Filters








1,691 Hits in 4.6 sec

Text Understanding from Scratch [article]

Xiang Zhang, Yann LeCun
2016 arXiv   pre-print
This article demontrates that we can apply deep learning to text understanding from character-level inputs all the way up to abstract text concepts, using temporal convolutional networks (ConvNets).  ...  We show that temporal ConvNets can achieve astonishing performance without the knowledge of words, phrases, sentences and any other syntactic or semantic structures with regards to a human language.  ...  To decide on the first question, we extract all replaceable words from the given text and randomly choose r of them to be replaced.  ... 
arXiv:1502.01710v5 fatcat:f4s7ymfranckfjerey2snotcvu

Multidisciplinary Approaches to Text Understanding

John B. Black
1982 Contemporary Psychology  
In these books they discuss text understanding from the perspectives of linguistics, psychology, philosophy, and artificial intelligence. Each book examines text understand- ing at a different level.  ...  Multidisciplinary Approaches to Text Understanding Robert de Beaugrande Text, Discourse, and Process: Toward a Multidisciplinary Science of Texts.  ... 
doi:10.1037/020596 fatcat:o2rau2pb2raljam6wjmqlcrsgy

Natural Scene Text Understanding [chapter]

Celine Mancas, Bernard Gosseli
2007 Vision Systems: Segmentation and Pattern Recognition  
In particular, natural scene text understanding aiming at extracting text from daily images is the main concern of this text.  ...  NS text understanding may help to properly extract text from backgrounds to apply a particular compression to scene text, as already forecast in DjVu.  ... 
doi:10.5772/4966 fatcat:2vx67sdtrzfpfousqoagdukutm

The KERNEL text understanding system

Martha S. Palmer, Rebecca J. Passonneau, Carl Weir, Tim Finin
1993 Artificial Intelligence  
This article describes KERNEL, a text understanding system developed at the Unisys Center for Advanced Information Technology.  ...  It has stressed the encoding of deep, general common sense 43 PROTEUS is an acronym for PROtotype TExt Understanding System. the kernel text understanding system 43 and domain knowledge as predicate  ...  All conceptual predicates represented in the knowledge base are indicated here by the a x -P. the kernel text understanding system 23 includesP period34 , moment34 in 13a. 30 13 a.  ... 
doi:10.1016/0004-3702(93)90014-3 fatcat:jdxpictivzfofedwia6aw5ly3y

A Framework for Procedural Text Understanding

Hirokuni Maeta, Tetsuro Sasada, Shinsuke Mori
2015 Proceedings of the 14th International Conference on Parsing Technologies  
In this paper we propose a framework for procedural text understanding.  ...  the concepts in the text at once.  ...  We extract features from labeled arcs (u, v, l) by two processes: first we extract features from the arc and input recipe text and then we concatenate the label l to each feature we extract.  ... 
doi:10.18653/v1/w15-2206 dblp:conf/iwpt/MaetaSM15 fatcat:f6i6ql5wrrbzde7jfeukz4n4s4

Knowledge-Aware Procedural Text Understanding with Multi-Stage Training [article]

Zhihan Zhang, Xiubo Geng, Tao Qin, Yunfang Wu, Daxin Jiang
2020 arXiv   pre-print
In this paper, we propose a novel KnOwledge-Aware proceduraL text understAnding (KOALA) model, which leverages external knowledge sources to solve these issues.  ...  Specifically, we retrieve informative knowledge triples from ConceptNet and perform knowledge-aware reasoning while tracking the entities.  ...  To sum up, we introduce our KnOwledge-Aware proceduraL text understAnding (KOALA) model, which incorporates commonsense knowledge from ConceptNet and is trained with a multi-stage schema.  ... 
arXiv:2009.13199v1 fatcat:dbh3yip5mvdpvim2hjoohm6eoi

Device-Dependent Readability for Improved Text Understanding

A-Yeong Kim, Hyun-Je Song, Seong-Bae Park, Sang-Jo Lee
2014 Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)  
Readability is used to provide users with highquality service in text recommendation or text visualization.  ...  They are extracted from 28 French Foreign Language (FFL) textbooks written for adults learning FFL.  ...  The texts written for adults actually contain more entities than those written for children (Barzilay and Lapata, 2008) . The same is true for VP.  ... 
doi:10.3115/v1/d14-1146 dblp:conf/emnlp/KimSPL14 fatcat:dyx7n5z4nrethiltevbaqjp55y

Contextual Text Understanding in Distributional Semantic Space

Jianpeng Cheng, Zhongyuan Wang, Ji-Rong Wen, Jun Yan, Zheng Chen
2015 Proceedings of the 24th ACM International on Conference on Information and Knowledge Management - CIKM '15  
Representing discrete words in a continuous vector space turns out to be useful for natural language applications related to text understanding.  ...  We demonstrate the effectiveness of the framework in a number of tasks on text understanding, including word/phrase similarity measurements, paraphrase identification and question-answer relatedness classification  ...  In all experiments related to text understanding, including contextual word/phrase similarity measurements, paraphrase identification and question-answer relatedness classification. we achieve impressive  ... 
doi:10.1145/2806416.2806517 dblp:conf/cikm/ChengWWYC15 fatcat:f7f6nsonn5h77bnpmppvnxqeuq

