2018 21st International Conference of Computer and Information Technology (ICCIT)最新文献

筛选

英文中文

DEEPGONET: Multi-Label Prediction of GO Annotation for Protein from Sequence Using Cascaded Convolutional and Recurrent Network DEEPGONET:基于级联卷积和递归网络的蛋白质序列GO注释多标签预测

2018 21st International Conference of Computer and Information Technology (ICCIT) Pub Date : 2018-10-31 DOI: 10.1109/ICCITECHN.2018.8631921

S. M. S. Islam, M. Hasan

{"title":"DEEPGONET: Multi-Label Prediction of GO Annotation for Protein from Sequence Using Cascaded Convolutional and Recurrent Network","authors":"S. M. S. Islam, M. Hasan","doi":"10.1109/ICCITECHN.2018.8631921","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631921","url":null,"abstract":"The present gap between the amount of available protein sequence due to the development of next generation sequencing technology (NGS) and slow and expensive experimental extraction of useful information, like annotation of protein sequence in different functional aspects, is ever widening. The gap can be reduced by employing automatic function prediction (AFP) approaches. Gene Ontology (GO), comprising of more than 40, 000 classes, defines three aspects of protein function named Biological Process (BP), Cellular Component (CC), and Molecular Function (MF). The availability of multiple functions of a single protein has rendered the automatic function prediction a large-scale, multi-class, and a multi-label task. In this paper, we present DEEPGONET, a novel cascaded convolutional and recurrent neural network, to predict the top-level hierarchy of GO ontology. The network takes the primary sequence of protein as input, making it more useful than other prevailing state-of-the-art deep learning based methods with multi-modal input, which are less applicable for proteins where only primary sequence is available. All the predictions of different protein functions of our network are performed by the same architecture, a proof of better generalization as demonstrated by promising performance on a variety of organisms while trained on Homo sapiens only. The task has been made possible by efficient exploration of vast output space by leveraging hierarchical relationship among GO classes. The promising performance of our model makes it a potential avenue for directing experimental protein functions exploration efficiently by vastly eliminating possible routes which is done by the exploring only the suggested routes from our model. Our proposed model is also very simple and efficient in terms of computational time and space compared to other architectures in literature.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115381576","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Total Recall: Understanding Traffic Signs Using Deep Convolutional Neural Network 全面回忆:使用深度卷积神经网络理解交通标志

2018 21st International Conference of Computer and Information Technology (ICCIT) Pub Date : 2018-08-30 DOI: 10.1109/ICCITECHN.2018.8631925

Sourajit Saha, Sharif Amit Kamran, A. Sabbir

{"title":"Total Recall: Understanding Traffic Signs Using Deep Convolutional Neural Network","authors":"Sourajit Saha, Sharif Amit Kamran, A. Sabbir","doi":"10.1109/ICCITECHN.2018.8631925","DOIUrl":"https://doi.org/10.1109/ICCITECHN.2018.8631925","url":null,"abstract":"Recognizing Traffic Signs using intelligent systems can drastically reduce the number of accidents happening worldwide. With the arrival of Self-driving cars it has become a staple challenge to solve the automatic recognition of Traffic and Hand-held signs in the major streets. Various machine learning techniques like Random Forest, SVM as well as deep learning models have been proposed for classifying traffic signs. Though they reach state-of-the-art performance on a particular dataset, yet fall short of tackling multiple Traffic Sign Recognition benchmarks. In this paper, we propose a novel and one-for-all architecture that aces multiple benchmarks with a better overall score than the state-of-the-art architectures. Our model is made of residual convolutional blocks with hierarchical dilated skip connections joined in steps. Intrinsically, our model achieves 99.33% Accuracy in German traffic sign recognition benchmark and 99.17% Accuracy in Belgian traffic sign classification benchmark, while classifying traffic signs in real time. Moreover, we propose a newly devised dilated residual learning representation technique which is very low in both memory and computational complexity.","PeriodicalId":355984,"journal":{"name":"2018 21st International Conference of Computer and Information Technology (ICCIT)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130499708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 8

首页上一页