2017 Intelligent Systems Conference (IntelliSys)最新文献_第8页

Predicting lung cancer survivability using ensemble learning methods 使用集成学习方法预测肺癌存活率

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324368

Ali Safiyari, R. Javidan

{"title":"Predicting lung cancer survivability using ensemble learning methods","authors":"Ali Safiyari, R. Javidan","doi":"10.1109/INTELLISYS.2017.8324368","DOIUrl":"https://doi.org/10.1109/INTELLISYS.2017.8324368","url":null,"abstract":"Ensemble methods are powerful techniques used in machine learning to improve the prediction accuracy of classifier learning systems. In this study, different ensemble learning methods for lung cancer survival prediction were evaluated on the Surveillance, Epidemiology, and End Results (SEER) dataset. Data were preprocessed in several steps before applying classification models. Five popular ensemble methods, Bagging, Dagging, AdaBoost, MultiBoosting and Random SubSpace, and eight classification algorithms, RIPPER, Decision Stump, Simple Cart, C4.5, SMO, Logistic Regression, Bayes Net and Random Forest, as base classifiers were evaluated for lung cancer survival prediction. Then, risk of mortality after 5 years of diagnosis has been estimated. The prediction performance is measured in terms of accuracy and area under ROC curve (AUC). AdaBoost Algorithm had the best efficiency in increasing base classifiers performance in comparison to other four ensemble methods. It increased the accuracy of RIPPER from 88.88% to 88.98%, the accuracy of decision stump algorithm from 81.21% to 87.67% and the accuracy of SMO algorithm from 83.41% to 87.16%. AdaBoost algorithm also increased the AUC of RIPPER from 91.5% to 94.9%, the AUC of decision stump algorithm from 81.2% to 93.9%, the AUC of J48 algorithm from 94.1% to 94.9% and the AUC of SMO algorithm from 50.0% to 92.1%. Random subspace algorithm was the worst method in comparison to other ensemble techniques used in this study. The results empirically showed that ensemble methods are able to evaluate the performance of their base classifiers and they are appropriate methods for analysis of cancer survival.","PeriodicalId":131825,"journal":{"name":"2017 Intelligent Systems Conference (IntelliSys)","volume":"186 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114744855","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: 20

SQbSN: JPEG2000 scalar quantizer implemented by means a statistical normalization JPEG2000标量量化器通过统计归一化的方式实现

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324353

Jesús Jaime Moreno-Escobar, O. Morales-Matamoros, Ricardo Tejeida-Padilla

{"title":"SQbSN: JPEG2000 scalar quantizer implemented by means a statistical normalization","authors":"Jesús Jaime Moreno-Escobar, O. Morales-Matamoros, Ricardo Tejeida-Padilla","doi":"10.1109/INTELLISYS.2017.8324353","DOIUrl":"https://doi.org/10.1109/INTELLISYS.2017.8324353","url":null,"abstract":"In this work we present an algorithm for quantizing wavelet coefficients taking in to account to be used by any image compression system that use wavelet transformation, we particularly implemented it in JPEG2000. In the literature is well-know that any wavelet-base compression encoder considers three stages: 1) Conversion of pixel into the frequency domain in order to obtain coefficients; 2) Scalar Quantization; and 3) Coding of the wavelet quantized coefficients. By one hand is important to highlight that just Scalar Quantization stage is responsible for degraded or maintaining precision of a certain coefficient, thus if the accuracy of inverse quantized coefficient is reduced we can consider a lossy reconstruction otherwise when inverse quantized coefficient is perfectly reconstructed we consider a lossless reconstruction with Scalar Quantization equal to one. By the other hand, we modify the state-of-the-art and classical JPEG2000 dead-zone scalar quantization modifying the process with a Statistical Normalization or better known as Z-Scores. We can define a Z-score as a expression in terms of standard deviations distributed along their mean. Thus, Z-scores can be defined as distribution with μ = 0 and σ2 = 0, in this way visual redundancies of the image are incremented, which gives as a result a lower compression rate.","PeriodicalId":131825,"journal":{"name":"2017 Intelligent Systems Conference (IntelliSys)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114797867","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: 0

Intra-class complex object shape representation towards high resolution 面向高分辨率的类内复杂对象形状表示

