检索结果-内蒙古大学图书馆

A Wide Evaluation of ChatGPT on affective computing Tasks

IEEE TRANSACTIONS ON affective computing 2024年第4期15卷 2204-2212页

作者： Amin, Mostafa M. Mao, Rui Cambria, Erik Schuller, Bjoern W. Univ Augsburg Chair Embedded Intelligence Hlth Careand Wellbeing D-86159 Augsburg Germany Tech Univ Munich CHI Chair Hlth Informat MRI D-81675 Munich Germany SYNCPILOT GmbH AI R&D Team D-86156 Augsburg Germany Nanyang Technol Univ Coll Comp & Data Sci Singapore 639798 Singapore Imperial Coll London GLAM Grp Language Audio & Mus London SW7 2BX England MDSI Munich Data Sci Inst D-85748 Munich Germany MCML Munich Ctr Machine Learning D-80538 Munich Germany

With the rise of foundation models, a new artificial intelligence paradigm has emerged, by simply using general purpose foundation models with prompting to solve problems instead of training a separate machine learning model for each problem. Such models have been shown to have emergent properties of solving problems that they were not initially trained on. The studies for the effectiveness of such models are still quite limited. In this work, we widely study the capabilities of the ChatGPT models, namely GPT-4 and GPT-3.5, on 13 affective computing problems, namely aspect extraction, aspect polarity classification, opinion extraction, sentiment analysis, sentiment intensity ranking, emotions intensity ranking, suicide tendency detection, toxicity detection, well-being assessment, engagement measurement, personality assessment, sarcasm detection, and subjectivity detection. We introduce a framework to evaluate the ChatGPT models on regression-based problems, such as intensity ranking problems, by modelling them as pairwise ranking classification. We compare ChatGPT against more traditional NLP methods, such as end-to-end recurrent neural networks and transformers. The results demonstrate the emergent abilities of the ChatGPT models on a wide range of affective computing problems, where GPT-3.5 and especially GPT-4 have shown strong performance on many problems, particularly the ones related to sentiment, emotions, or toxicity. The ChatGPT models fell short for problems with implicit signals, such as engagement measurement and subjectivity detection.

关键词： Chatbots Task analysis Computational modeling affective computing Sentiment analysis Analytical models Toxicology ChatGPT GPT-4 foundation models affective computing aspect-based sentiment analysis sentiment analysis sentiment intensity ranking emotions intensity ranking suicide tendency detection toxicity detection well-being assessment engagement measurement personality assessment sarcasm detection subjectivity detection

来源：评论

学校读者我要写书评

暂无评论

Unsupervised multi-modal representation learning for affective computing with multi-corpus wearable data

引用

JOURNAL OF AMBIENT INTELLIGENCE AND HUMANIZED computing 2021年第4期14卷 3199-3224页

作者： Ross, Kyle Hungler, Paul Etemad, Ali Queens Univ Dept Elect & Comp Engn Kingston ON K7L 3N6 Canada Queens Univ Fac Engn & Appl Sci Kingston ON Canada

There has been a growing focus on the use of artificial intelligence and machine learning for affective computing to further enhance user experience through emotion recognition. Typically, machine learning models used for affective computing are trained using manually extracted features from biological signals. Such features may not generalize well for large datasets. One approach to address this issue is to use fully supervised deep learning methods to learn latent representations. However, this method requires human supervision to label the data, which may be unavailable. In this work we propose an unsupervised framework for representation learning. The proposed framework utilizes two stacked convolutional autoencoders to learn latent representations from wearable electrocardiogram and electrodermal activity signals. The representations learned from this unsupervised framework are subsequently utilized within a random forest model to classify arousal. To validate this framework, an aggregation of the AMIGOS, ASCERTAIN, CLEAS, and MAHNOB-HCI datasets is created. The results of our proposed method are compared with other methods including convolutional neural networks, as well as methods that employ manual extraction of features. We show that our method outperforms current state-of-the-art results. The results show the wide-spread applicability for stacked convolutional autoencoders to be used for affective computing.