Hybridization of Intelligent Solutions Architecture for Text Understanding and Text Generation

Anton Ivaschenko, Arkadiy Krivosheev, Anastasia Stolbova, Oleg Golovnin
2021 Applied Sciences  
This study proposes a new logical model for intelligent software architecture devoted to improving the efficiency of automated text understanding and text generation in industrial applications.  ...  Data Availability Statement: The data presented in this study are available on request from the corresponding author upon reasonable request.  ...  from texts formed in natural languages.  ... 
doi:10.3390/app11115179 fatcat:i2ln6dgdmfgblkfk3jam7czwl4

LOT: A Story-Centric Benchmark for Evaluating Chinese Long Text Understanding and Generation [article]

Jian Guan, Zhuoer Feng, Yamei Chen, Ruilin He, Xiaoxi Mao, Changjie Fan, Minlie Huang
2022 arXiv   pre-print
We construct new datasets for these tasks based on human-written Chinese stories with hundreds of words.  ...  However, long text modeling requires many distinct abilities in contrast to short texts, such as the modeling of long-range discourse and commonsense relations, and the coherence and controllability of  ...  The right candidates are extracted from the original stories (at the position of "[MASK]") while the wrong candidates are written by crowd-sourced annotators.  ... 
arXiv:2108.12960v2 fatcat:6c4g5rhwureftcoao2d5tah6wa

Towards Better Text Understanding and Retrieval through Kernel Entity Salience Modeling

Chenyan Xiong, Zhengzhong Liu, Jamie Callan, Tie-Yan Liu
2018 The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval - SIGIR '18  
We also provide examples showing how KESM conveys its text understanding ability learned from entity salience to search.  ...  This paper presents a Kernel Entity Salience Model (KESM) that improves text understanding and retrieval by better estimating entity salience (importance) in documents.  ...  ranking features convey KESM's text understanding ability learned from entity salience labels to search.  ... 
doi:10.1145/3209978.3209982 dblp:conf/sigir/XiongLCL18 fatcat:52w27kyxlnavle7xgdjr4xy4re

Simulating Text Understanding for Educational Applications with Latent Semantic Analysis: Introduction to LSA

Thomas K. Landauer, Joseph Psotka
2000 Interactive Learning Environments  
The following seven articles describe new educational tools that rely on a unique capability of computer programs to abstract knowledge relationships from vast quantities of text, and using this, to determine  ...  the similarity of knowledge expressed in two or more texts.  ...  Cohen describes how Latent Semantic Indexing (LSI) was used to model expert knowledge of a complex domain using very short texts written by journeymen and experts alone, rather than from a broad knowledge  ... 
doi:10.1076/1049-4820(200008)8:2;1-b;ft073 fatcat:zxwbsb4vkvdjxnyhkt2ilxcdm4

Implementing of Reading Strategies in the Academic Process as the Productive Way of Foreign Text Understanding

Svetlana I. Pozdeeva, Liubov A. Sobinova
2015 Mediterranean Journal of Social Sciences  
The authors of the paper put the relevant issue associated with a perception of foreign texts presented in English for pupils and students at schools and universities.  ...  learning strategy" and "reading strategy" definitions the authors suggest the algorithm to be studied and applied in the academic process at schools and universities as the main way to understand authentic texts  ...  Differential reading algorithm is aimed at the main information extraction and written compression of a text due to key words' isolating, semantic numbers' building and textual dominants' identification  ... 
doi:10.5901/mjss.2015.v6n6s4p390 fatcat:7fx4fxjeubc67klgh4dhqhmyva

LOT: A Story-Centric Benchmark for Evaluating Chinese Long Text Understanding and Generation

Jian Guan, Zhuoer Feng, Yamei Chen, Ruilin He, Xiaoxi Mao, Changjie Fan, Minlie Huang
2022 Transactions of the Association for Computational Linguistics  
We construct new datasets for these tasks based on human-written Chinese stories with hundreds of words.  ...  However, long text modeling requires many distinct abilities in contrast to short texts, such as the modeling of long-range discourse and commonsense relations, and the coherence and controllability of  ...  The right candidates are extracted from the original stories (at the position of ''[MASK]'') while the wrong candidates are written by crowd-sourced annotators.  ... 
doi:10.1162/tacl_a_00469 fatcat:wzxedwqfnbawvo3gbgmo4cjvpi

A Survey of Computational Semantics: Representation, Inference and Knowledge in Wide-Coverage Text Understanding

Johan Bos
2011 Language and Linguistics Compass  
To make interesting inferences, often additional background knowledge is required (not expressed in the analysed text or speech parts).  ...  This can be derived (and turned into first-order logic) from raw text, semistructured databases or large-scale lexical databases such as WordNet.  ...  Extracting Semantic Relations from Text Instead of relying on manually crafted lexical resources, one could also calculate lexical knowledge on the basis of large corpora.  ... 
doi:10.1111/j.1749-818x.2011.00284.x fatcat:23b6p3m2w5bi7llo6vi6hie24y
« Previous Showing results 1 — 15 out of 1,691 results