scholarly journals AttendAffectNet–Emotion Prediction of Movie Viewers Using Multimodal Fusion with Self-Attention

Sensors ◽  
2021 ◽  
Vol 21 (24) ◽  
pp. 8356
Author(s):  
Ha Thi Phuong Thao ◽  
B T Balamurali ◽  
Gemma Roig ◽  
Dorien Herremans
Keyword(s):  
Network Architecture ◽  
Affective Responses ◽  
Attention Mechanism ◽  
The Self ◽  
Multimodal Fusion ◽  
Sequential Dependency ◽  
Emotional Impact ◽  
Audio Features ◽  
Text Features ◽  
Better Than

In this paper, we tackle the problem of predicting the affective responses of movie viewers, based on the content of the movies. Current studies on this topic focus on video representation learning and fusion techniques to combine the extracted features for predicting affect. Yet, these typically, while ignoring the correlation between multiple modality inputs, ignore the correlation between temporal inputs (i.e., sequential features). To explore these correlations, a neural network architecture—namely AttendAffectNet (AAN)—uses the self-attention mechanism for predicting the emotions of movie viewers from different input modalities. Particularly, visual, audio, and text features are considered for predicting emotions (and expressed in terms of valence and arousal). We analyze three variants of our proposed AAN: Feature AAN, Temporal AAN, and Mixed AAN. The Feature AAN applies the self-attention mechanism in an innovative way on the features extracted from the different modalities (including video, audio, and movie subtitles) of a whole movie to, thereby, capture the relationships between them. The Temporal AAN takes the time domain of the movies and the sequential dependency of affective responses into account. In the Temporal AAN, self-attention is applied on the concatenated (multimodal) feature vectors representing different subsequent movie segments. In the Mixed AAN, we combine the strong points of the Feature AAN and the Temporal AAN, by applying self-attention first on vectors of features obtained from different modalities in each movie segment and then on the feature representations of all subsequent (temporal) movie segments. We extensively trained and validated our proposed AAN on both the MediaEval 2016 dataset for the Emotional Impact of Movies Task and the extended COGNIMUSE dataset. Our experiments demonstrate that audio features play a more influential role than those extracted from video and movie subtitles when predicting the emotions of movie viewers on these datasets. The models that use all visual, audio, and text features simultaneously as their inputs performed better than those using features extracted from each modality separately. In addition, the Feature AAN outperformed other AAN variants on the above-mentioned datasets, highlighting the importance of taking different features as context to one another when fusing them. The Feature AAN also performed better than the baseline models when predicting the valence dimension.

scholarly journals Fighting Together against the Pandemic: Learning Multiple Models on Tomography Images for COVID-19 Diagnosis

AI ◽  
2021 ◽  
Vol 2 (2) ◽  
pp. 261-273
Author(s):  
Mario Manzo ◽  
Simone Pellino
Keyword(s):  
Network Architecture ◽  
State Of The Art ◽  
Ensemble Classification ◽  
Effective Vaccine ◽  
Rt Pcr ◽  
Neural Network Architecture ◽  
Experimental Phase ◽  
Different Types ◽  
Polymerase Chain ◽  
Better Than

COVID-19 has been a great challenge for humanity since the year 2020. The whole world has made a huge effort to find an effective vaccine in order to save those not yet infected. The alternative solution is early diagnosis, carried out through real-time polymerase chain reaction (RT-PCR) tests or thorax Computer Tomography (CT) scan images. Deep learning algorithms, specifically convolutional neural networks, represent a methodology for image analysis. They optimize the classification design task, which is essential for an automatic approach with different types of images, including medical. In this paper, we adopt a pretrained deep convolutional neural network architecture in order to diagnose COVID-19 disease from CT images. Our idea is inspired by what the whole of humanity is achieving, as the set of multiple contributions is better than any single one for the fight against the pandemic. First, we adapt, and subsequently retrain for our assumption, some neural architectures that have been adopted in other application domains. Secondly, we combine the knowledge extracted from images by the neural architectures in an ensemble classification context. Our experimental phase is performed on a CT image dataset, and the results obtained show the effectiveness of the proposed approach with respect to the state-of-the-art competitors.