关键词： affective computing Autoencoder Arousal classification Biosignals Wearables

来源：评论

学校读者我要写书评

暂无评论

What affective computing Reveals about Autistic Children's Facial Expressions of Joy or Fear

引用

COMPUTER 2018年第6期51卷 7-8页

作者： Schuller, Bjoern Univ Augsburg Embedded Intelligence Healthcare & Wellbeing Augsburg Germany Imperial Coll London Machine Learning London England

This installment of Computer's series highlighting the work published in IEEE Computer Society journals comes from IEEE Transactions on affective computing

关键词： Spotlight On Transactions affective computing Autism Facial Expression Facial Recognition Emotion

来源：评论

学校读者我要写书评

暂无评论

Child and Youth affective computing-Challenge Accepted

引用

IEEE INTELLIGENT SYSTEMS 2022年第6期37卷 69-76页

作者： Loechner, Johanna W. Schuller, Bjoern Eberhard Karls Univ D-72076 Teubingen Germany Univ Augsburg D-86159 Augsburg Germany

affective computing has been shown effective and useful in a range of use cases by now, including human-computer interaction, emotionally intelligent tutoring, or depression monitoring. While these could be very useful to the younger among us- including in particular also earlier recognition of developmental disorders, usually research and even working demonstrators have been largely targeting an adult population. Only a few studies, including the first-ever competitive emotion challenge, were based on children's data. In times where fairness is a dominating topic in the world of artificial intelligence, it seems timely to widen up to include children and youth more broadly as a user group and beneficiaries of the promises affective computing holds. To best support according to algorithmic and technological development, here, we summarize the emotional development of this group over the years, which poses considerable challenges for automatic emotion recognition, generation, and processing engines. We also provide a view on the steps to be taken to best cope with these, including drifting target learning, broadening up on the "vocabulary" of affective states modeled, transfer, few-shot, zero-shot, reinforced, and life-long learning in affective computing besides trustability.

关键词： affective computing Vocabulary Target recognition Computational modeling Roads Sociology Pressing

来源：评论

学校读者我要写书评

暂无评论

Toward Explainable affective computing: A Review

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024年第10期35卷 13101-13121页

作者： Cortinas-Lorenzo, Karina Lacey, Gerard Trinity Coll Dublin Sch Comp Sci & Stat Dublin D02 W272 Ireland Maynooth Univ Dept Elect Engn Kildare W23 A3HY Ireland

affective computing has an unprecedented potential to change the way humans interact with technology. While the last decades have witnessed vast progress in the field, multimodal affective computing systems are generally black box by design. As affective systems start to be deployed in real-world scenarios, such as education or healthcare, a shift of focus toward improved transparency and interpretability is needed. In this context, how do we explain the output of affective computing models? and how to do so without limiting predictive performance? In this article, we review affective computing work from an explainable AI (XAI) perspective, collecting and synthesizing relevant papers into three major XAI approaches: premodel (applied before training), in-model (applied during training), and postmodel (applied after training). We present and discuss the most fundamental challenges in the field, namely, how to relate explanations back to multimodal and time-dependent data, how to integrate context and inductive biases into explanations using mechanisms such as attention, generative modeling, or graph-based methods, and how to capture intramodal and cross-modal interactions in post hoc explanations. While explainable affective computing is still nascent, existing methods are promising, contributing not only toward improved transparency but, in many cases, surpassing state-of-the-art results. Based on these findings, we explore directions for future research and discuss the importance of data-driven XAI and explanation goals, and explainee needs definition, as well as causability or the extent to which a given method leads to human understanding.

关键词： affective computing Training Data models Computational modeling Task analysis Terminology Predictive models explainable AI (XAI) multimodal machine learning review

来源：评论

学校读者我要写书评

暂无评论

FAformer: parallel Fourier-attention architectures benefits EEG-based affective computing with enhanced spatial information

引用

NEURAL computing & APPLICATIONS 2024年第8期36卷 3903-3919页

作者： Gao, Ziheng Huang, Jiajin Chen, Jianhui Zhou, Haiyan Beijing Univ Technol Fac Informat Technol Beijing 100124 Peoples R China Beijing Int Collaborat Base Brain Informat & Wisdo Beijing 100124 Peoples R China Minist Educ Engn Res Ctr Intelligent Percept & Autonomous Con Beijing 100124 Peoples R China Minist Educ Engn Res Ctr Digital Community Beijing 100124 Peoples R China

