An analysis of graph convolutional networks and recent datasets for visual question answering

Yusuf, Abdulganiyu Abdu; Feng, Chong; Mao, Xian-Ling

doi:10.1007/s10462-022-10151-2

Cited by 27 publications

(7 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Formulated after the Ministry of Education "about supporting the construction of a number of high school network education institute to conduct modern distance education pilot work of several opinions" clearly 31 pilot colleges and universities is highly autonomy, such as can be combined with their own situation work out standards for admission and enrolment numbers, can be combined with their own professional colleges and universities set up professional advantage of directory, and have the right to issue state-recognized academic qualifications and diplomas (Chen et al) [14]. On July 31, 2000, 31 pilot schools established the "Modern Distance Education Collaboration Group of Higher Education Institutions" in Beijing to strengthen exchanges and cooperation among pilot universities and promote the construction and sharing of educational resources (Yusuf et al) [15].…”

Section: Methodsmentioning

confidence: 99%

Design and Research of Artificial Intelligence in Multimedia Intelligent Question-Answering System and Self-Test System

Yao¹

2022

Advances in Multimedia

View full text Add to dashboard Cite

With the development of computer science and technology, online education is gradually accepted by more and more users, and has become one of the main teaching methods in schools under the influence of the epidemic in recent years. In addition, online vocational education or higher education has also captured the psychology of national learning, and online tutoring has gradually entered into nonstudent groups such as office workers. Although online education has the advantages of low cost and no limitation of time and space, it also has some problems, such as low classroom satisfaction, delayed feedback from students, and poor teaching effect. Therefore, this paper aims at this problem by combining the development background of foreign intelligent answer system and self-test system, and introducing it into our computer application, using artificial intelligence for natural language processing, aiming at online teaching students cannot get timely answer, and self-test questions intelligent system design. The final experimental results show that the system effectively alleviates the problems of students’ delay in answering questions and poor self-test effect, and increases the satisfaction of online classroom by 13%.

show abstract

Section: Methodsmentioning

confidence: 99%

Design and Research of Artificial Intelligence in Multimedia Intelligent Question-Answering System and Self-Test System

Yao¹

2022

Advances in Multimedia

View full text Add to dashboard Cite

show abstract

“…Attention mechanisms allow the model to focus on relevant visual and textual information, improving the accuracy of the answers. Additionally, graph-based methods have been employed to model the relationships between different modalities [33]. Graph structures can capture the dependencies and interactions between visual and textual elements, enabling more comprehensive understanding and reasoning [34].…”

Section: Methods Used In Multimodal Medical Question Answering Researchmentioning

confidence: 99%

Survey of Multimodal Medical Question Answering

Demirhan,

Zadrozny

2023

BioMedInformatics

View full text Add to dashboard Cite

Multimodal medical question answering (MMQA) is a vital area bridging healthcare and Artificial Intelligence (AI). This survey methodically examines the MMQA research published in recent years. We collect academic literature through Google Scholar, applying bibliometric analysis to the publications and datasets used in these studies. Our analysis uncovers the increasing interest in MMQA over time, with diverse domains such as natural language processing, computer vision, and large language models contributing to the research. The AI methods used in multimodal question answering in the medical domain are a prominent focus, accompanied by applicability of MMQA to the medical field. MMQA in the medical field has its unique challenges due to the sensitive nature of medicine as a science dealing with human health. The survey reveals MMQA research to be in an exploratory stage, discussing different methods, datasets, and potential business models. Future research is expected to focus on application development by big tech companies, such as MedPalm. The survey aims to provide insights into the current state of multimodal medical question answering, highlighting the growing interest from academia and industry. The identified research gaps and trends will guide future investigations and encourage collaborative efforts to advance this transformative field.

show abstract

“…The attack is induced in the proposed work through the FGSM method, which is effectively handled by Simplified Graph Convolutional Networks (SGCN) [41], a variant of applying CNN in the perspective of graphs. The SGCN avoids adversarial attacks by stacking multiple layers of firstorder spectral filters, followed by an utterly non-linear activation function.…”

Section: Preventing the Adversarial Attack On Sculpture Images Using ...mentioning

confidence: 99%

Study on Network Virtual Printing Sculpture Design using Artificial Intelligence

Wang

Sirivesmas

2023

Int. j. commun. netw. inf. secur.

View full text Add to dashboard Cite

Sculptures are visionaries of a country’s culture from time immemorial. Chinese sculptures hold an aesthetic value in the global market, catalysed by opening the country's gates. On the other hand, this paved the way for many duplicates and replicates of the original sculptures, defaming the entire artwork. This work proposes a defrauding model that deploys a Siamese-based Convolutional Neural Network (S-CNN) that effectively detects the mimicked sculpture images. Nevertheless, adversarial attacks are gaining momentum, compromising the deep learning models to make predictions for faked or forged images. The work uses a Simplified Graph Convolutional Network (SGCN) to misclassify the adversarial images generated by the Fast Gradient Sign Method (FGSM) to combat this attack. The model's training is done with adversarial images of the Imagenet dataset. By transfer learning, the model is rested for its efficacy in identifying the adversarial examples of the Chinese God images dataset. The results showed that the proposed model could detect the generated adversarial examples with a reasonable misclassification rate.

show abstract

An analysis of graph convolutional networks and recent datasets for visual question answering

Cited by 27 publications

References 72 publications

Design and Research of Artificial Intelligence in Multimedia Intelligent Question-Answering System and Self-Test System

Design and Research of Artificial Intelligence in Multimedia Intelligent Question-Answering System and Self-Test System

Survey of Multimodal Medical Question Answering

Study on Network Virtual Printing Sculpture Design using Artificial Intelligence

Contact Info

Product

Resources

About