Filters








31 Hits in 5.4 sec

CATE: CAusality Tree Extractor from Natural Language Requirements [article]

Noah Jadallah, Jannik Fischbach, Julian Frattini, Andreas Vogelsang
2021 arXiv   pre-print
However, we lack an approach capable of extracting causal relations from natural language with reasonable performance.  ...  In this paper, we present our tool CATE (CAusality Tree Extractor), which is able to parse the composition of a causal relation as a tree structure.  ...  CATE: CAUSALITY TREE EXTRACTOR Principal Idea A RNN embraces the idea of natural language as a recursive structure.  ... 
arXiv:2107.10023v2 fatcat:kwkp76e4znfvrgl4uvyhbscpmu

Personalized musically induced emotions of not-so-popular Colombian music [article]

Juan Sebastián Gómez-Cañón and Perfecto Herrera and Estefanía Cano and Emilia Gómez
2021 arXiv   pre-print
Colombia (FARC) and (2) corridos from the "right-wing" paramilitaries Autodefensas Unidas de Colombia (AUC).  ...  In specific, we analyze traditional Colombian music containing politically charged lyrics of two types: (1) vallenatos and social songs from the "left-wing" guerrilla Fuerzas Armadas Revolucionarias de  ...  Recent Developments in OpenSMILE, the Munich Open-source Multimedia Feature Extractor.  ... 
arXiv:2112.04975v1 fatcat:f2oijtzkhbhnjkjwxmkbexdus4

Integrating Linguistic Theory and Neural Language Models [article]

Bai Li
2022 arXiv   pre-print
Transformer-based language models have recently achieved remarkable results in many natural language tasks.  ...  This has led many to question the relevance of linguistics for modern natural language processing.  ...  Model We use the transformer language model as a contextual embedding extractor (we write this as BERT for convenience). Let L be the layer index, which ranges from 0 to 12 on all of our models.  ... 
arXiv:2207.09643v1 fatcat:hz2roubcinfbrnq5hd427t62wq

A Review of Some Techniques for Inclusion of Domain-Knowledge into Deep Neural Networks [article]

Tirtharaj Dash, Sharad Chitlangia, Aditya Ahuja, Ashwin Srinivasan
2021 arXiv   pre-print
In many such instances, machine-based model construction may benefit significantly from being provided with human-knowledge of the domain encoded in a sufficiently precise form.  ...  Ethical-advice taker: Do language mod- ery. In IJCAI, vol. 7, 2462–2467 (Hyderabad, 2007). els understand natural language interventions? In [96] Pereira, F.  ...  In each cate- data. . .  ... 
arXiv:2107.10295v4 fatcat:ifkgq3cptbapfamr3dld57uruu

Proceeding of the 2nd Deep Learning Indaba-X Ethiopia Conference 2021

ejssd journal
2022 Ethiopian Journal of Science and Sustainable Development  
Its goal is to develop highly qualified, capable, competent, and innovative human resource in the field of science and technology so as to transfer relevant scientific knowledge and skills required for  ...  A similar predictive algorithm was made by other scholars from India [1], and had an accuracy rating of 80.05%. Our algorithm has an accuracy of 81.12%.  ...  The parser differs from conventional semantic parser in that it does not require lexicon-level rules to specify the mapping between natural language and logical form tokens.  ... 
doi:10.20372/ejssdastu:v0.i0.2022.412 doaj:6184f6f09a8e4521a49a11b6b41cd433 fatcat:jflueovu45aindmlhhvfv7xigy

Cortical response to naturalistic stimuli is largely predictable with deep neural networks [article]

Meenakshi Khosla, Gia H Ngo, Keith W Jamison, Amy Kuceyeski, Mert R Sabuncu
2020 bioRxiv   pre-print
Using movie watching data from the Human Connectome Project (HCP, N=158) database, we build group-level models of neural activity that incorporate several inductive biases about information processing  ...  Language 55b, SFL, PSL, 44, 45 5.  ...  accordance with pre-391 vious language fMRI studies (35).  ... 
doi:10.1101/2020.09.11.293878 fatcat:avodv4khorhube77talodegfk4

GELP: GAN-Excited Linear Prediction for Speech Synthesis from Mel-Spectrogram

Lauri Juvela, Bajibabu Bollepalli, Junichi Yamagishi, Paavo Alku
2019 Interspeech 2019  
However, for real-time applications, unidirectional (causal) models are required instead (Zen and Sak, 2015) .  ...  Language check The language of my dissertation has been checked by Lingsoft Language Services Oy. I have personally examined and accepted/rejected the results of the language check one by one.  ... 
doi:10.21437/interspeech.2019-2008 dblp:conf/interspeech/JuvelaBYA19 fatcat:bd6cc74arvb3joauc3wbeqnkba