The balance of brain functional segregation (i.e., the process in specialized local subsystems) and integration (i.e., the process in global cooperation of the subsystems) is crucial for cognition in human beings, and many deep learning models have been used to evaluate the spatial information during EEG-based affective computing. However, acquiring the intrinsic spatial representation in the topology of EEG channels is still challenging. To further address the issue, we propose the FAformer to enhance spatial information in EEG signals with parallel-branch architectures based on a vision transformer (ViT). In the encoder, there is a branch that utilizes Adaptive Neural Fourier Operators (AFNO) to model global spatial patterns using the Fourier transform in the electrode channel dimension. The other branch utilizes multi-head self-attention (MSA) to explore the dependence of emotion on different channels, which is conducive to building key local networks. Additionally, a self-supervised learning (SSL) task of adaptive feature dissociation (AdaptiveFD) is developed to improve the distinctiveness of spatial features generated from the parallel branches and guarantee robustness in different subjects. FAformer achieves superior performance over the competitive models on the DREAMER and DEAP. Moreover, the rationality and hyperparameters analysis are conducted to demonstrate the effectiveness of the FAformer. Finally, the visualization of features reveals the spatial global connections and key local patterns during the deep learning process in FAformer, which benefits EEG-based affective computing.

关键词： affective computing Deep learning Vision transformer (ViT) Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Fuzzy model of dominance emotions in affective computing

引用

NEURAL computing & APPLICATIONS 2014年第6期25卷 1467-1477页

作者： Bakhtiyari, Kaveh Husain, Hafizah Univ Duisburg Essen Dept Comp & Cognit Sci Fac Engn D-47048 Duisburg North Rhine Wes Germany Univ Kebangsaan Malaysia Fac Engn & Built Environm Dept Elect Elect & Syst Engn Bangi 43600 Selangor Malaysia

To date, most of the human emotion recognition systems are intended to sense the emotions and their dominance individually. This paper discusses a fuzzy model for multilevel affective computing based on the dominance dimensional model of emotions. This model can detect any other possible emotions simultaneously at the time of recognition. One hundred and thirty volunteers from various countries with different cultural backgrounds were selected to record their emotional states. These volunteers have been selected from various races and different geographical locations. Twenty-seven different emotions with their strengths in a scale of 5 were questioned through a survey. Recorded emotions were analyzed with the other possible emotions and their levels of dominance to build the fuzzy model. Then this model was integrated into a fuzzy emotion recognition system using three input devices of mouse, keyboard and the touch screen display. Support vector machine classifier detected the other possible emotions of the users along with the directly sensed emotion. The binary system (non-fuzzy) sensed emotions with an incredible accuracy of 93 %. However, it only could sense limited emotions. By integrating this model, the system was able to detect more possible emotions at a time with slightly lower recognition accuracy of 86 %. The recorded false positive rates of this model for four emotions were measured at 16.7 %. The resulted accuracy and its false positive rate are among the top three accurate human emotion recognition (affective computing) systems.

关键词： Fuzzy emotion Dominance emotion Multilevel emotion Human-computer interaction affective computing

来源：评论

学校读者我要写书评

暂无评论

Toward a Taxonomy of affective computing

引用

INTERNATIONAL JOURNAL OF HUMAN-COMPUTER INTERACTION 2015年第11期31卷 761-768页

作者： Schwark, Jeremy D. Exponent Inc Human Factors Phoenix AZ 85027 USA

affective computing, which is the topic concerned with the design of emotionally intelligent machines, has been a growing area of interest in the field of human-computer interaction. Many of the current questions in the field will need to be answered in order for progress to be made toward a truly affective computing platform. The current article seeks to summarize much of the existing research in affective computing focused in three primary areas: affective generation, affective understanding, and application. A taxonomy of affective computing applications is then proposed in an effort to help organize the efforts of researchers and designers alike. This taxonomy contains a hierarchy of five tiers: the purpose or goal of the system, the level of integration, the method of affective understanding, the method of affective generation, and the platform of use. The article concludes by highlighting several key issues that the field of affective computing faces moving forward.

