Abstract
A popular research topic in Graph Convolutional Networks (GCNs) is to speedup the training time of the network. The main bottleneck in training GCN is the exponentially growing of computations. In Cluster-GCN based on this fact that each node and its neighbors are usually grouped ...
Read More
A popular research topic in Graph Convolutional Networks (GCNs) is to speedup the training time of the network. The main bottleneck in training GCN is the exponentially growing of computations. In Cluster-GCN based on this fact that each node and its neighbors are usually grouped in the same cluster, considers the clustering structure of the graph, and expand each node's neighborhood within each cluster when training GCN.The main assumption of Cluster-GCN is the weak relation between clusters; which is not correct at all graphs. Here we extend their approach by overlapped clustering, instead of crisp clustering which is used in Cluster-GCN. This is achieved by allowing the marginal nodes to contribute to training in more than one cluster. The evaluation of the proposed method is investigated through the experiments on several benchmark datasets.The experimental results show that the proposed method is more efficient than Cluster-GCN, in average.
Mahmood Amintoosi
Abstract
A popular research topic in Graph Convolutional Networks (GCNs) is to speedup the training time of the network. The main bottleneck in training GCN is the exponentially growing of computations. In Cluster-GCN based on this fact that each node and its neighbors are usually grouped in the same ...
Read More
A popular research topic in Graph Convolutional Networks (GCNs) is to speedup the training time of the network. The main bottleneck in training GCN is the exponentially growing of computations. In Cluster-GCN based on this fact that each node and its neighbors are usually grouped in the same cluster, considers the clustering structure of the graph, and expand each node's neighborhood within each cluster when training GCN.The main assumption of Cluster-GCN is the weak relation between clusters; which is not correct at all graphs. Here we extend their approach by overlapped clustering, instead of crisp clustering which is used in Cluster-GCN. This is achieved by allowing the marginal nodes to contribute to training in more than one cluster. The evaluation of the proposed method is investigated through the experiments on several benchmark datasets.The experimental results show that the proposed method is more efficient than Cluster-GCN, in average.
Zohre Kiapasha; Iraj Mahdavi; Hamed Fazlollahtabar; Zahra Kiapasha
Abstract
Analyzing eyes performance is essential for effective functioning of human. Therefore, following their motion could help doctors to make quick and accurate diagnoses for disorders like Autism, schizophrenia, or attention deficit hyperactivity disorder. Recently, several studies investigated autism disorder ...
Read More
Analyzing eyes performance is essential for effective functioning of human. Therefore, following their motion could help doctors to make quick and accurate diagnoses for disorders like Autism, schizophrenia, or attention deficit hyperactivity disorder. Recently, several studies investigated autism disorder diagnosis and treatment. Meanwhile, various algorithms have been provided for eye tracking. In this paper, it is intended to identify diagnosis parameters of autism disorder using eye tracking concept. The eye tracking algorithm that has been used in this research is simple and sufficient accurate to appropriate function on videos with varying quality. The direct analysis of gaze and study of the interactions of its features are employed a useful method for diagnosis of autism. For this purpose, two separate groups of ordinary children and children with autism are considered. By tracking their eyes while watching television and performing the necessary analyses then their eye movements are compared and discussed. To identify pupils is face detection Viola Jones algorithm is implemented.
Gholamreza Hesamian; Mohammad Ghasem Akbari
Abstract
Here are many situations in real applications of decision making where we deal with uncertain conditions. Due to the different sources of uncertainty, since its original definition of fuzzy sets in 1965 \cite{zadeh1965}, different generalizations and extensions of fuzzy sets have been ...
Read More
Here are many situations in real applications of decision making where we deal with uncertain conditions. Due to the different sources of uncertainty, since its original definition of fuzzy sets in 1965 \cite{zadeh1965}, different generalizations and extensions of fuzzy sets have been introduced: Type-2 fuzzy sets \cite{6,13}, Intuitionistic fuzzy sets \cite{1}, fuzzy multi-sets \cite{37} and etc. However, in such cases, it is suitable for experts to provide their preferences or assessments by using linguistic information rather than quantitative values.
