检索结果-内蒙古大学图书馆

Bilinear pooling in video-QA: empirical challenges and motivational drift from neurological parallels

PEERJ COMPUTER SCIENCE 2022年 8卷 e974页

作者： Winterbottom, Thomas Xiao, Sarah McLean, Alistair Al Moubayed, Noura Univ Durham Dept Comp Sci Durham England Univ Durham Business Sch Durham England Carbon AI Middlesbrough Cleveland England

Bilinear pooling (BLP) refers to a family of operations recently developed for fusing features from different modalities predominantly for visual question answering (VQA) models. Successive BLP techniques have yielded higher performance with lower computational expense, yet at the same time they have drifted further from the original motivational justification of bilinear models, instead becoming empirically motivated by task performance. Furthermore, despite significant success in text-image fusion in VQA, BLP has not yet gained such notoriety in video question answering (video-QA). Though BLP methods have continued to perform well on video tasks when fusing vision and non-textual features, BLP has recently been overshadowed by other vision and textual feature fusion techniques in video-QA. We aim to add a new perspective to the empirical and motivational drift in BLP. We take a step back and discuss the motivational origins of BLP, highlighting the often-overlooked parallels to neurological theories (dual coding theory and The Two-Stream Model of Vision). We seek to carefully and experimentally ascertain the empirical strengths and limitations of BLP as a multimodal text-vision fusion technique in video-QA using two models (TVQA baseline and heterogeneous-memory-enchanced 1-IME' model) and four datasets (TVQA, TGif-QA, MSVD-QA, and EgoVQA). We examine the impact of both simply replacing feature concatenation in the existing models with BLP, and a modified version of the TVQA baseline to accommodate BLP that we name the 'dual-stream' model. We find that our relatively simple integration of BLP does not increase, and mostly harms, performance on these video-QA benchmarks. Using our insights on recent work in BLP for video-QA results and recently proposed theoretical multimodal fusion taxonomies, we offer insight into why BLP-driven performance gain for video-QA benchmarks may be more difficult to achieve than in earlier VQA models. We share our perspective on, and s

关键词： Video question answering Bilinear pooling Multimodal fusion Deep-CCA TVQA Ego-VQA MSVD-QA TGif-QA dual coding theory Two-stream model of vision

来源：评论

学校读者我要写书评

暂无评论

?I have to translate the colors?: Description and implications of a genuine case of phoneme color synaesthesia

引用

CONSCIOUSNESS AND COGNITION 2023年第1期111卷 103509页

作者： Bouvet, Lucie Magnen, Cynthia Bled, Clara Tardieu, Julien Ehrle, Nathalie Univ Toulouse CERPPS Toulouse France Univ Toulouse CNRS Maison Sci Homme Toulouse Toulouse France CHRU Maison Blanche Serv Neurol Reims France Univ Paris Cite Lab MC2 UR 7536 Paris France Univ Toulouse Jean Jaures CERPPS 5 Allees Antonio Machado F-31058 Toulouse France

The aim of this paper was to assess the genuineness of a case of phoneme color synesthesia (VA) by evaluating the influence of several psycholinguistic levels with different stimuli (isolated vowels, nonsense syllables, and words). Results demonstrate the robustness of the synesthesia no matter what the type of stimuli. To explore how this form of synesthesia manifested itself in everyday conversation, interviews were also conducted. VA reported that, in the context of conversation, phonemes still evoked colors that she had to translate in order to access the meaning. She also reported that her mental representations were multisensorial and that the verbal dimension was almost non-existent. We address several implications of this phoneme color synesthesia: the atypical speech perception that it implies, the cognitive cost of this stable system, and its relation to a specific cognitive functioning.

关键词： Phoneme color synesthesia Speech perception Multisensory representations Veridical mapping Segmental speech processing dual coding theory Thinking in pictures

来源：评论

学校读者我要写书评

暂无评论

Analýza grafické informace v digitálních vzdělávacích zdrojích pro obor Informatika

Analýza grafické informace v digitálních vzdělávacích...

引用

作者： Marschnerová, Michaela Charles University of Prague

This bachelor thesis is focused on applied graphics used on online learning platforms and its impact on learning process. The aspects and impacts of applied graphics are analysed in the theoretical part. The practical part presents the evaluation criteria for design principles used in the educational materials. There are also suggested analysis methods that are immediately applied to five online platforms chosen from ***. These materials are intended for education or self-study of information and communication technologies at lower-secondary schools. KEYWORDS dual coding theory, applied graphics, information visualization, online education resources

关键词： dual coding theory applied graphics information visualization online education resources

来源：评论

学校读者我要写书评

暂无评论

Examining the Role of Semantic Similarity in Online Restaurant Review Evaluations 26

Examining the Role of Semantic Similarity in Online Restaura...