关键词： Taxonomy Fields affective computing Platform Social Hierarchy Designers computer human interaction Alike

来源：评论

学校读者我要写书评

暂无评论

affective computing: A Topic-Based SER Approach on Collaborative Discussions in Academic Setting 53

Affective Computing: A Topic-Based SER Approach on Collabora...

引用

53rd IEEE ASEE Frontiers in Education International Conference, FIE 2023

作者： Dehbozorgi, Nasrin Kunuku, Mourya Teja Kennesaw State University Department of Software Engineering MariettaGA United States Kennesaw State University Department of Computer Science MariettaGA United States

ISBN: (纸本)9798350336429

One of the biggest concerns in the modern day especially in the educational domain centers on the student's mental health. High rates of anxiety and depression have especially brought the attention of researchers in engineering education to apply affective computing to help with students' academic performance. It is known that a person's emotional states cause physiological and physical changes in the body. Emotions may impact facial expression, tone of speech, blood pressure, pulse, etc. Since visual and auditory signals are two variables that can be measured without the need to attach any physical device to the individuals, they are most studied in this field. Speech in particular has been known as a means that transfers much information about the mental and emotional states of the person. Speech Emotion Recognition (SER) is a growing field that has been applied in several domains including engineering education. Recent advancements in AI, Natural Language Understanding (NLU), and Large Language Models (LLM) have significantly streamlined this line of research. In this work which is a continuation of our prior work, we propose a speech analysis model that extracts both the emotions and topics from verbal discussions in a computer science classroom to understand if the expressed emotions were mostly about the course related topics or not. The goal of this research is to develop a tool that helps educators gain insights into the students' emotional states in teamwork and also understand the context of their conversations. We further analyze if the expressed emotions in the verbal class discussions are mostly about the course content or other subjects outside class setting. To expand the emotion analysis module we added a new layer to our developed pipeline by passing the speech data into the ChatGPT API to generate summarized scripts and extract additional classes of emotion. The preliminary results from this study are promising, indicating the potential value of th

关键词： affective computing ChatGPT API Engineering education Large Language Models (LLM) NLP Speech Emotion Recognition (SER) Teamwork Topic modeling

来源：评论

学校读者我要写书评

暂无评论

Opacity, Transparency, and the Ethics of affective computing

引用

IEEE TRANSACTIONS ON affective computing 2024年第1期15卷 4-17页

作者： Kumar, Manohar Aijaz, Aisha Chattar, Omkar Shukla, Jainendra Mutharaju, Raghava Indraprastha Inst Informat Technol Delhi 110020 India IIIT Comp Sci & Engn Dept Delhi 110020 India

Human opacity is the intrinsic quality of unknowability of human beings with respect to machines. The descriptive relationship between humans and machines, which captures how much information one can gather about the other, can be explicated using an opacity-transparency relationship. This relationship allows us to describe and normatively evaluate a spectrum of opacity where humans and machines may be either opaque or transparent. In this paper, we argue that the advent of affective computing (AC) has begun to shift the ideal position of humans on this spectrum towards greater transparency, while much of this technology is shifting towards opacity. We explore the implications of this shift with regard to the affective information of humans and how the threat to human opacity by AC systems has various adverse repercussions, such as infringement of one's autonomy, deception, manipulation, and increased anxiety. There are also distributive consequences that expose vulnerable groups to unjustified burdens and reduce them to mere profiles. We further provide an assessment of current AC technology, which follows the descriptive relationship between humans and machines from the lens of opacity and transparency. Finally, we foresee and address three possible objections to our claims. These are the beneficence of AC systems, their relation to privacy, and their restrictive capacity to capture human affects. Through these arguments, the paper aims to bring attention to the ontological relationship between humans and machines from the perspective of opacity and transparency while emphasizing on the gravity of the ethical concerns raised by their threat to human opacity.

关键词： Ethics affective computing Privacy Security Safety Resists Face recognition opacity transparency ethics autonomy distributive justice manipulation

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：