A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Filters
Blending Sentence Optimization Weights of Unsupervised Approaches for Extractive Speech Summarization
2015
Procedia Computer Science
Preserved Fulltext
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]
2010
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
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
2018
Indian Journal of Computer Science and Engineering
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
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]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
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
2005
IEEE Signal Processing Magazine
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
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]
2012
Multi-source, Multilingual Information Extraction and Summarization
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
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
2022
IEEE Access
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
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
Contents
2015
Procedia Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
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
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)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf.
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
2018
2018 Third International Conference on Informatics and Computing (ICIC)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
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
IGlasses
2008
Proceedings of the 10th international conference on Multimodal interfaces - IMCI '08
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2011; you can also visit the original URL.
The file type is application/pdf.
/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]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
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
2013
IEEE Transactions on Audio, Speech, and Language Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2015; you can also visit the original URL.
The file type is application/pdf.
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]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
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]
2020
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
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