引用

Conference of the Association-for-Information-Systems (AMCIS)

作者： Li, Lin Ren, Gang Hong, Taeho Yang, Sung-Byung Kyung Hee Univ Seoul South Korea Kookmin Univ Seoul South Korea Pusan Natl Univ Busan South Korea

ISBN: (纸本)9781733632546

Both language and image are critical for the grasp of information embedded in online reviews. While a large quantity of research has focused on the role of textual features and visual features separately, the specific role of similarity between textual and visual information in online review evaluations (e.g., review usefulness and review enjoyment) remains unaddressed. Thus, drawing on dual coding theory, this study attempts to investigate the impacts of textual and visual features on review evaluations by employing the Latent Dirichlet Allocation (LDA) topic modeling and Google Vision API's web detection techniques in the context of online restaurant review (ORR). Moreover, the moderating role of semantic similarity is examined in the relationships between textual/visual features and ORR evaluations. It is believed that this study could provide implications on information comprehension, draw consumer interest, and provide suggestions for restaurant managers to tune levels of review evaluation in a proper manner.

关键词： Online restaurant review review evaluation image mining text mining semantic similarity dual coding theory

来源：评论

学校读者我要写书评

暂无评论

When online reviews meet virtual reality: Effects on consumer hotel booking

引用

ANNALS OF TOURISM RESEARCH 2020年第0期81卷 102860-000页

作者： Zeng, Guojun Cao, Xinning Lin, Zhibin Xiao, Sarah H. Sun Yat Sen Univ Sch Tourism Management Guangzhou Peoples R China Univ Durham Business Sch Mill Hill Lane Durham DH1 3LB England

This study examines the direct and interaction effects of online reviews (quality and quantity) and virtual reality on consumer hotel booking. Data were collected from two 2 x 2 experimental studies. The results show a direct effect of both online review and virtual reality's application on behavioral intention, and the influence of online reviews on behavioral intention was weakened when virtual reality was applied. Moreover, online reviews and virtual reality had a significant combined effect on behavioral intention, with greater strength than that of online reviews alone. The findings provide insights for travel and tourism managers to enhance marketing communication effects by proper use of virtual reality to complement online reviews.

关键词： Online review Virtual reality dual coding theory Sensory information Textual information

来源：评论

学校读者我要写书评

暂无评论

Are users attracted by playlist titles and covers? Understanding playlist selection behavior on a music streaming platform

引用

JOURNAL OF INNOVATION & KNOWLEDGE 2022年第3期7卷

作者： Li, Zhi Song, Min Duan, Shen Wang, Zhe Tiangong Univ Sch Econ & Management Tianjin 300387 Peoples R China Renmin Univ China Sch Business Beijing 100872 Peoples R China Sichuan Tourism Univ Sch Econ & Management Chengdu 610100 Sichuan Peoples R China

Playlists have become the main entry point for users to obtain music resources. This study aimed to investigate the features of playlist titles and covers that attract users and the consequences of playlist selection on music streaming platforms. In this study, 7,606 playlist data were collected from the NetEase Cloud Music platform. The linguistic style of playlist titles was classified based on the language expectancy theory (LET). Artificial intelligence technology was used to recognize and classify the image style of playlist covers according to the content of image features. We then employed hierarchical multiple linear regression and grouping regression to study the moderating effects of playlist features (linguistic style of titles and image style of covers) on the relationship between the number of playlist comments and the number of plays. The findings revealed that (1) the number of playlist comments (comment-count) has a stronger positive relationship with the number of plays (play-count) when the linguistic style of titles, such as concrete, perceptual, interactive, and social languages, appears more frequently and (2) the image style of playlist covers, such as natural, non-natural, painting and text, and portrait images, strengthens the relationships between playlist comment-count and linguistic style of playlist titles on play-count.

