Semantic composition of distributed representations for query subtopic mining

2018 ◽  
Vol 19 (11) ◽  
pp. 1409-1419 ◽  
Author(s):  
Wei Song ◽  
Ying Liu ◽  
Li-zhen Liu ◽  
Han-shi Wang
2021 ◽  
Vol 31 (1) ◽  
pp. 139-154
Author(s):  
Svetlana Efimova

Abstract Im Kontext des neu aufgekommenen theoretischen Interesses für die Werkkategorie wird überwiegend das ,Einzelwerk‘ fokussiert. Im vorliegenden Aufsatz wird das heuristische Potenzial des Konzepts ,Gesamtwerk‘ als ein anderer Teil der Werkkategorie herausgearbeitet. Neben dem üblichen Aspekt eines ,Lebenswerks‘ schließt das Gesamtwerk vielfältige Werkkomplexe und Werkgruppierungen ein, die auf Produktions- oder Rezeptionsseite entstehen. Analysiert werden feste und variable Anordnungen, Ab- und Entgrenzungen zwischen Einzelwerken eines Autors, die das Gesamtwerk zu einem dynamischen Gefüge machen. Ein Werkkomplex bildet eine Zwischenstufe und ein Bindeglied zwischen ,Einzelwerk‘ und ,Gesamtwerk‘. Daher besitzt er eine besondere Relevanz für die semantische Zusammensetzung der Werkkategorie als Trias, deren Teile sich aufeinander beziehen: Opus – Werkkomplex – Œuvre.The newly arisen theoretical research on the literary work category focuses mainly on the ,single work‘. This paper argues for the heuristic potential of the ,oeuvre‘ as another part of the work category. In addition to the usual aspect of a ,life’s work‘, the oeuvre includes diverse work complexes and work groupings that emerge on the production or reception side. The paper analyzes a dynamic structure of the oeuvre: fixed and variable arrangements, boundaries and dissolution of borders between single works by the same author. A work complex forms an intermediate stage and a link between ,single work‘ and ,oeuvre‘. It therefore has a special relevance for the semantic composition of the work category as a triad, the parts of which refer to each other: single work – work complex – oeuvre.


2020 ◽  
Vol 1 (4) ◽  
pp. 381-401
Author(s):  
Ryan Staples ◽  
William W. Graves

Determining how the cognitive components of reading—orthographic, phonological, and semantic representations—are instantiated in the brain has been a long-standing goal of psychology and human cognitive neuroscience. The two most prominent computational models of reading instantiate different cognitive processes, implying different neural processes. Artificial neural network (ANN) models of reading posit nonsymbolic, distributed representations. The dual-route cascaded (DRC) model instead suggests two routes of processing, one representing symbolic rules of spelling–to–sound correspondence, the other representing orthographic and phonological lexicons. These models are not adjudicated by behavioral data and have never before been directly compared in terms of neural plausibility. We used representational similarity analysis to compare the predictions of these models to neural data from participants reading aloud. Both the ANN and DRC model representations corresponded to neural activity. However, the ANN model representations correlated to more reading-relevant areas of cortex. When contributions from the DRC model were statistically controlled, partial correlations revealed that the ANN model accounted for significant variance in the neural data. The opposite analysis, examining the variance explained by the DRC model with contributions from the ANN model factored out, revealed no correspondence to neural activity. Our results suggest that ANNs trained using distributed representations provide a better correspondence between cognitive and neural coding. Additionally, this framework provides a principled approach for comparing computational models of cognitive function to gain insight into neural representations.


2011 ◽  
Vol 18 (3) ◽  
pp. 343-374
Author(s):  
HAKKI C. CANKAYA ◽  
EDUARDO BLANCO ◽  
DAN MOLDOVAN

AbstractThis paper presents a method for the composition of at-location with other semantic relations. The method is based on inference axioms that combine two semantic relations yielding another relation that otherwise is not expressed. An experimental study conducted on PropBank, WordNet, and eXtended WordNet shows that inferences have high accuracy. The method is applicable to combining other semantic relations and it is beneficial to many semantically intense applications.


2021 ◽  
pp. 115418
Author(s):  
Haoqing Wang ◽  
Huiyu Mai ◽  
Zhi-hong Deng ◽  
Chao Yang ◽  
Luxia Zhang ◽  
...  

2003 ◽  
Vol 39 (2) ◽  
pp. 245-273 ◽  
Author(s):  
GEORGE M. HORN

Jackendoff (1997), whose analysis of idioms is based, in part, on work by Nunberg, Sag & Wasow (1994), discusses VP idioms and addresses the question of mobility. Both works identify fixed idioms, such as kick the bucket, and mobile idioms, such as spill the beans and take advantage of. Fixed idioms are ones whose NP objects are impervious to syntactic operations, as illustrated by the unacceptability, in their idiomatic sense, of sentences like *The bucket was kicked by Bill; while mobile idioms occur in sentences like The beans were spilled by Fred and Advantage was taken of Bill. Jackendoff correlates the mobility of VP idioms with a property that he refers to as metaphorical semantic composition. However, he observes that this property is not a sufficient condition for mobility.I will argue that the property of metaphorical semantic composition be replaced by a property of thematic composition, and that this property is a sufficient condition for mobility. A closer inspection of mobile idioms that have thematic composition reveals that they fall into two subtypes: expressions that have a property of ‘transparency of interpretation’, and ones that do not have this property. I refer to members of the first subtype as METAPHORS. I will demonstrate that there are no idiosyncratic constraints on their syntactic mobility, and will conclude that they need not be encoded in lexical entries as phrasal idioms. In these respects, they are distinct from members of the second subtype, whose degree of mobility is more limited, and which must be encoded in lexical entries as phrasal idioms. Finally, I will address the question of the necessity of thematic composition for mobility. Throughout the paper, I will assume that phrasal idioms are appropriately encoded in lexical entries of the types proposed by Jackendoff for fixed and mobile expressions.


Sign in / Sign up

Export Citation Format

Share Document