Filters








188 Hits in 4.0 sec

Populating Web Scale Knowledge Graphs using Distantly Supervised Relation Extraction and Validation [article]

Sarthak Dash, Michael R. Glass, Alfio Gliozzo, Mustafa Canim
2019 arXiv   pre-print
The designed system leverages a deep learning based technology for relation extraction that can be trained by a distantly supervised approach.  ...  In addition to that, the system uses a deep learning approach for knowledge base completion by utilizing the global structure information of the induced KG to further refine the confidence of the newly  ...  It consists of two main components: Relation extraction, a deep learning based distantly supervised system to detect relations from text; Relation validation, a deep learning based knowledge base validation  ... 
arXiv:1908.08104v2 fatcat:6r6yxn2ptbh2ppbrms3ogwt3ou

Populating Web-Scale Knowledge Graphs Using Distantly Supervised Relation Extraction and Validation

Sarthak Dash, Michael R. Glass, Alfio Gliozzo, Mustafa Canim, Gaetano Rossiello
2021 Information  
The designed system leverages a deep-learning-based technology for relation extraction that can be trained by a distantly supervised approach.  ...  In addition, the system uses a deep learning approach for knowledge base completion by utilizing the global structure information of the induced KG to further refine the confidence of the newly discovered  ...  It consists of two main components: relation extraction, a deep-learning-based distantly supervised system to detect relations from text, and relation validation, a deep-learningbased knowledge base validation  ... 
doi:10.3390/info12080316 fatcat:ceaagl6ptrhsdpg2m73xwpzgwu

UNIQORN: Unified Question Answering over RDF Knowledge Graphs and Natural Language Text [article]

Soumajit Pramanik, Jesujoba Alabi, Rishiraj Saha Roy, Gerhard Weikum
2021 arXiv   pre-print
Conversely, systems from the IR and NLP communities have addressed QA over text, but barely utilize semantic data and knowledge.  ...  Our method, called UNIQORN, builds a context graph on the fly, by retrieving question-relevant triples from the RDF data and/or the text corpus, where the latter case is handled by automatic information  ...  Hence this focus on distantly supervised methods.  ... 
arXiv:2108.08614v2 fatcat:72so4mutjzfsrgrgsjrdsrthbm

Robust Retrieval Augmented Generation for Zero-shot Slot Filling [article]

Michael Glass, Gaetano Rossiello, Md Faisal Mahbub Chowdhury, Alfio Gliozzo
2021 arXiv   pre-print
Automatically inducing high quality knowledge graphs from a given collection of documents still remains a challenging problem in AI.  ...  The recent works in the field try to solve this task in an end-to-end fashion using retrieval-based language models.  ...  The community proposed different solutions to improve relation extraction performance, such as rule-based (Angeli et al., 2015) , supervised (Zhang et al., 2017) , or distantly supervised (Glass et  ... 
arXiv:2108.13934v2 fatcat:q7476utkl5b5hlhjkwvgdlsnjm

Embedding Time Expressions for Deep Temporal Ordering Models

Tanya Goyal, Greg Durrett
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
representations of neural net models.  ...  Data-driven models have demonstrated stateof-the-art performance in inferring the temporal ordering of events in text.  ...  This demonstrates that the temporal model does not use the knowledge from time expressions when making temporal relation predictions.  ... 
doi:10.18653/v1/p19-1433 dblp:conf/acl/GoyalD19 fatcat:xlg4anjlsjboxiesyuitypw5ie

Learning Consistent Deep Generative Models from Sparse Data via Prediction Constraints [article]

Gabriel Hope, Madina Abdrakhmanova, Xiaoyin Chen, Michael C. Hughes, Michael C. Hughes, Erik B. Sudderth
2020 arXiv   pre-print
Our approach enables advances in generative modeling to directly boost semi-supervised classification performance, an ability we demonstrate by augmenting deep generative models with latent variables capturing  ...  We develop a new framework for learning variational autoencoders and other deep generative models that balances generative and discriminative goals.  ...  Related work on consistency.  ... 
arXiv:2012.06718v1 fatcat:hka7e23vs5ge7nbus7mmbz32vm

Learning Representations of Social Media Users [article]

Adrian Benton
2018 arXiv   pre-print
We then show how user features can be employed as distant supervision to improve topic model fit.  ...  In this thesis, we begin by showing how user representations can be learned from multiple types of user behavior on social media.  ...  We also present a new topic model, deep Dirichlet Multinomial Regression (dDMR), that can better make use of highy-dimensional and only distantly related features, improving upon a previous supervised  ... 
arXiv:1812.00436v1 fatcat:qp2hf6f6nfe7djyjakkns36epq

ClassiNet -- Predicting Missing Features for Short-Text Classification [article]

Danushka Bollegala, Vincent Atanasov, Takanori Maehara, Ken-ichi Kawarabayashi
2018 arXiv   pre-print
Moreover, we propose a method based on graph propagation to find features that are indirectly related to a given short-text.  ...  for finding related features.  ...  FTS can be considered as a method that uses clusters of features induced from the data instances to overcome the feature sparseness problem.  ... 
arXiv:1804.05260v1 fatcat:b3ev546g5jgvxo4noo5nsd7xii