关键词： User-generated playlist (UGP) Playlist selection Language expectancy theory (LET) dual coding theory Music streaming platform

来源：评论

学校读者我要写书评

暂无评论

The effects of computer assisted English listening comprehension learning on listening and reading performance of Taiwanese EFL learners

应用语文学报

引用

应用语文学报 2020年第11期 45-67页

作者：張旭琪張喬妍蔡世峰

The performance of Taiwanese undergraduate students' English proficiency in the global proficiency index is always a pain in the English teaching community in Taiwan, but how to improve the English proficiency of Taiwanese students is still at a loss. To this end, the author proposes a practical and feasible way, a dual-coding learning theory that uses auditory and visual sensory to assist English language teaching. The current study aimed to investigate whether the implementation of complementary aural and visual English (dual-coding) language instruction impacted results on the TOEIC test given at a private university in central Taiwan. Both ANCOVA and mixed-design ANOVA results on multiple measures indicated that students who were placed in the TV-CLT instruction group performed significantly higher in listening comprehension and reading comprehension than some of the other groups after one semester of instruction. The results have confirmed the effects of dual coding theory (DCT) in listening and reading comprehension. Therefore, the author strongly recommends that in addition to the current English teaching model in Taiwan, English listening comprehension should be developed as early as possible.

关键词： English Language Teaching Communicative Language Teaching dual coding theory TOEIC test Studio Classroom (TV program)

来源：评论

学校读者我要写书评

暂无评论

Gesture and Vocabulary Learning in a Second Language

引用

LANGUAGE LEARNING 2019年第1期69卷 177-197页

作者： Huang, Xiaoyi Kim, Nayoung Christianson, Kiel Univ Illinois 1310 S 6th StRm 240MC-708 Champaign IL 61820 USA

When introducing new words in a second language (L2), presenting vocabulary with concurrent gestures might facilitate learners' recollection of new words. Previous research has suggested that this gestural advantage might hold only for gestures that overlap with the semantics of the words. dual coding theory predicts that learners should learn input better when multiple sensory routes act as aids to retrieval. Our research replicated and extended previous studies examining the limits of gestures in learning L2 vocabulary. A within-participant design directly compared the effects of pairing low idiosyncratic gestures (gestures traditionally iconic with word meanings) versus high idiosyncratic gestures (gestures that likely need to be idiosyncratically paired with word meanings) with L2 vocabulary presentation, relative to using no gestures. Results supported dual coding theory: All gestures were helpful if they were not confusable with other to-be-learned words and if the number of words presented was limited. Open Practices This article has been awarded an Open Data badge. All data are publicly accessible via the Open Science Framework at . Learn more about the Open Practices badges from the Center for Open Science: .

关键词： gesture lexical learning Mandarin Chinese dual coding theory second language vocabulary

来源：评论

学校读者我要写书评

暂无评论

Užitá grafika ve vzdělávacích zdrojích pro oblast informačních a komunikačních technologií

Užitá grafika ve vzdělávacích zdrojích pro oblast info...

引用

作者： Stejskalová, Michaela Charles University of Prague

The bachelor thesis deals with applied graphics in the educational resources and its impact on learning process. The aspects and impacts of applied graphics are analysed in the theoretical part. Secondly, there are suggested the evaluation criteria for design principles in the educational materials. Subsequently, the analysis methods are set and applied to five selected publications that have been intended for education or self-study of information and communication technologies at lower-secondary schools. KEYWORDS dual coding theory, textbooks, applied graphics, information visualization

关键词： dual coding theory textbooks applied graphics information visualization

来源：评论

学校读者我要写书评

暂无评论

The effects of visuospatial cueing on EFL learners' science text and picture processing through mobile phones

引用

TECHNOLOGY IN SOCIETY 2019年 59卷 101191-000页

作者： Yang, Hui-Yu Guangdong Univ Petrochem Technol Foreign Language Dept Maoming Peoples R China

This quasi-experimental study examined whether or not attention cueing can benefit learners in learning the kinematic operation of a mechanical system. This study investigated the interaction effects of the dynamic and cued conditions on the test results as well as cognitive load. Self-regulatory, mobile-phone-based visualizations depicting the kinematics of a flushing cistern comprised the instrument. The participants consisted of 174 English as a foreign language (EFL) learners from four sections of a reading course. After assessing their backgrounds, they were randomly assigned to one of four experimental conditions-pictures only, pictures plus cueing, animation only, and animation plus cueing. Immediately, after administering the experimental treatment, the participants answered comprehension tests and a cognitive load questionnaire. The results demonstrated that visuospatial cues, particularly in static diagrams, ameliorated the learners' learning effectiveness by facilitating their local comprehension better than their global comprehension. While animation benefited the learners in terms of solving directions and troubleshooting problems, the static diagrams and animations complemented each other in helping the learners understand the different kinematic functions of a mechanical system. The static diagrams were beneficial in enhancing local comprehension, whereas the animations benefited the learners' global comprehension as indicated in the learners' transfer test results. The experimental results revealed that the effectiveness of animation and visual cueing depend on the subject matter and purpose of the tests.

关键词： Animation Cognitive load theory dual coding theory Signaling Principle Mobile learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：