmi/MassaroCMSBPP08 fatcat:af5yk6nxqjeojfrq7yrkyzwvwa

Neuron-level Interpretation of Deep NLP Models: A Survey [article]

Hassan Sajjad and Nadir Durrani and Fahim Dalvi
2021 arXiv   pre-print
The proliferation of deep neural networks in various domains has seen an increased need for interpretability of these methods.  ...  neuron analysis such as model behavior control and domain adaptation along with potential directions for future work.  ...  Using this approach helps to avoid catastrophic forgetting of the general domain while also obtaining an optimal in-domain model.  ... 
arXiv:2108.13138v1 fatcat:enbc53qb35d63kedhi3xiq2wzi

Machine Learning Paradigms for Speech Recognition: An Overview

Li Deng, Xiao Li
2013 IEEE Transactions on Audio, Speech, and Language Processing  
sequential and dynamic nature of speech.  ...  On the other hand, even though ASR is available commercially for some applications, it is largely an unsolved problem-for almost all applications, the performance of ASR is not on par with human performance  ...  Jeff Bilmes for contributions during the early phase (2010) of developing this paper, and for valuable discussions with Geoff Hinton, John Platt, Mark Gales, Nelson Morgan, Hynek Hermansky, Alex Acero,  ... 
doi:10.1109/tasl.2013.2244083 fatcat:fv4qulshnrh4fgzmzb45mkqwmq

Pre-trained Models for Natural Language Processing: A Survey [article]

Xipeng Qiu, Tianxiang Sun, Yige Xu, Yunfan Shao, Ning Dai, Xuanjing Huang
2020 arXiv   pre-print
Next, we describe how to adapt the knowledge of PTMs to the downstream tasks. Finally, we outline some potential directions of PTMs for future research.  ...  Recently, the emergence of pre-trained models (PTMs) has brought natural language processing (NLP) to a new era. In this survey, we provide a comprehensive review of PTMs for NLP.  ...  Acknowledgements We thank Zhiyuan Liu, Wanxiang Che, Minlie Huang, Danqing Wang and Luyao Huang for their valuable feedback on this manuscript.  ... 
arXiv:2003.08271v3 fatcat:ze64wcfecfgs7bguq4vajpsgpu

Deep Learning for Political Science [article]

Kakia Chatsiou, Slava Jankin Mikhaylov
2020 arXiv   pre-print
The latest advances in deep learning methods for NLP are also reviewed, together with their potential for improving information extraction and pattern recognition from political science texts.  ...  New developments in the areas of machine learning, deep learning, natural language processing (NLP), and, more generally, artificial intelligence (AI) are opening up new opportunities for testing theories  ...  a small number of instances while at the same time optimize which elements it needs to learn labels for.  ... 
arXiv:2005.06540v1 fatcat:kz2cbxjrmrfhdlfss5gqziefoq
« Previous Showing results 1 — 15 out of 478 results