From the heavy to the light verb

Begoña Sanromán Vilas
2018 Lingvisticæ investigationes  
Examples from the National Corpus of Russian Language (www.ruscorpora.ru) were used. Acknowledgments The draft of this paper has been read and commented upon by Ju. Apresjan, D. Beck, I.  ...  Acknowledgements This paper has benefited from helpful comments of François Lareau, Igor Mel' uk, Owen Rambow, and two reviewers.  ...  highest scoring projective tree, typically the highest scoring non-projective tree is only a small number of transformations away from.  ... 
doi:10.1075/li.00005.san fatcat:aqtad7regneflhblei7agmwzj4

Program Committee

2006 2006 Sixth IEEE International Workshop on Source Code Analysis and Manipulation  
There were papers from 24 different countries with every continent except Antarctica represented.  ...  Preface The Ninth Global Wordnet Conference was held at Nanyang Technological University, Singapore from 9-12th January 2018.  ...  Acknowledgments We wish to thank Georgia Sanna for the first joint analysis about the possibility of creating a wordnet for the Sardinian language.  ... 
doi:10.1109/scam.2006.23 dblp:conf/scam/X06c fatcat:2dhsf7loj5hlffu2jxpmlo2qcq

Why-query support in graph databases

Elena Vasilyeva
2016 2016 IEEE 32nd International Conference on Data Engineering Workshops (ICDEW)  
Considering graph queries to be complex structures with multiple constraints, iii we investigate different ways of generating explanations starting from the most general one that considers only a query  ...  From them, we formulate the following features of why-queries, which we discuss in this thesis, namely: holistic support of different cardinality-based problems, explanation of unexpected results and query  ...  Ten Cate et al.  ... 
doi:10.1109/icdew.2016.7495652 dblp:conf/icde/Vasilyeva16 fatcat:snct4vcbsrf4rpdjjbqtxyzpq4

Participatory rural appraisal (PRA): Analysis of experience

Robert Chambers
1994 World Development  
Explanations include reversals and shifts of emphasis: from etic to emit, closed to open, individual to group, verbal to visual, and measuring to comparing; and from extracting information to empowering  ...  Evidence to date shows high validity and reliability of information shared by local people through PRA compared with data from more traditional methods.  ...  standardized interviewing require close support from participant observation and general informal interview ing.  ... 
doi:10.1016/0305-750x(94)90003-5 fatcat:z4j7m4hgrjbpbmsj5phcfbxgta

A Survey of Unsupervised Generative Models for Exploratory Data Analysis and Representation Learning

Mohanad Abukmeil, Stefano Ferrari, Angelo Genovese, Vincenzo Piuri, Fabio Scotti
2021 ACM Computing Surveys  
From this perspective, exploratory data analysis plays a substantial role in learning the hidden structures that encompass the significant features of the data in an ordered manner by extracting patterns  ...  and learning representations; specifically, this article covers three families of methods based on their usage in the era of big data: blind source separation, manifold learning, and neural networks, from  ...  Recursive AEs (RAEs) were proposed in [171] and extended in [170] for natural language processing; the underlying idea of an RAE is to utilize a sparse tree to represent the general structure of sentences  ... 
doi:10.1145/3450963 fatcat:hqzhdz6turd73grqkkee2pwnbe

A systematic review of software architecture visualization techniques

Mojtaba Shahin, Peng Liang, Muhammad Ali Babar
2014 Journal of Systems and Software  
Results: We selected 53 papers from the initially retrieved 23,056 articles for data extraction, analysis, and synthesis based on pre-defined inclusion and exclusion criteria.  ...  The results from the data analysis enabled us to classify the identified VTs into four types based on the usage popularity: graph-based, notation-based, matrix-based, and metaphor-based VTs.  ...  Acknowledgement This research has been partially sponsored by the Natural Science Foundation of China (NSFC) under Grant No. 61170025. Appendix A. Selected studies See Table 19 .  ... 
doi:10.1016/j.jss.2014.03.071 fatcat:3xnzsi2nh5eizj3awarwjcwwsq

A context-aware traffic congestion estimation framework to overcome missing sensory data in Bangkok

Panraphee Raphiphan
2017
A TIS collects data from multiple sources, including sensors. However, data from sensors may become unavailable due to some reasons such as sensor damage or lost communication.  ...  In contrast, the CATE framework utilizes available external context information to infer the traffic situation.  ...  This was achieved by using natural language processing to extract traffic information from tweets and to reflect that data onto a map [38] .  ... 
doi:10.4225/03/58b6269ba045f fatcat:jko3xyigffhorhvynt4hhj3tiy

Affective Brain-Computer Interfaces [chapter]

Rafael Calvo, Sidney D'Mello, Jonathan Gratch, Arvid Kappas, Christian Mühl, Dirk Heylen, Anton Nijholt
2015 The Oxford Handbook of Affective Computing  
To achieve robust passive BCIs, efforts from applied and basic sciences have to be combined.  ...  On the one hand, applied fields such as affective computing aim at the development of applications that adapt to changes in the user states and thereby enrich the interaction, leading to a more natural  ...  In other words, dialogue partners tend to copy aspects of each other's language.  ... 
doi:10.1093/oxfordhb/9780199942237.013.024 fatcat:oia3yvk7hbf7ta6magl6bzjx6u
« Previous Showing results 1 — 15 out of 31 results