944 Hits in 4.2 sec

Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing [article]

Sarah Wiegreffe, Ana Marasović
2021 arXiv   pre-print
and shortcomings of existing collection methodologies, and give recommendations for collecting ExNLP datasets in the future.  ...  Explainable NLP (ExNLP) has increasingly focused on collecting human-annotated textual explanations.  ...  Acknowledgements We are grateful to Yejin Choi, Peter Clark, Gabriel Ilharco, Alon Jacovi, Daniel Khashabi, Mark Riedl, Alexis Ross, and Noah Smith for valuable feedback.  ... 
arXiv:2102.12060v4 fatcat:ufmg2rdexbbhlnqborhhsivy6u

Explaining NLP Models via Minimal Contrastive Editing (MiCE) [article]

Alexis Ross, Ana Marasović, Matthew E. Peters
2021 arXiv   pre-print
We present Minimal Contrastive Editing (MiCE), a method for producing contrastive explanations of model predictions in the form of edits to inputs that change model outputs to the contrast case.  ...  Despite the influential role that contrastivity plays in how humans explain, this property is largely missing from current methods for explaining NLP models.  ...  Then, we turn to a key motivation for this work: the potential for contrastive explanations to assist in NLP system development.  ... 
arXiv:2012.13985v2 fatcat:v2f7ll62a5d7vkkyp4vgrzlmuq

WT5?! Training Text-to-Text Models to Explain their Predictions [article]

Sharan Narang, Colin Raffel, Katherine Lee, Adam Roberts, Noah Fiedel, Karishma Malkan
2020 arXiv   pre-print
We show that this approach not only obtains state-of-the-art results on explainability benchmarks, but also permits learning from a limited set of labeled explanations and transferring rationalization  ...  abilities across datasets.  ...  In addition to producing state-of-the-art results on explainability datasets, this approach also allows for both "semi-supervised" training (where explanations are only provided on a subset of the dataset  ... 
arXiv:2004.14546v1 fatcat:hfothpcngzbixnhgbcxhri2pvy

Explainable AI for B5G/6G: Technical Aspects, Use Cases, and Research Challenges [article]

Shen Wang, M.Atif Qureshi, Luis Miralles-Pechuaán, Thien Huynh-The, Thippa Reddy Gadekallu, Madhusanka Liyanage
2021 arXiv   pre-print
Such a 6G network will lead to an excessive number of automated decisions made every second.  ...  The promising explainable AI (XAI) methods can mitigate such risks by enhancing the transparency of the black box AI decision-making process.  ...  The goal of pre-modelling explainability is to family, how an explanation is going to be presented to a user; describe the dataset to gain better insights into the dataset  ... 
arXiv:2112.04698v1 fatcat:y7ss4opmrjbsbjm3ip2vgkkgky

Lessons Learned from Teaching Machine Learning and Natural Language Processing to High School Students

Narges Norouzi, Snigdha Chaturvedi, Matthew Rutledge
This paper describes an experience in teaching Machine Learning (ML) and Natural Language Processing (NLP) to a group of high school students over an intense one-month period.  ...  These measures include employing a combination of objectivist and constructivist pedagogies, reviewing/introducing basic programming concepts at the beginning of the course, and addressing gender discrepancies  ...  Conclusion In this work, we shared the curriculum we designed to teach ML and NLP to a group of high school students over a onemonth period.  ... 
doi:10.1609/aaai.v34i09.7063 fatcat:ee7lz2qre5f55olwqnmzgbvu5a

Aspects Based Opinion Mining for Teacher and Course Evaluation

2019 Sukkur IBA Journal of Computing and Mathematical Sciences  
Using, this two-step strategy combining with NLP, machine learning techniques and data from past seven years of real feedback at a public sector university in Pakistan, we are able to achieve a recall  ...  To the best of our knowledge, this is the first rule-based approach for such problem with quite satisfactory results.  ...  applied this approach on movie reviews dataset and got overall accuracy of 84%.  ... 
doi:10.30537/sjcms.v3i1.375 fatcat:3fm5up2webc4di4acpyt3z73m4

Teaching NLP outside Linguistics and Computer Science classrooms: Some challenges and some opportunities [article]

Sowmya Vajjala
2021 arXiv   pre-print
We see people using NLP methods in a range of academic disciplines from Asian Studies to Clinical Oncology.  ...  This paper takes a closer look at some issues related to teaching NLP to these diverse audiences based on my classroom experiences, and identifies some challenges the instructors face, particularly when  ...  ) that gave me the opportunities to teach them.  ... 
arXiv:2105.00895v1 fatcat:aljjaqglbferfb7x35sf5td3eu

Towards Process-Oriented, Modular, and Versatile Question Generation that Meets Educational Needs [article]

Xu Wang, Simin Fan, Jessica Houghton, Lu Wang
2022 arXiv   pre-print
We argue that building effective human-NLP collaborative QG systems that emphasize instructor control and explainability is imperative for real-world adoption.  ...  We call for QG systems to provide process-oriented support, use modular design, and handle diverse sources of input.  ...  We thank all our participants for their time. We thank the anonymous reviewers for their valuable suggestions on various aspects of this work.  ... 
arXiv:2205.00355v1 fatcat:4twiro2lkjbehcyvtyvjcjl7r4