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324278

Xinhan Di

引用次数: 0

Generalized relevance vector machine 广义相关向量机

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324361

Yuheng Jia, S. Kwong, Wenhui Wu, Wei Gao, Ran Wang

{"title":"Generalized relevance vector machine","authors":"Yuheng Jia, S. Kwong, Wenhui Wu, Wei Gao, Ran Wang","doi":"10.1109/INTELLISYS.2017.8324361","DOIUrl":"https://doi.org/10.1109/INTELLISYS.2017.8324361","url":null,"abstract":"This paper considers the generalized version of relevance vector machine (RVM), which is a sparse Bayesian kernel machine for classification and ordinary regression. Generalized RVM (GRVM) follows the work of generalized linear model (GLM), which is a natural generalization of ordinary linear regression model and shares a common approach to estimate the parameters. GRVM inherits the advantages of GLM, i.e., unified model structure, same training algorithm, and convenient task-specific model design. It also inherits the advantages of RVM, i.e., probabilistic output, extremely sparse solution, hyperparameter auto-estimation. Besides, GRVM extends RVM to a wider range of learning tasks beyond classification and ordinary regression by assuming that the conditional output belongs to exponential family distribution (EFD). Since EFD results in inference intractable problem in Bayesian analysis, in this paper, Laplace approximation is adopted to solve this problem, which is a common approach in Bayesian inference. Further, several task-specific models are designed based on GRVM including models for ordinary regression, count data regression, classification, ordinal regression, etc. Besides, the relationship between GRVM and traditional RVM models are discussed. Finally, experimental results show the efficiency of the proposed GRVM model.","PeriodicalId":131825,"journal":{"name":"2017 Intelligent Systems Conference (IntelliSys)","volume":"105 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123851560","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: 2

Using confidence factors to share control between a mobile robot tele-operater and ultrasonic sensors 利用置信度因子在移动机器人远程操作者和超声波传感器之间共享控制

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324255

D. Sanders, Benjamin John Sanders, A. Gegov, D. Ndzi

引用次数: 10

Partially squeezing the resources of the web of data towards applications 将数据网络的部分资源压缩到应用程序中

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324319

Ángel Castellanos, E. D. De Luca, Juan Cigarán, A. G. Serrano

{"title":"Partially squeezing the resources of the web of data towards applications","authors":"Ángel Castellanos, E. D. De Luca, Juan Cigarán, A. G. Serrano","doi":"10.1109/INTELLISYS.2017.8324319","DOIUrl":"https://doi.org/10.1109/INTELLISYS.2017.8324319","url":null,"abstract":"The Web of Data (WOD) contains a large amount of formalized and interconnected data, offering a valuable help for experimental tasks requiring an accurate data representation. However, the practical application of such data is often limited by the complexity when it comes to extracting the necessary information, mainly because of the lack of a proper structure and organization in the WOD-resources. The (re)organization of the knowledge contained in these resources might facilitate the identification of the necessary information and, consequently, limit the problems arising in their practical application. In this context, this paper proposes the application of Formal Concept Analysis (FCA) to create a concept-based abstraction that better organizes the knowledge contained in the WOD-resources. In order to test, to what extent this enhanced organization is able to improve the data representation process, the obtained FCA models will be tested in a practical application to represent a set of Twitter contents in a specific task: the Topic Detection task at Replab 2013. The results demonstrate that the better data representation obtained through FCA improves the operation of the topic detection process, outperforming state-of-the-art results.","PeriodicalId":131825,"journal":{"name":"2017 Intelligent Systems Conference (IntelliSys)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121052720","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

An efficient rank based Arabic root extractor 一个高效的基于等级的阿拉伯语词根提取器

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324232

M. El-Defrawy, Nahla A. Belal, Y. El-Sonbaty

引用次数: 1

Pervasive audio playback in cyber-physical environments 网络物理环境中普遍存在的音频回放

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324346

Yannick Körber, M. Feld, Tim Schwartz

引用次数: 3

Stacking approach for CNN transfer learning ensemble for remote sensing imagery 遥感图像CNN迁移学习集成的叠加方法

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324356

O. Korzh, Gregory Cook, T. Andersen, Edoardo Serra

{"title":"Stacking approach for CNN transfer learning ensemble for remote sensing imagery","authors":"O. Korzh, Gregory Cook, T. Andersen, Edoardo Serra","doi":"10.1109/INTELLISYS.2017.8324356","DOIUrl":"https://doi.org/10.1109/INTELLISYS.2017.8324356","url":null,"abstract":"In this paper we propose a stacking approach for Convolutional Neural Network (CNN) transfer learning ensemble for remote sensing imagery, in particular for the task of scene classification. We propose to use a combination of features produced by an ensemble of CNNs as one feature vector for classification. At the same time the original data set can be processed with different up-sampling and image enhancement methods and then used to obtain more features from pretrained networks. We investigate both fine-tuning and non fine-tuning approaches for transfer learning. We have selected Brazilian Coffee Scenes data set as a benchmark to measure the classification accuracy. Proposed method in case of a non fine-tuned model shows 89.18% classification accuracy. For a fine-tuned model the best classification rate is 96.11%. We analyzed how networks that have appeared recently (VGG-19 and SqueezeNet), can be applied to the task of transfer learning for remote sensing. Also we describe a method of decreasing processing time and memory consumption while preserving classification accuracy by using feature selection based on feature importance.","PeriodicalId":131825,"journal":{"name":"2017 Intelligent Systems Conference (IntelliSys)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132175133","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: 9

Data mining and model-predictive approach for blast furnace thermal control 高炉热控制的数据挖掘与模型预测方法

2017 Intelligent Systems Conference (IntelliSys) Pub Date : 2017-09-01 DOI: 10.1109/INTELLISYS.2017.8324364

D. Shnayder, L. Kazarinov, T. Barbasova, A. Lipatnikov

引用次数: 4