SOCIAL LOAFING AND SELF-BELIEFS: PEOPLE'S COLLECTIVE EFFORT DEPENDS ON THE EXTENT TO WHICH THEY DISTINGUISH THEMSELVES AS BETTER THAN OTHERS

1998 ◽  
Vol 26 (4) ◽  
pp. 329-340 ◽  
Author(s):  
Emmanuelle Charbonnier ◽  
Pascal Huguet ◽  
Markus Brauer ◽  
Jean-Marc Monteil
Keyword(s):  
Social Life ◽  
The Self ◽  
Social Loafing ◽  
Significant Component ◽  
Easy Task ◽  
Collective Effort ◽  
Better Than

An experiment tested whether the belief that one is better than others on performance aspects of the self moderates social loafing, the tendency for individuals to expend less effort when working collectively than when working individually. French students performed an easy task either coactively or collectively. They were then asked to rate how they individuate themselves from others on personal abilities within various domains of social life. The loafing effect was especially strong in participants who perceived themselves as better than others, suggesting that self-beliefs related to one's feeling of uniqueness is a significant component of social loafing. This finding is discussed from the perspective of Karau and Williams' (1993) Collective Effort Model.

When Praising Yourself Insults Others: Self-Superiority Claims Provoke Aggression

2017 ◽  
Vol 43 (7) ◽  
pp. 1008-1019 ◽  
Author(s):  
Carolien Van Damme ◽  
Eliane Deschrijver ◽  
Eline Van Geert ◽  
Vera Hoorens
Keyword(s):  
The Self ◽  
Decision Task ◽  
Actual Decision ◽  
Better Than

We tested the prediction, derived from the hubris hypothesis, that bragging might serve as a verbal provocation and thus enhance aggression. Experiments 1 and 2 were vignette studies where participants could express hypothetical aggression; Experiment 3 was an actual decision task where participants could make aggressive and/or prosocial choices. Observers disliked an explicit braggart (who claimed to be “better than others”) or a competence braggart as compared with an implicit braggart (who claimed to be “good”) or a warmth braggart, respectively. Showing that explicit and competence bragging function as verbal provocations, observers responded more aggressively to the explicit and competence braggart than to the implicit and warmth braggart, respectively. They did so because they inferred that an explicit and a competence braggart viewed other people and them negatively, and therefore disliked the braggart. Rather than praising the self, braggarts are sometimes viewed as insulting others.

scholarly journals DeepIsoFun: a deep domain adaptation approach to predict isoform functions

2018 ◽  
Vol 35 (15) ◽  
pp. 2535-2544 ◽  
Author(s):  
Dipan Shaw ◽  
Hao Chen ◽  
Tao Jiang
Keyword(s):  
Network Architecture ◽  
Domain Adaptation ◽  
Multiple Instance Learning ◽  
Training Data ◽  
Supplementary Information ◽  
Operating Characteristics ◽  
Neural Network Architecture ◽  
Receiver Operating Characteristics Curve ◽  
Gene Functions ◽  
Better Than

AbstractMotivationIsoforms are mRNAs produced from the same gene locus by alternative splicing and may have different functions. Although gene functions have been studied extensively, little is known about the specific functions of isoforms. Recently, some computational approaches based on multiple instance learning have been proposed to predict isoform functions from annotated gene functions and expression data, but their performance is far from being desirable primarily due to the lack of labeled training data. To improve the performance on this problem, we propose a novel deep learning method, DeepIsoFun, that combines multiple instance learning with domain adaptation. The latter technique helps to transfer the knowledge of gene functions to the prediction of isoform functions and provides additional labeled training data. Our model is trained on a deep neural network architecture so that it can adapt to different expression distributions associated with different gene ontology terms.ResultsWe evaluated the performance of DeepIsoFun on three expression datasets of human and mouse collected from SRA studies at different times. On each dataset, DeepIsoFun performed significantly better than the existing methods. In terms of area under the receiver operating characteristics curve, our method acquired at least 26% improvement and in terms of area under the precision-recall curve, it acquired at least 10% improvement over the state-of-the-art methods. In addition, we also study the divergence of the functions predicted by our method for isoforms from the same gene and the overall correlation between expression similarity and the similarity of predicted functions.Availability and implementationhttps://github.com/dls03/DeepIsoFun/Supplementary informationSupplementary data are available at Bioinformatics online.

