Filters








513 Hits in 5.7 sec

Discovering Connotations as Labels for Weakly Supervised Image-Sentence Data

Aditya Mogadala, Bhargav Kanuparthi, Achim Rettinger, York Sure-Vetter
2018 Companion of the The Web Conference 2018 on The Web Conference 2018 - WWW '18  
(iii) works effectively with large-scale weakly supervised data.  ...  In this paper, we aim to annotate such image-sentence pairs with connotations as labels to capture the intrinsic "intension".  ...  CONCLUSION AND FUTURE WORK In this paper, we presented an approach to automatically extract connotations as labels for images by leveraging weakly supervised image-tweet data.  ... 
doi:10.1145/3184558.3186352 dblp:conf/www/MogadalaKRS18 fatcat:7tvmbfqrv5cg3h4ccj7pahszqe

Learning Everything about Anything: Webly-Supervised Visual Concept Learning

Santosh K. Divvala, Ali Farhadi, Carlos Guestrin
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
Our approach leverages vast resources of online books to discover the vocabulary of variance, and intertwines the data collection and modeling steps to alleviate the need for explicit human supervision  ...  How can we learn a model for any concept that exhaustively covers all its appearance variations, while requiring minimal or no human supervision for compiling the vocabulary of visual variance, gathering  ...  We thank Neeraj Kumar for his helpful comments.  ... 
doi:10.1109/cvpr.2014.412 dblp:conf/cvpr/DivvalaFG14 fatcat:py5e4bza6ndyfnkcxrd5357cfu

Deep Learning for Sentiment Analysis : A Survey [article]

Lei Zhang, Shuai Wang, Bing Liu
2018 arXiv   pre-print
Deep learning has emerged as a powerful machine learning technique that learns multiple layers of representations or features of the data and produces state-of-the-art prediction results.  ...  It contains a two-step learning process: it first learns a sentence representation weakly supervised by overall review ratings and then uses the sentence (and aspect) level labels for finetuning.  ...  Guan et al. 62 employed a weakly-supervised CNN for sentence (and also aspect) level sentiment classification.  ... 
arXiv:1801.07883v2 fatcat:nplicfgaozb6fbfx4eyts4zt7e

Opinion Mining and Sentiment Analysis [chapter]

Bing Liu
2011 Web Data Mining  
text, and readers of on-line text -from consumers to sports fans to news addicts to governments -can benefit from automatic methods that synthesise useful opinion-orientated information from the sea of data  ...  One issue for supervised approaches to opinion expression identification is that they require training data; and unfortunately, such data is not as easy to come by for this task as it is for, say, sentiment  ...  lexicon, then the sentence is deemed subjective; otherwise, it is labelled as objective.  ... 
doi:10.1007/978-3-642-19460-3_11 fatcat:5epiy3et3jed5n24ejzyeavpby

CATs are Fuzzy PETs: A Corpus and Analysis of Potentially Euphemistic Terms [article]

Martha Gavidia, Patrick Lee, Anna Feldman, Jing Peng
2022 arXiv   pre-print
Secondly, we observe cases of disagreement in an annotation task, where humans are asked to label PETs as euphemistic or not in a subset of our corpus text examples.  ...  Additionally, we present a subcorpus of texts where these PETs are not being used euphemistically, which may be useful for future applications.  ...  Acknowledgements We thank our annotators Raz Besaleli, Kira Horiuchi, Kelly Ortega, and Kenna Reagan for their time and attention to our corpus annotation task as well as Brad McNamee and Avery Field for  ... 
arXiv:2205.02728v1 fatcat:hyywpltxovc5fedknqthkllicq

Sentiment Analysis and Opinion Mining [chapter]

Lei Zhang, Bing Liu
2017 Encyclopedia of Machine Learning and Data Mining  
This method is unsupervised as it does not use any manually labeled data for training. Clearly, with the labeled data supervised learning can be applied as well.  ...  The most dominant methods are based on sequential learning (or sequential labeling). Since these are supervised techniques, they need manually labeled data for training.  ... 
doi:10.1007/978-1-4899-7687-1_907 fatcat:iy5ty44cyzbrtodxfo7osy3iu4

Multi-Dimensional Gender Bias Classification [article]

Emily Dinan, Angela Fan, Ledell Wu, Jason Weston, Douwe Kiela, Adina Williams
2020 arXiv   pre-print
We show our classifiers prove valuable for a variety of important applications, such as controlling for gender bias in generative models, detecting gender bias in arbitrary text, and shed light on offensive  ...  Machine learning models are trained to find patterns in data. NLP models can inadvertently learn socially undesirable patterns when training on gender biased text.  ...  However, this weakly supervised data provides somewhat noisy training signal -particularly for the masculine and feminine classes -as the labels are automatically annotated or inferred.  ... 
arXiv:2005.00614v1 fatcat:o3lgzjeouvhepmp6bkmw2jk7jm

Word Sense Disambiguation in Hindi Language Using Score Based Modified Lesk Algorithm

