{"title":"Selection of Kernel Function for Least Squares Support Vector Machines in Downburst Wind Speed Forecasting","authors":"Zhou Li, Chunxiang Li","doi":"10.1109/ISCID.2018.10178","DOIUrl":"https://doi.org/10.1109/ISCID.2018.10178","url":null,"abstract":"In recent years, kernel machine learning algorithm (e.g. least square support vector machine, LSSVM) is a commonly method for time series prediction, as the main unit, the type of kernel function is very crucial. In this paper, a hybrid model of variational mode decomposition (VMD), particle swarm optimization (PSO) and LSSVM is proposed to forecast the downburst wind speed series. Linear kernel function, polynomial kernel function, radial basis function, Morlet wavelet kernel function, Mexican Hat wavelet kernel function and their combinations are selected to demonstrate the influence of different kernel functions on prediction results. The results indicate that the Morlet+RBF combined kernel function is considerably effective in enhancing the forecasting accuracy of the VMD-PSO-LSSVM model.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116646937","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}
{"title":"Usability Study on Railway Self-Service Terminal Interface for the Elderly","authors":"Zhizhao Luo, Ying Wang, Pingping Wang","doi":"10.1109/ISCID.2018.00043","DOIUrl":"https://doi.org/10.1109/ISCID.2018.00043","url":null,"abstract":"Based on the user performance test and the retrospective interview, this study carried out a usability evaluation of the railway self-service terminal interface. The results showed that the existing self-service terminal interface had many usability problems, especially for the elderly. First of all, the elderly had more difficulties than the young users in operating. The goal of operating should be clarified and all steps should be simplified to reduce the difficulty of the tasks. Secondly, with the decline of learning ability, the elderly needed guiding operation and timely feedback, which could help them to complete the task better. Thirdly, the elderly paid more attention to the function of the interface. The interface should be designed to be functional and simple. The habitual operation can help the elderly to be more confident and pleasant in their using experience.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"111 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117194155","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}
{"title":"Sponsors, Committees, and Reviewers","authors":"","doi":"10.1109/iscid.2018.10101","DOIUrl":"https://doi.org/10.1109/iscid.2018.10101","url":null,"abstract":"","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125841883","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}
{"title":"Identification of Functional States of the Cardiovascular System According to Flowmetry Data Using Machine Learning Methods","authors":"Konnova Natalia Sergeevna, Basarab Mikhail Alekseevich","doi":"10.1109/ISCID.2018.00057","DOIUrl":"https://doi.org/10.1109/ISCID.2018.00057","url":null,"abstract":"The aim of this work is to analyze the applicability of machine learning methods to the problems of diagnosing on the cardiovascular system data and develop a novel technique for automating this process to support decision making in cardiology. Here, the results obtained with the help of classifiers based on random decision forests are examined, and a proprietary numerical experiment is performed with the Doppler flowmetry data. Considerable attention is paid to data processing and reduction of the input vector dimension for analysis.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"25 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125893379","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}
{"title":"Virtual Simulation System for Finite Element Analysis Based on Web Server","authors":"Minghui Zhao, Jiaxin Han, Sai Ma","doi":"10.1109/ISCID.2018.10151","DOIUrl":"https://doi.org/10.1109/ISCID.2018.10151","url":null,"abstract":"The virtual simulation system for finite element analysis on web server is to apply virtual reality technology, virtual lab technology and web technology to system simulation by finite element method. It enables users to browse the experiment process on the internet, design characteristic experiments independently, and carry out finite element analysis results. System simulation is just a variety of related data, including the experimental operation process and the analysis of various results, in the form of text, sound, pictures, animation and others expressed.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"7 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123668457","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}