Study of Self-Propelled Pufferfish Driven by Multiple Fins: A Comparison Between Rigid and Deformable Fins

2017 ◽  
Author(s):  
Ruoxin Li ◽  
Qing Xiao ◽  
Lijun Li ◽  
Hao Liu
Keyword(s):  
Underlying Mechanism ◽  
Operating Conditions ◽  
The Self ◽  
Fish Swimming ◽  
Propulsion Efficiency ◽  
Live Fish ◽  
Body Dynamics ◽  
Computational Fluid Dynamics Cfd ◽  
Multi Body ◽  
Better Than

In this work, we numerically studied the steady swimming of a pufferfish driven by the undulating motion of its dorsal, anal and caudal fins. The simulations are based on experimentally measured kinematics. To model the self-propelled fish swimming, a Computational Fluid Dynamics (CFD) tool was coupled with a Multi-Body-Dynamics (MBD) technique. It is widely accepted that deformable/flexible or undulating fins are better than rigid fins in terms of propulsion efficiency. To elucidate the underlying mechanism, we established an undulating fins model based on the kinematics of live fish, and conducted a simulation under the same operating conditions as rigid fins. The results presented here agree with this view by showing that the contribution of undulating fins to propulsion efficiency is significantly larger than that of rigid fins.

scholarly journals A Multi-User Schedule Method for Non-Orthogonal Multiple Access in 5G Heterogeneous Network

2019 ◽  
Vol 37 (2) ◽  
pp. 337-343
Author(s):  
Yi Xie
Keyword(s):  
Heterogeneous Network ◽  
Network Architecture ◽  
Multiple Access ◽  
Average Rate ◽  
Small Cells ◽  
Fifth Generation ◽  
Scheduling Method ◽  
New Type ◽  
Multiuser Scheduling ◽  
Better Than

Heterogeneous network is supposed to be the dominant network architecture of the fifth generation (5G) cellular network, which means small cells are overlaid on the macrocell. The beamforming (BF) and cell expansion are two important approaches to serve users in small cells. Furthermore, non-orthogonal multiple access (NOMA) is a new type of multiple access multiplexing which improves system performance without taking up extra spectrum resources. Therefore, it becomes one promising technique in 5G. In this paper, NOMA is applied in a 5G heterogeneous network with biased small cells. The BF strategy and the multiuser scheduling method are proposed. The main user in NOMA is scheduled inside the original coverage of the small cell while the side user is chosen from the biased expansion area. The BF strategy that is executed depends on the channel of main user. The multiuser scheduling method is to maximize the rate performance. The proposed method can provide performance benefits. Simulation results show that the proposed methods can be well applied in heterogeneous networks. The achieved performance gain is approximately twice better than traditional OMA and has 10% improvement to the stochastic schedule method. In addition, the average rate of cell edge users is improved.

scholarly journals Engagement of Occupational Safety and Health at Mahkamah’s Welding Worker Medan

2018 ◽  
Vol 1 (1) ◽  
pp. 48-53
Author(s):  
Eka Mahyuni ◽  
Kalsum ◽  
Muhammad Makmur Sinaga
Keyword(s):  
Occupational Safety ◽  
Occupational Safety And Health ◽  
The Self ◽  
Safety And Health ◽  
Before And After ◽  
Physical Risk ◽  
Result Show ◽  
Rest Time ◽  
Self Protection ◽  
Better Than

Welding worker was not the easy task because it has a very high physical risk and the process requires special skills and equipment to prevent accident exposed. This devotional activity is carried out in the welding industry at Jl. Mahkamah with two partners, namely CV. M. Nauli and CV. Cahaya. The aim of training activity made the worker able to analyze the hazards in the workplace so that it will be more careful in their work. The result show that the training could develop the worker to be aware about safety and health work patterns. In order to support the work in accordance with occupational safety and health standards, workers are also given pocket books that contain safety and health working methods and also given the self-protection of welding like welding clothes, welding gloves, welding mask, welding glasses and masks. Based on the evaluation of activities, it show that the worker has develop and always using the self protector in their work evenly. It build the good collaboration between them and they are could arrage the rest time with ergonomics relaxation in 5-10 minutes. The workshop station looks better than before and the workshop doing good house keeping before and after their work.

