Agricultural robots rely on semantic segmentation for distinguishing between crops and weeds to perform selective treatments and increase yield and crop health while reducing the amount of chemicals ...used. Deep‐learning approaches have recently achieved both excellent classification performance and real‐time execution. However, these techniques also rely on a large amount of training data, requiring a substantial labeling effort, both of which are scarce in precision agriculture. Additional design efforts are required to achieve commercially viable performance levels under varying environmental conditions and crop growth stages. In this paper, we explore the role of knowledge transfer between deep‐learning‐based classifiers for different crop types, with the goal of reducing the retraining time and labeling efforts required for a new crop. We examine the classification performance on three datasets with different crop types and containing a variety of weeds and compare the performance and retraining efforts required when using data labeled at pixel level with partially labeled data obtained through a less time‐consuming procedure of annotating the segmentation output. We show that transfer learning between different crop types is possible and reduces training times for up to 80%. Furthermore, we show that even when the data used for retraining are imperfectly annotated, the classification performance is within 2% of that of networks trained with laboriously annotated pixel‐precision data.
We investigate the effects of two types of relationship between the words of a sentence or text - predictability and semantic similarity - by reanalysing electroencephalography (EEG) and functional ...magnetic resonance imaging (fMRI) data from studies in which participants comprehend naturalistic stimuli. Each content word's predictability given previous words is quantified by a probabilistic language model, and semantic similarity to previous words is quantified by a distributional semantics model. Brain activity time-locked to each word is regressed on the two model-derived measures. Results show that predictability and semantic similarity have near identical N400 effects but are dissociated in the fMRI data, with word predictability related to activity in, among others, the visual word-form area, and semantic similarity related to activity in areas associated with the semantic network. This indicates that both predictability and similarity play a role during natural language comprehension and modulate distinct cortical regions.
Domain invariance and discrimination of learned features as two crucial factors affect the performance of unsupervised domain adaptation (UDA) person re-identification (Re-ID). Person attributes ...(such as "backpack", "boots", "handbag", etc) remaining unchanged across multiple domains have been used as mid-level visual-semantic information in UDA person Re-ID. As two main challenges, both misalignment of attribute-related regions across multiple images and domain shift between source and target domains affect the learning of domain-invariant features (DIF). To address the above two challenges, this article proposes to take advantage of the stability of person attributes and the complementarity of person attributes and the corresponding low-level visual features to guide the learning of discriminative DIF. Specifically, the proposed solution contains the generation of latent attribute-correlated visual features (GLAVF), DIF learning under the guidance of person attributes, and the alignment of person attributes corresponding to the local regions of pedestrian images. Due to the gap between person attributes and visual features, person attributes are first converted into latent attribute-correlated visual features (LAVF) without any specific domain information in GLAVF, and then LAVF are used as the substitutions of person attributes to guide the learning of DIF. To enhance the discrimination of learned features, the proposed solution mainly explores the alignment between person attributes and corresponding local regions, and the alignment of the same person attributes across multiple pedestrian images. A fully connected layer is used to achieve the above two types of alignment in the proposed framework, which reduces the adverse impacts of inference information and ensures the semantic consistency between person attributes and corresponding local regions across multiple pedestrian images. The effectiveness of the proposed solution is confirmed on four existing datasets by comparative experiments.
Semantic similarity measures play an important role in information retrieval and information integration. Traditional approaches to modeling semantic similarity compute the semantic distance between ...definitions within a single ontology. This single ontology is either a domain-independent ontology or the result of the integration of existing ontologies. We present an approach to computing semantic similarity that relaxes the requirement of a single ontology and accounts for differences in the levels of explicitness and formalization of the different ontology specifications. A similarity function determines similar entity classes by using a matching process over synonym sets, semantic neighborhoods, and distinguishing features that are classified into parts, functions, and attributes. Experimental results with different ontologies indicate that the model gives good results when ontologies have complete and detailed representations of entity classes. While the combination of word matching and semantic neighborhood matching is adequate for detecting equivalent entity classes, feature matching allows us to discriminate among similar, but not necessarily equivalent entity classes.
We argue that the pictorial nature of certain constructions in signs and in gestures explains surprising properties of their syntax. In several sign languages, the standard word order (e.g. SVO) gets ...turned into SOV (with preverbal arguments) when the predicate is a classifier, a distinguished construction with highly iconic properties (e.g. Pavlič, 2016). In silent gestures, participants also prefer an SOV order in extensional constructions, irrespective of the word order of the language they speak (Goldin-Meadow et al., 2008). But in silent gestures and in Brazilian Sign Language (Libras), intensional constructions can override these SOV preferences, yielding SVO instead (Schouwstra & de Swart, 2014; Napoli et al., 2017). This distinction was argued to be due to iconicity: arguments are expressed before the verb if they correspond to entities that are present before the action, otherwise they follow the verb. While agreeing with this intuition, we propose that the extensional/intensional distinction is neither empirically nor theoretically appropriate. In new data from American Sign Language, we replicate the distinction among extensional classifier predicates: for
x ate up the ball
, the ball is typically seen before the eating and a preposed object is preferred; but for
x spit out the ball
, the ball is typically seen after the spitting and a postposed object is preferred, although both
eat up
and
spit out
are used extensionally. We extend this finding to data involving pro-speech (= speech-replacing) gestures embedded in French sentences. We argue for a Visibility Generalization according to which arguments appear before the verb if their denotations are typically visible before the action, and we develop a new formal account within a pictorial semantics for visual animations (inspired by Greenberg and Abusch). It derives the observed word order preferences, it explains how the semantics of classifier predicates combines iconic and conventional properties, and it makes a more general point: sign language semantics combines logical semantics with pictorial semantics.
