Filters








42,527 Hits in 6.7 sec

Combining learn-based and lexicon-based techniques for sentiment detection without using labeled examples

Songbo Tan, Yuefen Wang, Xueqi Cheng
2008 Proceedings of the 31st annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '08  
In this work, we propose a novel scheme for sentiment classification (without labeled examples) which combines the strengths of both "learn-based" and "lexicon-based" approaches as follows: we first use  ...  a lexicon-based technique to label a portion of informative examples from given task (or domain); then learn a new supervised classifier based on these labeled ones; finally apply this classifier to the  ...  In this work, we propose a novel scheme for sentiment classification which combines the strengths of both approaches as follows: we first use unsupervised technique to label a portion of informative examples  ... 
doi:10.1145/1390334.1390481 dblp:conf/sigir/TanWC08 fatcat:6pg7ljvcvja7dmplarp7ph4tp4

Unsupervised Few-Shot Action Recognition via Action-Appearance Aligned Meta-Adaptation [article]

Jay Patravali, Gaurav Mittal, Ye Yu, Fuxin Li, Mei Chen
2021 arXiv   pre-print
Moreover, unlike previous few-shot action recognition methods that are supervised, MetaUVFS needs neither base-class labels nor a supervised pretrained backbone.  ...  MetaUVFS comprises a novel Action-Appearance Aligned Meta-adaptation (A3M) module that learns to focus on the action-oriented video features in relation to the appearance features via explicit few-shot  ...  Related Work Supervised Few-shot Learning A typical supervised few-shot learning setting has a set of base-classes with a large number of labeled samples and a set of novel classes with few labeled samples  ... 
arXiv:2109.15317v2 fatcat:36fm5sd6azhzpdcfr2hoh3lyjm

Style transfer matrix learning for writer adaptation

Xu-Yao Zhang, Cheng-Lin Liu
2011 CVPR 2011  
We combine STM learning with a specific nearest prototype classifier: learning vector quantization (LVQ) with discriminative feature extraction (DFE), where both the prototypes and the subspace transformation  ...  To adapt the basic classifier (trained with writerindependent data) to particular writers, we first propose two supervised models, one based on incremental learning and the other based on supervised STM  ...  Acknowledgements This work was supported by the National Natural Science Foundation of China (NSFC) under grants no.60825301 and no.60933010.  ... 
doi:10.1109/cvpr.2011.5995661 dblp:conf/cvpr/ZhangL11 fatcat:iszsvrvan5feba7teb33k3g65e

NEURAL NETWORK BASED DEEP LEARNING AND ENSEMBLE TECHNIQUES FOR DATA CLASSIFICATION

K. G. Nandha Kumar
2017 International Journal of Advanced Research in Computer Science  
It is found from this research that the combination of supervised learning method and stack architecture leads to better performance.  ...  The four neural networks are constructed based on deep learning and ensemble architectures. Supervised and unsupervised learning paradigms are adopted.  ...  They are biologically inspired machine learning techniques. Supervised learning and unsupervised learning are the two significant paradigms of this soft computing approach.  ... 
doi:10.26483/ijarcs.v8i7.4373 fatcat:pjst5r2lgrbx7mf7eqqz6oxzzq

Integrating dynamic stopping, transfer learning and language models in an adaptive zero-training ERP speller

Pieter-Jan Kindermans, Michael Tangermann, Klaus-Robert Müller, Benjamin Schrauwen
2014 Journal of Neural Engineering  
The time spent on calibration would allow a novel user to spell 29 symbols with our unsupervised approach. It could be of use for various clinical and non-clinical ERP-applications of BCI.  ...  A simulation study compares the proposed probabilistic zero framework (using transfer learning and task structure) to a state-of-the-art supervised model on n = 22 subjects.  ...  Table 1 contains the result on the symbol level for a subset of the unsupervised methods and baseline supervised models.  ... 
doi:10.1088/1741-2560/11/3/035005 pmid:24834896 fatcat:cjoefkibm5dnvnoppifrhjns2q

Towards Purely Unsupervised Disentanglement of Appearance and Shape for Person Images Generation [article]

Hongtao Yang, Tong Zhang, Wenbing Huang, Xuming He, Fatih Porikli
2020 arXiv   pre-print
weakly-supervised or even supervised methods.  ...  Experimental results on DeepFashion and Market1501 demonstrate that the proposed method achieves clean disentanglement and is able to synthesis novel images of comparable quality with state-of-the-art  ...  Our task is thus to learn an image decoder D I (z a a , z b s ) that combines the appearance from I a with the shape of I b to produce a novel imageÎ mix when a = b.  ... 
arXiv:2007.13098v2 fatcat:sz2s7b4iabdsninwxazyipqynm

OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in an Open World

Zhun Zhong, Linchao Zhu, Zhiming Luo, Shaozi Li, Yi Yang, Nicu Sebe
2021 Zenodo  
Existing methods typically first pre-train a model with labeled data, and then identify new classes in unlabeled data via unsupervised clustering.  ...  The challenge is that the labeled and unlabeled examples are from non-overlapping classes, which makes it difficult to build a learning relationship between them.  ...  Acknowledgements This work is supported by the EU H2020 SPRING No. 871245 and AI4Media No. 951911 projects; the Italy-China collaboration project TALENT:2018YFE0118400; and the Caritro Deep Learning Lab  ... 
doi:10.5281/zenodo.5014205 fatcat:vcoo7qusjrfhxipjtm7igm5o24

Learning Representations for Outlier Detection on a Budget [article]