Coleridge’s Exile from the Social Contract, 1795–1829

2016 ◽  
Author(s):  
Zoe Beenstock
Keyword(s):  
Social Contract ◽  
The State ◽  
The Self ◽  
Breaking Point ◽  
General Will ◽  
Social Crisis ◽  
State Of Nature ◽  
The Social ◽  
The Individual ◽  
Better Than

Coleridge wrote frequently about Rousseau throughout his varied career. His early lectures and letters draw on Rousseau’s critique of luxury and frequently allude to the general will, depicting Rousseau as a Christ-like figure. Coleridge’s subsequent disappointment with Pantisocracy led him to reject Rousseau and the social contract. Comparing Rousseau to Luther in The Friend, Coleridge argues that Rousseau’s unhappiness arises from a conflict between an age of individualism and an ongoing need for community. According to Coleridge, poetry tolerates this conflict better than philosophy. In ‘Reflections on Having Left a Place of Retirement’ Coleridge suggests that social retreat offers illusory solace from war and social crisis. He critiques the state of nature, sympathy, and even religion for failing to balance the self with its environment. Thematically and formally The Rime of the Ancient Mariner explores this crisis in cohering systems. Through the mariner’s relationship to the albatross, the wedding that frames the poem, and episodes of the supernatural that disrupt the ballad form, Coleridge defines a breaking point between the individual and general wills.

Network Blueprint for Maximizing the Lifetime of Smart Devices in Low Power IoT Networks

2021 ◽  
Vol 13 (2) ◽  
pp. 21-38
Author(s):  
Sarwesh P. ◽  
K. Chandrasekaran ◽  
Thamizharasan S.
Keyword(s):  
Network Lifetime ◽  
Network Architecture ◽  
Data Transfer ◽  
Network Reliability ◽  
Smart Devices ◽  
Node Placement ◽  
The Standard Model ◽  
Device Lifetime ◽  
Placement Technique ◽  
Better Than

In the modern communication and computation era, internet of things (IoT) is developing as the key technology that satisfies the requirements of various applications. Prolonging device lifetime and maintaining network reliability is the evident objective for IoT network. Therefore, the authors come up with the network architecture that integrates node placement technique and routing technique. In the architecture, node placement is implemented by varying the density of nodes, by varying battery level of nodes, and by varying transmission range of nodes. Energy efficient and reliable path computation is addressed by routing technique. Therefore, enhancing the features of routing and node placement technique and integrating them together in network architecture can efficiently prolong the network lifetime. From the results, the authors observed that the proposed network architecture prolongs the network lifetime two times better than the standard model and also outperforms EQSR protocol and maintains the reliable data transfer.

Equity and the Right to Do Wrong

2020 ◽  
pp. 72-91
Author(s):  
Andrew S Gold
Keyword(s):  
The State ◽  
The Self ◽  
Role Of The State ◽  
The Right ◽  
The Relationship ◽  
Private Litigation ◽  
Better Than

This chapter discusses how the ‘stickler-enjoining’ account of equity has important limits. While many distinctive doctrines of equity can be understood to limit stickler behaviour, equity in fact often turns a blind eye to, and sometimes even enables, stickler behaviour. One can sort cases in which equity restrains sticklers from those in which it is indifferent to stickler behaviour if one attends to the role of the state in private litigation. Sometimes the state’s responsibilities require it to protect plaintiffs against sticklers. Other times, it requires it to protect the stickler, as a means, for example, of keeping as open as possible each person’s sphere of choices. Ultimately, the self-regarding account of equity sheds light on the question of the relationship between equity and justice: from the distinct perspective of the judgment, sometimes equitable justice is better than legal justice and sometimes legal justice is better than equitable justice.

Sign in / Sign up

Export Citation Format

Share Document