{"title":"Hybrid Image Edge Detection Algorithm Based on Fractional Differential and Canny Operator","authors":"Liangyan Wang, Xiubo Ma, Honghai Wang","doi":"10.1109/ISCID.2018.10149","DOIUrl":"https://doi.org/10.1109/ISCID.2018.10149","url":null,"abstract":"The Canny operator can detect most edges in an image very well, but some rich details cannot be detected. Edge detection operator based on fractional differential can detect more details such as textures in an image. A hybrid edge detection algorithm combining fractional differential and Canny operator is proposed to improve edge detection effect. Texture details are preserved and strong edges are removed from the result of the fractional differential algorithm. Then, the preserved texture details are merged into the result of Canny. The experimental results show that the hybrid algorithm not only retains the strong edges detected by Canny, but also increases the texture detail edges detected by the fractional differential algorithm, which is better than the other two algorithms.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116058314","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}
{"title":"Stock Price Trend Prediction Model Based on Deep Residual Network and Stock Price Graph","authors":"Heng Liu, Bowen Song","doi":"10.1109/ISCID.2018.10176","DOIUrl":"https://doi.org/10.1109/ISCID.2018.10176","url":null,"abstract":"Consider that people often use stock price graph to make decisions, this paper introduce a deep residual network (ResNet) model for prediction, using the stock price graph as input. The results show that the ResNet model has the average accuracy of 0.40, which is higher than the stochastic indicator of 0.33.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116399228","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}
{"title":"2018 11th International Symposium on Computational Intelligence and Design","authors":"","doi":"10.1109/iscid.2018.10097","DOIUrl":"https://doi.org/10.1109/iscid.2018.10097","url":null,"abstract":"","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129685194","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}
Chao Wang, Peng Liu, Peng Wang, Yalong Zhang, P. Gao
{"title":"Design of Dual-Band Dual-Port MIMO Antenna for LTE and WIMAX Applications","authors":"Chao Wang, Peng Liu, Peng Wang, Yalong Zhang, P. Gao","doi":"10.1109/ISCID.2018.00011","DOIUrl":"https://doi.org/10.1109/ISCID.2018.00011","url":null,"abstract":"A dual-band Multiple-Input-Multiple-Output (MIMO) antenna with two ports for Long-Term Evolution (LTE) and Worldwide Interoperability for Microwave Access (WIMAX) applications is presented. The antenna is derived from a non-radiation equivalent circuit with enough isolation achieved by meeting the decoupling requirement $mathbf{S}_{21}=0$. The circuit consisted of two pairs of symmetrical open-ended stubs connected by a transmission line realizes good intrinsic port isolation without employing complicated decoupling mechanisms. Ultimately, desired radiation can be obtained through impedance matching while port isolation unaffected almost. The antenna operates at 1.9-2.2 GHz and at 3.4-3.6 GHz and the measured minimum value of isolation between two ports is more than 20 dB in both operating frequencies.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114329773","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}
Mu Zhang, Qingyun Guo, Liang Li, Deng Pan, Sihong Zhou
{"title":"Art Deco Style Generation Method Based on Block Filling and Style Optimization","authors":"Mu Zhang, Qingyun Guo, Liang Li, Deng Pan, Sihong Zhou","doi":"10.1109/ISCID.2018.00060","DOIUrl":"https://doi.org/10.1109/ISCID.2018.00060","url":null,"abstract":"There are layered methods for generating art-works such as Pollock's drip style and Picasso stereoscopic painting. The disadvantage is that these layered methods do not consider the relationship between layers when they are implemented. In this paper, we propose a block-based method to generate the patterns of Art Deco style. Our method not only includes the details of generated artwork, but also makes up for the deficiencies of the layered methods. Having analyzed the characteristics of the Art Deco style, we divide Art Deco pattern into certain blocks and then fill with stylized blocky elements. All blocks are filled independently, then there would be many times of refilling after the first filling to smooth the generated images. Moreover, each block is semi-randomly drawn with featured lines. Finally, we implement our block-based algorithm, and demonstrate our approach can systematically generate patterns of Art Deco style.","PeriodicalId":334495,"journal":{"name":"2018 11th International Symposium on Computational Intelligence and Design (ISCID)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132420454","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}