Filters








17,389 Hits in 5.7 sec

Task-oriented Domain-specific Meta-Embedding for Text Classification

Xin Wu, Yi Cai, Yang Kai, Tao Wang, Qing Li
2020 Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)   unpublished
Moreover, the importance of general and domain word embeddings is related to downstream tasks, how to regularize meta-embedding to adapt downstream tasks is an unsolved problem.  ...  In this paper, we propose a method to incorporate both domain-specific and taskoriented information into meta-embeddings.  ...  Acknowledgement This work was supported by the Fundamental Research Funds for the Central Universities, SCUT (No.2017ZD048, D2182480), the Science and Technology Planning Project of Guangdong Province  ... 
doi:10.18653/v1/2020.emnlp-main.282 fatcat:ufrtdmftnzei3k5kst3gg6ns4q

ProtoDA: Efficient Transfer Learning for Few-Shot Intent Classification [article]

Manoj Kumar, Varun Kumar, Hadrien Glaude, Cyprien delichy, Aman Alok, Rahul Gupta
2021 arXiv   pre-print
Practical sequence classification tasks in natural language processing often suffer from low training data availability for target classes.  ...  Recent works towards mitigating this problem have focused on transfer learning using embeddings pre-trained on often unrelated tasks, for instance, language modeling.  ...  DATASETS We use two source corpora: the task-oriented dialog corpus from Facebook (FB) [26] containing crowd-sourced annotations for queries from the navigation and event management domains, and the  ... 
arXiv:2101.11753v1 fatcat:fmyl7vidyrdutlo2uumbhebjsq

Few-shot learning for medical text: A systematic review [article]

Yao Ge, Yuting Guo, Yuan-Chi Yang, Mohammed Ali Al-Garadi, Abeed Sarker
2022 arXiv   pre-print
Concept extraction/named entity recognition was the most frequently addressed task (13/31; 42%), followed by text classification (10/31; 32%).  ...  Discussion: Despite the potential for FSL in biomedical NLP, progress has been limited compared to domain-independent FSL.  ...  Specifically, we review FSL methods for medical NLP tasks, and characterize each reviewed article in terms of type of task (eg., text classification, NER), primary aim(s), dataset(s), evaluation metrics  ... 
arXiv:2204.14081v1 fatcat:ageqcud25fh3xeuctrgeqytmhe

A Review of Text Corpus-Based Tourism Big Data Mining

Qin Li, Shaobo Li, Sen Zhang, Jie Hu, Jianjun Hu
2019 Applied Sciences  
We summarize and discuss different text representation strategies, text-based NLP techniques for topic extraction, text classification, sentiment analysis, and text clustering in the context of tourism  ...  As an effective expression means of tourists' opinions, text mining of such data has big potential to inspire innovations for tourism practitioners.  ...  of metrics from the meta-training task [161] .  ... 
doi:10.3390/app9163300 fatcat:chb3pbtj5jgq7fauniomsb22yu

Low-resource Learning with Knowledge Graphs: A Comprehensive Survey [article]

Jiaoyan Chen and Yuxia Geng and Zhuo Chen and Jeff Z. Pan and Yuan He and Wen Zhang and Ian Horrocks and Huajun Chen
2022 arXiv   pre-print
knowledge extraction), but also tasks for KG curation (e.g., inductive KG completion), and some typical evaluation resources for each task.  ...  We next presented different applications, including not only KG augmented prediction tasks in Computer Vision and Natural Language Processing (e.g., image classification, visual question answering and  ...  ACKNOWLEDGMENTS This work was supported by the SIRIUS Centre for Scalable Data Access (Research Council of Norway, project 237889), eBay, Samsung Research UK, Siemens AG, and the EPSRC projects OASIS (  ... 
arXiv:2112.10006v5 fatcat:vxl5hnqe5jaafgwuznbz556fmm

A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities [article]

Yisheng Song, Ting Wang, Subrota K Mondal, Jyoti Prakash Sahoo
2022 arXiv   pre-print
For the sake of avoiding conceptual confusion, we first elaborate and compare a set of similar concepts including few-shot learning, transfer learning, and meta-learning.  ...  Despite the recent creative works in tackling FSL tasks, learning valid information rapidly from just a few or even zero samples still remains a serious challenge.  ...  As a general learning framework, meta-learning is independent of specific problems and more oriented to future tasks instead of optimizing the current one.  ... 
arXiv:2205.06743v2 fatcat:xmxht2ileja53o2o5b4vrw32ey

Challenge Closed-book Science Exam: A Meta-learning Based Question Answering System [article]

Xinyue Zheng, Peng Wang, Qigang Wang, Zhongchao Shi
2020 arXiv   pre-print
We evaluate our method on AI2 Reasoning Challenge (ARC), and the experimental results show that meta-classifier yields considerable classification performance on emerging question types.  ...  Specifically, our method based on meta-learning method and large language model BERT, which can efficiently solve science problems by learning from related example questions without relying on external  ...  q t , l t ) query samples , f θ } (1) For each task, specific supporting samples and query samples are formed to train a task-specific classification modelf θ by minimizing the loss function L(f θ (q t  ... 
arXiv:2004.12303v1 fatcat:5xzmebvh2vbtpi2z7ogfkvmhau

ProtoInfoMax: Prototypical Networks with Mutual Information Maximization for Out-of-Domain Detection [article]