Praffullit Tripathi, Prasenjit Mukherjee, Manik Hendre, Manish Godse, Baisakhi Chakraborty
2021 International Journal of Computing and Digital Systems  
After a data set generation, we implement supervised, unsupervised, and knowledge-based the Lesk algorithm for further processing of sentences. techniques and that can be utilized to improve  ...  The sample data from SENSEVAL-2 has been utilized for method evaluation and this system has achieved 32% overall accuracy as in [33].  ... 
doi:10.12785/ijcds/100185 fatcat:35aaf3v2wnggrnfb6st2goesa4

Knowledge-rich image gist understanding beyond literal meaning

Lydia Weiland, Ioana Hulpuş, Simone Paolo Ponzetto, Wolfgang Effelsberg, Laura Dietz
2018 Data & Knowledge Engineering  
We investigate the problem of understanding the message (gist) conveyed by images and their captions as found, for instance, on websites or news articles.  ...  the identification of connotations, i.e., iconic meanings of objects, to understand the message of images.  ...  Acknowledgement We gratefully acknowledge the support of NVIDIA Corporation with the donation of the GeForce Titan X GPU used for this research.  ... 
doi:10.1016/j.datak.2018.07.006 fatcat:2t73wjr3nnci3dz6xxar3c3boi

Opinion Mining and Sentiment Analysis [chapter]

2016 Text Data Management and Analysis: A Practical Introduction to Information Retrieval and Text Mining  
With the growing availability and popularity of opinion-rich resources such as online review sites and personal blogs, new opportunities and challenges arise as people now can, and do, actively use information  ...  sacred as human . • Persuasion suggests a belief grounded on assurance (as by evidence) of its truth was of the persuasion that everything changes . • Sentiment suggests a settled opinion reflective of  ...  sentence [130] , or to incorporate information drawn from some labeled data as well [33] .  ... 
doi:10.1145/2915031.2915050 fatcat:3ypahumv4rhz5alxmh34ikvm3i

Open-Domain Question–Answering

John Prager
2006 Foundations and Trends in Information Retrieval  
With the growing availability and popularity of opinion-rich resources such as online review sites and personal blogs, new opportunities and challenges arise as people now can, and do, actively use information  ...  sacred as human . • Persuasion suggests a belief grounded on assurance (as by evidence) of its truth was of the persuasion that everything changes . • Sentiment suggests a settled opinion reflective of  ...  sentence [130] , or to incorporate information drawn from some labeled data as well [33] .  ... 
doi:10.1561/1500000001 fatcat:5xq5eb7idrb37lz3wyzfl4cb34

Design and Evaluation of Metaphor Processing Systems

Ekaterina Shutova
2015 Computational Linguistics  
A wide range of methods have been proposed and investigated by the community, including supervised (Gedigian et 2 Shutova ).  ...  This hampers our progress as a community in this area.  ...  Supervised classification A number of approaches trained classifiers on manually annotated data to recognise metaphor.  ... 
doi:10.1162/coli_a_00233 fatcat:xndjednwznb6ldr54xh6r4u6wu

Linguistic Reflexes of Well-Being and Happiness in Echo

Jiaqi Wu, Marilyn Walker, Pranav Anand, Steve Whittaker
2017 Proceedings of the 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis  
Different theories posit different sources for feelings of well-being and happiness.  ...  We show that recurrent event types, such as OBLIGATION and IN-COMPETENCE, which affect people's feelings of well-being are not captured in current lexical or semantic resources.  ...  Linguistic Pattern Learning We also apply Autoslog-TS, a weakly supervised linguistic-pattern learner as a way of learning some compositional patterns.  ... 
doi:10.18653/v1/w17-5211 dblp:conf/wassa/WuWAW17 fatcat:vgr5dwfabjg77cnpbeyh7yzmzu

Exploring the Effect of Spreading Fake News Debunking Based on Social Relationship Networks

Xin Wang, Fan Chao, Ning Ma, Guang Yu
2022 Frontiers in Physics  
developing effective strategies for debunking fake news.  ...  Additionally, more celebrity accounts, larger node sizes with follower-followee relationships in the SRNs, and more weakly connected components were found to lead to a faster growth rate in the dissemination  ...  The authors would like to thank Zhiwei Data Sharing Platform (http://university.zhiweidata.com/) for our data support.  ... 
doi:10.3389/fphy.2022.833385 fatcat:w3vlrav7mndqpfbxo3n57qude4

Multilingual Stance Detection in Social Media Political Debates

Mirko Lai, Alessandra Teresa Cignarella, Delia Irazú Hernández Farías, Cristina Bosco, Viviana Patti, Paolo Rosso
2020 Computer Speech and Language  
First of all, a set of resources on topics related to politics for English, French, Italian, Spanish and Catalan is provided which includes: novel corpora collected for the purpose of this study, and benchmark  ...  analysis of the misclassified tweets for each of the observed languages, devoted to reflect on the open challenges.  ...  Weakly Supervised Framework. Unlike in Task A, for this task only data for testing were released, and they were only about one target, "Donald Trump".  ... 
doi:10.1016/j.csl.2020.101075 fatcat:j3cwbbnairatpb4czoqccvuehi
« Previous Showing results 1 — 15 out of 513 results