Improving Classification through Weak Supervision in Context-specific Conversational Agent Development for Teacher Education [article]

Debajyoti Datta, Maria Phillips, Jennifer Chiu, Ginger S. Watson, James P. Bywater, Laura Barnes, Donald Brown
2020 arXiv   pre-print
Machine learning techniques applied to the Natural Language Processing (NLP) component of conversational agent development show promising results for improved accuracy and quality of feedback that a conversational  ...  We demonstrate the validity of this method on the Google Jigsaw data set and then propose a scenario to apply this method using the Instructional Quality Assessment(IQA) to define the categories for labeling  ...  Acknowledgments Want to acknowledge Labelbox for providing us an academic license. This work was supported in part by an Anonymous Grant Foundation.  ... 
arXiv:2010.12710v1 fatcat:y4uebbnbuvdrrd5t5bi3kaq2eu

Summarize-then-Answer: Generating Concise Explanations for Multi-hop Reading Comprehension [article]

Naoya Inoue, Harsh Trivedi, Steven Sinha, Niranjan Balasubramanian, Kentaro Inui
2021 arXiv   pre-print
Instead, we advocate for an abstractive approach, where we propose to generate a question-focused, abstractive summary of input paragraphs and then feed it to an RC system.  ...  Given a limited amount of human-annotated abstractive explanations, we train the abstractive explainer in a semi-supervised manner, where we start from the supervised model and then train it further through  ...  We thank the anonymous reviewers for the insightful feedback.  ... 
arXiv:2109.06853v1 fatcat:mdiqqhpitjesrap3ilgwld5rnu

Comparing the Performance of NLP Toolkits and Evaluation measures in Legal Tech [article]

Muhammad Zohaib Khan
2021 arXiv   pre-print
Recent developments in Natural Language Processing have led to the introduction of state-of-the-art Neural Language Models, enabled with unsupervised transferable learning, using different pretraining  ...  We use domain-specific pretraining and additional legal vocabulary to adapt BERT Model further to the Legal Domain.  ...  Currently, NLP is playing a vital role in the Legal Technology in quite a few areas, including extracting relevant information about judicial decisions, automated review of contracts for error checking  ... 
arXiv:2103.11792v1 fatcat:iivdeljmvzbwpnoxz4ycm7f7li

Survey on Sentiment Analysis Using Machine Learning

Parth Deshmukh, Adesh Gadge, Aniket Ganbote, Swapnali Garud, Prof. D. S. Kulkarni
2019 Zenodo  
SA plays important role in our day to day life while taking decisions about buying online products & for movie reviews.  ...  Sentiment Analysis (SA) or Opinion Mining (OM) is the study of people's opinions, monitoring social media & other online resources for customer reviews to understand customer understanding of significant  ...  ME requires more time to train datasets as compared to Naive Bayes [6] .  ... 
doi:10.5281/zenodo.2614704 fatcat:7apwg2j4rbhw5e6v7itf24fdme

Conversational artificial intelligence - demystifying statistical vs linguistic NLP solutions

Kulvinder Panesar
2020 Journal of Computer-Assisted Linguistic Research  
To demonstrate this, a deep linguistically aware and knowledge aware text based conversational agent (LING-CSA) presents a proof-of-concept of a non-statistical conversational AI solution.  ...  Natural language is the most easily understood knowledge representation for people, but certainly not the best for computers because of its inherent ambiguous, complex and dynamic nature.  ...  TABLE 1 . 1 REVIEW OF NLP ADVANCEMENTS. CBOW) -based a neighbouring word predict a given word Skip gram -reverse of CBOW.  ... 
doi:10.4995/jclr.2020.12932 fatcat:oogpuyd6zvhixi22k33xawe3dm

Gamified crowdsourcing for idiom corpora construction

GülŞen Eryiğit, Ali Şentaş, Johanna Monti
2022 Natural Language Engineering  
The approach has been shown to have the potential to speed up the construction of idiom corpora for different natural languages to be used as second-language learning material, training data for supervised  ...  As opposed to classical crowd-processing annotation efforts in the field, for the first time in the literature, a crowd-creating & crowd-rating approach is implemented and tested for idiom corpora construction  ...  The authors would like to offer their special thanks to Cihat Eryigit for the discussions during the  ... 
doi:10.1017/s1351324921000401 fatcat:i36rqivxbrgnro6mmgf7yexvvq

Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance [article]

Gagan Bansal, Tongshuang Wu, Joyce Zhou, Raymond Fok, Besmira Nushi, Ece Kamar, Marco Tulio Ribeiro, Daniel S. Weld
2021 arXiv   pre-print
We conduct mixed-method user studies on three datasets, where an AI with accuracy comparable to humans helps participants solve a task (explaining itself in some conditions).  ...  Rather, explanations increased the chance that humans will accept the AI's recommendation, regardless of its correctness.  ...  ACKNOWLEDGMENTS This material is based upon work supported by ONR grant N00014-18-1-2193, the University of Washington WRF/Cable Professorship, and the Allen Institute for Artificial Intelligence (AI2)  ... 
arXiv:2006.14779v3 fatcat:763byqoxrrdttm3wyhhzys3qla
« Previous Showing results 1 — 15 out of 944 results