A Survey on Knowledge Graphs: Representation, Acquisition and Applications [article]

Shaoxiong Ji and Shirui Pan and Erik Cambria and Pekka Marttinen and Philip S. Yu
2021 IEEE Transactions on Neural Networks and Learning Systems   accepted
Knowledge graph embedding is organized from four aspects of representation space, scoring function, encoding models, and auxiliary information.  ...  We further explore several emerging topics, including meta relational learning, commonsense reasoning, and temporal knowledge graphs.  ...  Relation extraction suffers from noisy patterns under the assumption of distant supervision, especially in text corpus of different domains.  ... 
doi:10.1109/tnnls.2021.3070843 pmid:33900922 arXiv:2002.00388v4 fatcat:4l2yxnf3wbg4zpzdumduvyr4he

Intelligent mining of large-scale bio-data: Bioinformatics applications

Farahnaz Sadat Golestan Hashemi, Mohd Razi Ismail, Mohd Rafii Yusop, Mahboobe Sadat Golestan Hashemi, Mohammad Hossein Nadimi Shahraki, Hamid Rastegari, Gous Miah, Farzad Aslani
2017 Biotechnology & Biotechnological Equipment  
Data mining, as biology intelligence, attempts to find reliable, new, useful and meaningful patterns in huge amounts of data.  ...  Bioinformatics creates heuristic approaches and complex algorithms using artificial intelligence and information technology in order to solve biological problems.  ...  Besides pairwise alignments, Multiple Sequence Alignments (MSAs) have been used to align closely related sequences, distantly related sequences or both [42] .  ... 
doi:10.1080/13102818.2017.1364977 fatcat:qmbiss53wfggtc7ayj2ysgt5rq

Machine Knowledge: Creation and Curation of Comprehensive Knowledge Bases [article]

Gerhard Weikum, Luna Dong, Simon Razniewski, Fabian Suchanek
2021 arXiv   pre-print
Over the last decade, large-scale knowledge bases, also known as knowledge graphs, have been automatically constructed from web contents and text sources, and have become a key asset for search engines  ...  . • Deep Learning: The most recent addition to the methodogical repertoire is deep neural networks, trained in a supervised or distantly supervised manner (see, e.g., [71, 200] ).  ...  Distantly Supervised Extraction from DOM Trees Like with distant supervision for extraction from text, we start with seed pairs (S, O) of entities for a given property P and make two key assumptions.  ... 
arXiv:2009.11564v2 fatcat:vh2lqfmhhbcwpf6dcsej3hhvgy

Information Seeking in the Spirit of Learning: a Dataset for Conversational Curiosity [article]

Pedro Rodriguez, Paul Crook, Seungwhan Moon, Zhiguang Wang
2020 arXiv   pre-print
Responses using a user's prior knowledge increase engagement.  ...  and user responses such as "liking" messages or asking followup questions, we design a Wizard-of-Oz dialog task that tests the hypothesis that engagement increases when users are presented with facts related  ...  always taking the role of an assistant, (3) are dialog responses constrained, (4) are document groundings annotated-as opposed to distantly supervised-and fine-grained, (5) Table 5 : indicates a  ... 
arXiv:2005.00172v2 fatcat:ogflfn7uczfbbjbh4xv2snyttu

A Survey Of Cross-lingual Word Embedding Models [article]

Sebastian Ruder, Ivan Vulić, Anders Søgaard
2019 arXiv   pre-print
Cross-lingual representations of words enable us to reason about word meaning in multilingual contexts and are a key facilitator of cross-lingual transfer when developing natural language processing models  ...  In such cases, using a distantly supervised seed lexicon of identical strings in both languages is often preferable .  ...  Moreover, it seems that signals from other modalities are typically more useful as an additional source of information to complement the uni-modal signals from text, rather than using other modalities  ... 
arXiv:1706.04902v3 fatcat:lts6uop77zaazhzlbygqmdsama

A Survey of Cross-lingual Word Embedding Models

Sebastian Ruder, Ivan Vulić, Anders Søgaard
2019 The Journal of Artificial Intelligence Research  
Cross-lingual representations of words enable us to reason about word meaning in multilingual contexts and are a key facilitator of cross-lingual transfer when developing natural language processing models  ...  In such cases, using a distantly supervised seed lexicon of identical strings in both languages is often preferable .  ...  Moreover, it seems that signals from other modalities are typically more useful as an additional source of information to complement the uni-modal signals from text, rather than using other modalities  ... 
doi:10.1613/jair.1.11640 fatcat:vwlgtzzmhfdlnlyaokx2whxgva

One Model to Rule them all: Multitask and Multilingual Modelling for Lexical Analysis [article]

Johannes Bjerva
2017 arXiv   pre-print
., language preservation, as speakers of minority languages will have a lower degree of pressure to using high-resource languages.  ...  In the short term, answering the specific research questions posed should be of use to NLP researchers working towards the same goal.  ...  For instance, MTL can have the role of being a distant supervision signal, in the sense that the tasks used might be fairly distantly related.  ... 
arXiv:1711.01100v1 fatcat:njotlhapbjggxkfndwa3wkwg4q
« Previous Showing results 1 — 15 out of 188 results