Filters








6,190 Hits in 6.4 sec

A high-performance semi-supervised learning method for text chunking

Rie Kubota Ando, Tong Zhang
2005 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics - ACL '05  
This paper presents a novel semi-supervised method that employs a learning paradigm which we call structural learning.  ...  Although a number of semi-supervised methods have been proposed, their effectiveness on NLP tasks is not always clear.  ...  First, we present a novel robust semi-supervised method based on a new learning model and its application to chunking tasks.  ... 
doi:10.3115/1219840.1219841 dblp:conf/acl/AndoZ05 fatcat:b222o7nponbn3dqifbdci5ng2i

Unsupervised acquisition of idiomatic units of symbolic natural language: An n-gram frequency-based approach for the chunking of news articles and tweets

Dario Borrelli, Gabriela Gongora Svartzman, Carlo Lipizzi, Christopher M. Danforth
2020 PLoS ONE  
Existing methods are based primarily on supervised and semi-supervised learning approaches; however, in this study, a novel unsupervised approach is proposed based on the existing concept of n-grams, which  ...  In this paper, an unsupervised approach for the chunking of idiomatic units of sequential text data is presented.  ...  It is still used as a benchmark corpus, and a high accuracy has been obtained by applying different supervised and semi-supervised learning methods [31, 32] .  ... 
doi:10.1371/journal.pone.0234214 pmid:32511252 fatcat:vztyjlnkuzer3jvtxgs2rz3psm

Machine learning classification of surgical pathology reports and chunk recognition for information extraction noise reduction

Giulio Napolitano, Adele Marshall, Peter Hamilton, Anna T. Gavin
2016 Artificial Intelligence in Medicine  
., Machine learning classification of surgical pathology reports and chunk recognition for information extraction noise reduction, Abstract Background and aims: Machine learning techniques for the text  ...  For semi-structured reports the performance ranged from 0.97 to 0.94 and from 0.92 to 0.83 in precision and recall, while for unstructured reports performance ranged from 0.91 to 0.64 and from 0.68 to  ...  the supervised learning was manual.  ... 
doi:10.1016/j.artmed.2016.06.001 pmid:27431038 fatcat:xub3mz7chraczjsemck4by6eeu

A Clustering-based Framework for Classifying Data Streams [article]

Xuyang Yan, Abdollah Homaifar, Mrinmoy Sarkar, Abenezer Girma, Edward Tunstel
2021 arXiv   pre-print
Experimental results and quantitative comparison studies reveal that the proposed method provides statistically better or comparable performance than the existing methods.  ...  A density-based stream clustering procedure is used to capture novel concepts with a dynamic threshold and an effective active label querying strategy is introduced to continuously learn the new concepts  ...  The authors would like to thank them for their support.  ... 
arXiv:2106.11823v1 fatcat:56fcfm2ohzbwzn4pmfgomcs6bm

Learning Speaker Representations with Mutual Information [article]

Mirco Ravanelli, Yoshua Bengio
2019 arXiv   pre-print
Our experiments consider both unsupervised and semi-supervised settings and compare the performance achieved with different objective functions.  ...  Mutual Information (MI) or similar measures of statistical dependence are promising tools for learning these representations in an unsupervised way.  ...  Beyond unsupervised learning, this paper explores two semi-supervised variations for learning speaker representations.  ... 
arXiv:1812.00271v2 fatcat:s4m2ziaedjdqpdgubg25ht7cwe

Opinion Stream Mining [chapter]

Myra Spiliopoulou, Eirini Ntoutsi, Max Zimmermann
2016 Encyclopedia of Machine Learning and Data Mining  
In this chapter, we overview methods for polarity learning in a stream environment focusing especially on how these methods deal with the challenges imposed by the stream nature of the data, namely the  ...  Opinion stream mining aims at learning and adaptation of a polarity model over a stream of opinionated documents, i.e., documents associated with a polarity.  ...  Next to fully supervised solutions, there are also semi-supervised learning methods and active learning methods, in response to challenge (d).  ... 
doi:10.1007/978-1-4899-7502-7_905-1 fatcat:d2fgtvgjzbhz7pkmt44amrfpmy

Chinese Chunking with Tri-training Learning [chapter]

Wenliang Chen, Yujie Zhang, Hitoshi Isahara
2006 Lecture Notes in Computer Science  
This paper presents a practical tri-training method for Chinese chunking using a small amount of labeled training data and a much larger pool of unlabeled data.  ...  We propose a novel selection method for tri-training learning in which newly labeled sentences are selected by comparing the agreements of three classifiers.  ...  Ando and Zhang [14] proposed a semi-supervised learning method that employs the structural learning for English chunking. Steedman et al.  ... 
doi:10.1007/11940098_49 fatcat:no4hdsc44jc3lm7qmeabuaquea

Learning Speaker Representations with Mutual Information

Mirco Ravanelli, Yoshua Bengio
2019 Interspeech 2019  
Our experiments consider both unsupervised and semi-supervised settings and compare the performance achieved with different objective functions.  ...  Mutual Information (MI) or similar measures of statistical dependence are promising tools for learning these representations in an unsupervised way.  ...  Beyond unsupervised learning, this paper explores two semi-supervised variations for learning speaker representations.  ... 
doi:10.21437/interspeech.2019-2380 dblp:conf/interspeech/RavanelliB19 fatcat:c7q4ijjzbrg2pmvsl5ajndzmtu

Recognizing Named Entities in Tweets

Xiaohua Liu, Shaodian Zhang, Furu Wei, Ming Zhou
2011 Annual Meeting of the Association for Computational Linguistics  
The semi-supervised learning plus the gazetteers alleviate the lack of training data.  ...  We propose to combine a K-Nearest Neighbors (KNN) classifier with a linear Conditional Random Fields (CRF) model under a semi-supervised learning framework to tackle these challenges.  ...  Acknowledgments We thank Long Jiang, Changning Huang, Yunbo Cao, Dongdong Zhang, Zaiqing Nie for helpful discussions, and the anonymous reviewers for their valuable comments.  ... 
dblp:conf/acl/LiuZWZ11 fatcat:ekp4qwci25cypbgnpakjrrcxmu

KEC@DPIL-FIRE2016: Detection of Paraphrases in Indian Languages (Tamil)

R. Thangarajan, S. V. Kogilavani, A. Karthic, S. Jawahar
2016 Forum for Information Retrieval Evaluation  
They have been classified into Paraphrase and Not-a-Paraphrase for task1 and Paraphrase, Not-a-Paraphrase and Semi-Paraphrase for task2.  ...  The accuracy and performance of these methods are measured on the basis of evaluation parameters like accuracy, precision, recall, f-measure and macro f-measure.  ...  Paraphrase(P), Not-a-Paraphrase(NP) and Semi-Paraphrase (SP) for task2.  ... 
dblp:conf/fire/ThangarajanKKJ16 fatcat:zlhrcmqzlrd2jhcj45hiif6ad4

Supervised Contrastive Learning for Interpretable Long-Form Document Matching [article]

Akshita Jha, Vineeth Rakesh, Jaideep Chandrashekar, Adithya Samavedhi, Chandan K. Reddy
2022 arXiv   pre-print
CoLDE uses unique positional embeddings and a multi-headed chunkwise attention layer in conjunction with a supervised contrastive learning framework to capture similarity at three different levels: (i)  ...  high-level similarity scores between a pair of documents, (ii) similarity scores between different sections within and across documents, and (iii) similarity scores between different chunks in the same  ...  Figure 8 shows the model performance for all the baseline methods.  ... 
arXiv:2108.09190v2 fatcat:qaylpfkgd5davcah6qrc2hfelq

On the Learning Dynamics of Semi-Supervised Training for ASR

Electra Wallington, Benji Kershenbaum, Ondrej Klejch, Peter Bell
2021 Conference of the International Speech Communication Association  
The use of semi-supervised training (SST) has become an increasingly popular way of increasing the performance of ASR acoustic models without the need for further transcribed speech data.  ...  This paper undertakes a comprehensive study of the improvements gained with respect to variation in the initial systems, the quantity of untranscribed data used, and the learning schedules.  ...  final learning rate for the i-th chunk is equal to lri+1.  ... 
doi:10.21437/interspeech.2021-1777 dblp:conf/interspeech/WallingtonKK021 fatcat:jypacbj6bzfgzpkdosbht74nba

Semi-supervised Music Tagging Transformer

Minz Won, Keunwoo Choi, Xavier Serra
2021 Zenodo  
We present Music Tagging Transformer that is trained with a semi-supervised approach.  ...  The Music Tagging Transformer is further improved by noisy student training, a semi-supervised approach that leverages both labeled and unlabeled data combined with data augmentation.  ...  music tagging performances via semi-supervised learning, (iii) we provide a new split of the MSD to solve known issues of the previous one.  ... 
doi:10.5281/zenodo.5624405 fatcat:jxi3c6edbrelplsmbayqjo6e24

Segment-Level Sequence Modeling using Gated Recursive Semi-Markov Conditional Random Fields

Jingwei Zhuo, Yong Cao, Jun Zhu, Bo Zhang, Zaiqing Nie
2016 Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
Semi-Markov Conditional Random Fields (Semi-CRFs) model segments directly but extracting segment-level features for Semi-CRFs is still a very challenging problem.  ...  Our experiments on text chunking and named entity recognition (NER) demonstrate that grSemi-CRFs generally outperform other neural models.  ...  For text chunking, grSemi-CRF outperforms all reported supervised models, except JESS-CM (Suzuki and Isozaki, 2008) , a semi-supervised model using giga-word scale unlabeled data in training 14 .  ... 
doi:10.18653/v1/p16-1134 dblp:conf/acl/ZhuoCZZN16 fatcat:mhzmy24zuzbzbkh2efew6xoy4a

Phone-to-audio alignment without text: A Semi-supervised Approach [article]

Jian Zhu, Cong Zhang, David Jurgens
2022 arXiv   pre-print
The proposed Wav2Vec2-FS, a semi-supervised model, directly learns phone-to-audio alignment through contrastive learning and a forward sum loss, and can be coupled with a pretrained phone recognizer to  ...  Here we introduce two Wav2Vec2-based models for both text-dependent and text-independent phone-to-audio alignment.  ...  First, we present a semi-supervised model to perform forced alignment, but can also be combined with a phone recognizer for text-independent alignment.  ... 
arXiv:2110.03876v2 fatcat:kbpiirclyrayrn7wkcv4h2agfe
« Previous Showing results 1 — 15 out of 6,190 results