Barbora Micenková, Brian McWilliams, Ira Assent
2015 arXiv   pre-print
Supervised approaches require a sufficient amount of labeled data and are challenged by novel types of outliers and inherent class imbalance, whereas unsupervised methods do not take advantage of available  ...  We propose BORE (a Bagged Outlier Representation Ensemble) which uses unsupervised outlier scoring functions (OSFs) as features in a supervised learning framework.  ...  Supervised approaches require a sufficient amount of labeled data and are challenged by novel types of outliers and inherent class imbalance, whereas unsupervised methods do not take advantage of available  ... 
arXiv:1507.08104v1 fatcat:kbrhb4vjszhhbbp2azhu5hbg74

Combining Supervised and Unsupervised Enembles for Knowledge Base Population

Nazneen Fatema Rajani, Raymond Mooney
2016 Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing  
We propose an algorithm that combines supervised and unsupervised methods to ensemble multiple systems for two popular Knowledge Base Population (KBP) tasks, Cold Start Slot Filling (CSSF) and Tri-lingual  ...  The success of our technique on two different and challenging problems demonstrates the power and generality of our combined approach to ensembling.  ...  Combining Supervised and Unsupervised We propose a novel approach to combine the aforementioned supervised and unsupervised methods using a stacked meta-classifier as the final arbiter for accepting a  ... 
doi:10.18653/v1/d16-1201 dblp:conf/emnlp/RajaniM16 fatcat:narfv26nyjblth3wvowenmu23i

PUD: Social Spammer Detection Based on PU Learning [chapter]

Yuqi Song, Min Gao, Junliang Yu, Wentao Li, Junhao Wen, Qingyu Xiong
2017 Lecture Notes in Computer Science  
Numerous notable studies have been done to detect social spammers, and these methods can be categorized into three types: unsupervised, supervised and semi-supervised methods.  ...  Numerous notable studies have been done to detect social spammers, and these methods can be categorized into three types: unsupervised, supervised and semi-supervised methods.  ...  Conclusion and Future Work In this paper, we proposed a novel method PUD based on PU Learning, it aims to construct a detection classifier by a few positive samples and plenty of unlabeled data.  ... 
doi:10.1007/978-3-319-70139-4_18 fatcat:olrn6prl2nbupgwiq4sgpijpfy

Object Based Unsupervised Classification of VHR Panchromatic and Multispectral Satellite Images by Combining the HDP, IBP and K-Mean on Multiple Scenes

Dipika R. Parate
2017 International Journal for Research in Applied Science and Engineering Technology  
The main contribution of this paper is a novel application framework to solve the problems of traditional probabilistic topic models and achieve the effective unsupervised classification of very high resolution  ...  process (HDP) and Indian buffet process (IBP) are combined on multiple scenes.  ...  The supervised method requires the availability of a training set for learning the classifier.  ... 
doi:10.22214/ijraset.2017.3022 fatcat:enhvbahfyjawnoblbwvitsglje

Improve Deep Learning with Unsupervised Objective [chapter]

Shufei Zhang, Kaizhu Huang, Rui Zhang, Amir Hussain
2017 Lecture Notes in Computer Science  
Compared with traditional methods which combine supervised and unsupervised learning, our proposed model avoids the needs for layer-wise pre-training and complicated model learning e.g. in deep autoencoder  ...  We propose a novel approach capable of embedding the unsupervised objective into hidden layers of the deep neural network (DNN) for preserving important unsupervised information.  ...  Specifically, we built a novel network with two parts: supervised part and unsupervised part.  ... 
doi:10.1007/978-3-319-70087-8_74 fatcat:k4j74arvvrachap3muexbhdqlm

Generalized clustering, supervised learning, and data assignment

Annaka Kalton, Pat Langley, Kiri Wagstaff, Jungsoon Yoo
2001 Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '01  
This framework views clustering as a general process of iterative optimization that includes modules for supervised learning and instance assignment.  ...  In this paper, we investigate experimentally the efficacy of these algorithms and test some hypotheses about the relation between such unsupervised techniques and the supervised methods embedded in them  ...  In the broader arena, there have been some efforts to link methods for supervised and unsupervised learning.  ... 
doi:10.1145/502512.502555 fatcat:bss7tia42rfxlput6tfzsux65u

A Novel Application of Unsupervised Machine Learning and Supervised Machine Learning-Derived Radiomics in Anterior Cruciate Ligament Rupture

De-Sheng Chen, Tong-Fu Wang, Jia-Wang Zhu, Bo Zhu, Zeng-Liang Wang, Jian-Gang Cao, Cai-Hong Feng, Jun-Wei Zhao
2021 Risk Management and Healthcare Policy  
The input dataset was automatically classified by the unsupervised machine learning algorithm. Then, we used a supervised machine learning algorithm to construct a radiomics model.  ...  We found that the combination of LASSO selection method and random forest classifier has the highest sensitivity, specificity, accuracy, and AUC.  ...  We found that the combination of LASSO selection method and random forest classifier has the highest sensitivity, specificity, accuracy, and AUC.  ... 
doi:10.2147/rmhp.s312330 pmid:34188576 pmcid:PMC8236276 fatcat:mipwbqez3zdm7ftu27laugyndq

An Improvement on Deep Time Growing Neural Network on Biological Signals: Review

Manpreet Kaur, Jashanpreet Kaur
2018 International Journal of Advanced Research in Computer Science and Software Engineering  
The DTGNN combines supervised and unsupervised methods in different levels of learning for an enhanced performance.  ...  A novel method for learning the cyclic contents of stochastic time series: the deep time-growing neural network (DTGNN).  ...  The DTGNN combines supervised and unsupervised methods in different levels of learning for an enhanced performance.  ... 
doi:10.23956/ijarcsse.v8i5.658 fatcat:vxlm6ovnvjgq5dutucjq7xnnja
« Previous Showing results 1 — 15 out of 42,527 results