Iftitahu Ni'mah, Meng Fang, Vlado Menkovski, Mykola Pechenizkiy
2021 arXiv   pre-print
Experimental results show that our proposed method can substantially improve performance up to 20% for OOD detection in low resource settings of text classification.  ...  The ability to detect Out-of-Domain (OOD) inputs has been a critical requirement in many real-world NLP applications. For example, intent classification in dialogue systems.  ...  Acknowledgment This research is supported by Indonesian Endowment Fund for Education (LPDP) Scholarship under Beasiswa Pendidikan Indonesia (BPI) -ID Number 0003194/SC/D/9/LPDP2016.  ... 
arXiv:2108.12229v5 fatcat:ykcw6gjlabfvpb5lr5qeu732xa

Cross-Domain Deep Code Search with Few-Shot Meta Learning [article]

Yitian Chai, Hongyu Zhang, Beijun Shen, Xiaodong Gu
2022 arXiv   pre-print
This restricts their practicality in domain-specific languages with relatively scarce and expensive data. In this paper, we propose CDCS, a novel approach for domain-specific code search.  ...  can be best reused in a domain-specific language.  ...  Solidity is an object-oriented language that is specifically designed for smart contracts [36] .  ... 
arXiv:2201.00150v2 fatcat:xfci4rxv35ggvhrnwfycp6en3y

Capturing Contextual Factors in Sentiment Classification: An Ensemble Approach

Thien Khai Tran, Tuoi Thi Phan
2020 IEEE Access  
We found that the combination of word embedding representation and the attention mechanism, along with pre-defined rules and specific-domain sentiment dictionaries are helpful in dealing with numerous  ...  In this paper, we propose an effective ensemble learning model for the sentiment classification problem.  ...  ACKNOWLEDGMENT The authors would like to thank Ho Chi Minh City University of Technology (HCMUT), and VNU-HCM for their support.  ... 
doi:10.1109/access.2020.3004180 fatcat:ro74lbephjgkfj3llir3lqz7y4

Transforming task representations to perform novel tasks [article]

Andrew K. Lampinen, James L. McClelland
2020 arXiv   pre-print
Across these domains, meta-mapping is successful, often achieving 80-90% performance, without any data, on a novel task, even when the new task directly contradicts prior experience.  ...  By contrast, models that achieve superhuman performance in specific tasks often fail to adapt to even slight task alterations.  ...  See section B for the specific classifications that were used in each domain.  ... 
arXiv:2005.04318v3 fatcat:taydgkl5hvgude3rf7vpcvluke

ProtAugment: Unsupervised diverse short-texts paraphrasing for intent detection meta-learning [article]

Thomas Dopierre, Christophe Gravier, Wilfried Logerais
2021 arXiv   pre-print
In this work, we propose ProtAugment, a meta-learning algorithm for short texts classification (the intent detection task).  ...  ProtAugment is the state-of-the-art method for intent detection meta-learning, at no extra labeling efforts and without the need to fine-tune a conditional language model on a given application domain.  ...  We also would like to thank ANRT 5 for making partnerships between companies and universities happen.  ... 
arXiv:2105.12995v1 fatcat:yrverod7uzgldpq6vzqvw2as6y

Teaching Domain-Specific Language Engineering and Model-Driven Software Development: A Competence-oriented Approach

Volkhard Pfeiffer
2016 ACM/IEEE International Conference on Model Driven Engineering Languages and Systems  
Teaching and learning domain-specific language (DSL) engineering and model-driven software development (MDSD) concepts are difficult tasks: either it requires a deep understanding of the nature of a domain  ...  This paper explains a competence-oriented approach for model-driven software development course design to reduce the above learning difficulties.  ...  Technical Learning Outcomes The course covers MDSD terminology, meta-modeling, model-to-model and modelto-text transformations, internal and external domain specific languages and model validation.  ... 
dblp:conf/models/Pfeiffer16 fatcat:zynd3xcmkre3xggjlohw5dusae

JuriBERT: A Masked-Language Model Adaptation for French Legal Text [article]

Stella Douka, Hadi Abdine, Michalis Vazirgiannis, Rajaa El Hamdani, David Restrepo Amariles
2022 arXiv   pre-print
We explore the use of smaller architectures in domain-specific sub-languages and their benefits for French legal text.  ...  We conclude that some specific tasks do not benefit from generic language models pre-trained on large amounts of data.  ...  Related Work Previous work on domain-specific text data has indicated the importance of creating domain-specific language models.  ... 
arXiv:2110.01485v2 fatcat:dwbxhd5acnfgli5dgittlspyzu

Multi-Aspect Oriented Sentiment Classification: Prior Knowledge Topic Modelling and Ensemble Learning Classifier Approach

Najwa AlGhamdi, Shaheen Khatoon, Majed Alshamari
2022 Applied Sciences  
Therefore, this paper proposes an aspect-oriented sentiment classification using a combination of the prior knowledge topic model algorithm (SA-LDA), automatic labelling (SentiWordNet) and ensemble method  ...  The study concluded that the proposed approach is equally adaptable across multi-domain applications.  ...  User-generated content usually contains unstructured text that is used in classification tasks such as information extraction (IE), text analysis and natural language processing (NLP).  ... 
doi:10.3390/app12084066 fatcat:yjtn3523yfetjhkxzdww3wbd5u
« Previous Showing results 1 — 15 out of 17,389 results