Executive functions recently have gained attention as important contributors to language performance in aphasia. Picture/word interference embraces both executive control and lexical processing ...through distracted confrontation naming. Thus, we created a novel approach that embedded interfered-naming into an established lexical therapy framework. We aimed to (1) investigate patients' behavioural response to distractor types and treatment methods of interfered-naming and (2) determine specific therapy effects on linguistic versus executive processing and initial eligibility criteria.
Persons with word-finding difficulties in chronic aphasia received 4-week therapy in a block design with thorough pre-post-testing including computational modelling. During therapy, picture naming was distracted by auditory stimuli, which were primed by a preceding comprehension task and directly assisted by increasing semantic or phonological cues.
Nineteen participants were included in the diagnostic study, 12 of which also completed the therapy study. Distractor types did not generally yield differential effects at baseline. The novel linguistic-executive treatment significantly improved pure naming in most (9 out of 12) cases, fostered generalisation to untrained items, increased semantic weights in the computational model, and reduced automated speech. Therapy gains correlated positively with initial distractor comprehension, lexical-semantics, and word discrimination and negatively with automated speech and conceptual-semantics.
The interference paradigm combined with computational modelling offers a useful tool for aphasia diagnosis, and the new treatment approach revealed to be effective. Semantic and executive processing appear to be the core source of improvements.
Orthographic systems vary dramatically in the extent to which they encode a language's phonological and lexico-semantic structure. Studies of the effects of orthographic transparency suggest that ...such variation is likely to have major implications for how the reading system operates. However, such studies have been unable to examine in isolation the contributory effect of transparency on reading because of covarying linguistic or sociocultural factors. We first investigated the phonological properties of languages using the range of the world's orthographic systems (alphabetic, alphasyllabic, consonantal, syllabic, and logographic), and found that, once geographical proximity is taken into account, phonological properties do not relate to orthographic system. We then explored the processing implications of orthographic variation by training a connectionist implementation of the triangle model of reading on the range of orthographic systems while controlling for phonological and semantic structure. We show that the triangle model is effective as a universal model of reading, able to replicate key behavioral and neuroscientific results. The model also generates new predictions deriving from an explicit description of the effects of orthographic transparency on how reading is realized and defines the consequences of orthographic systems on reading processes. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
The purpose of this paper is to provide an overview of the terminology designating humours as it appears in Romania. From medicine to literature, from the Hippocratic theory of the four humours ...(umori) to the etymological doublet umor, this research paper will investigate the semantic and onomasiological fluctuations of three lexical fields: umoare (fluid), umoare (temperament) and umor (humour). The study does not aim to provide a comprehensive inventory of the terms (and their contexts) belonging to these semantic categories, but to identify the terms denoting the three concepts and their evolution in the Romanian language in its process of modernisation.
This article discusses translation strategies involved in reproducing ludic effects in Russian and Ukrainian translations of Coetzee’s novel
Disgrace
. Ludic effects, embedded in the text, outline ...the potential result(s) of literary gaming. Created in different games, a number of ludic effects trigger ludic stylistics – a new heuristic area of linguistic “ludology”. The paper defines ludic stylistics as an artistic phenomenon manifested in literary text due to unconventional combinations of various linguistic means. In Coetzee’s
Disgrace
, ludic stylistics is the result of psychonarrative games aimed to transform plot-driven narratives into experience-centred. Psychonarrative games are governed by two principles – “external via internal” and semantic intrusiveness through semantic, plot-building, and compositional games represented at the macro- and microlevels of literary text. The study focuses on sematic games which enable construing new, emergent textual senses which bring the personage’s / narrator’s unceasing, obsessive experience of traumatic events to the fore. The paper looks at the translation strategies of rendering ludic effects in the translations of Coetzee’s
Disgrace
from two perspectives – intentional and receptive. The reproduction of such effects in Russian and Ukrainian translations of
Disgrace
is grounded in lexico-semantic, syntactic, associative-figurative, and functional equivalence, as well as respective loss, and gain.
Lingvistică și filologie română Siserman, Mihaela Munteanu
Philologica Jassyensia,
01/2021, Letnik:
33, Številka:
1
Journal Article
Recenzirano
Odprti dostop
Nicolae FELECAN, Lingvistică şi filologie română, Editura Mega, Cluj-Napoca, 2020, 366 p. Justificarea tipăririi sub un titlu integrator a unor studii şi articole apărute în ultimii zece ani în ...diferite volume ale unor manifestări ştiinţifice din ţară sau din străinătate este menţionată de însuşi autor în Cuvântul-înainte al cărţii: „Faţă de versiunile iniţiale, unele materiale au fost supuse revizuirii sub aspect formal sau îmbogăţite cu date noi" (p. 9). Termenii inventariaţi nu sunt, de regulă, menţionaţi în dicţionarele curente ale limbii române sau, atunci când sunt înregistraţi, se pot constata diferite glisări semantice (a se vedea, de exemplu, semnificaţia „zonală", specifică păcurarilor năsăudeni, a cuvântului galite ori a lexemului mărturie („târg"), al cărui transfer semantic poate fi explicat prin asocierea mersului la târg în ziua în care târgoveţii „veneau la oraş în calitate de martori" (p. 31). Edificiile religioase (biserici, mănăstiri, schituri) reprezintă reflecţia studiului Denumiri ale lăcaşurilor de cult din România din perspectivă onomastică, nume cunoscute, în mentalul colectiv, ca hram (în unele situaţii particulare acesta putând fi modificat în istoricul lăcaşului de cult sau cazuri în care pentru acelaşi aşezământ religios există mai multe hramuri). În afara numelor de sfinţi sau a unor evenimente biblice nou-testamentare, denumirile lăcaşurilor de cult pot fi raportate şi la poziţionarea cadastrală: