2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)最新文献_第8页

Fast variable-frame-rate decoding of speech recognition based on deep neural networks 基于深度神经网络的语音识别快速变帧率解码

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393381

Ge Zhang, Pengyuan Zhang, Jielin Pan, Yonghong Yan

{"title":"Fast variable-frame-rate decoding of speech recognition based on deep neural networks","authors":"Ge Zhang, Pengyuan Zhang, Jielin Pan, Yonghong Yan","doi":"10.1109/FSKD.2017.8393381","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393381","url":null,"abstract":"Deep neural networks (DNN) have recently shown impressive performance as acoustic models for large vocabulary continuous speech recognition (LVCSR) tasks. Typically, the frame shift of the output of neural networks is much shorter than the average length of the modeling units, so the posterior vectors of neighbouring frames are likely to be similar. The similarity, together with the better discrimination of neural networks than typical acoustic models, shows a possibility of removing frames of neural network outputs according to the distance of posterior vectors. Then, the computation costs of beam searching can be effectively reduced. Based on that, the paper introduces a novel variable-frame-rate decoding approach based on neural network computation that accelerates the beam searching for speech recognition with minor loss of accuracy. By computing the distances of posterior vectors and removing frames with a posterior vector similar to the previous frame, the approach can make use of redundant information between frames and do a much quicker beam searching. Experiments on LVCSR tasks show a 2.4-times speed up of decoding compared to the typical framewise decoding implementation.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128157478","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

Improving semantic role labeling using high-level classification in complex networks 在复杂网络中使用高级分类改进语义角色标注

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393113

M. Carneiro, J. Rosa, Qiusheng Zheng, Xiaoming Liu, Liang Zhao

{"title":"Improving semantic role labeling using high-level classification in complex networks","authors":"M. Carneiro, J. Rosa, Qiusheng Zheng, Xiaoming Liu, Liang Zhao","doi":"10.1109/FSKD.2017.8393113","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393113","url":null,"abstract":"While traditional supervised learning methods perform classification based only on the physical features of the data (e.g. distribution, similarity or distance), the high-level classification is characterized by its ability to capture topological features of the input data by using complex network measures. Recent works have shown that a variety of patterns can be detected by combining both features of the data, although the physical features alone are unable to uncover them. In this article we investigate such a hybrid method for the Semantic Role Labeling (SRL) task, which consists of the identification and classification of arguments in a sentence with roles that indicate semantic relations between an event and its participants. Due to its potential to improve many other natural language processing tasks, such as information extraction and plagiarism detection to name a few, we consider the SRL task over a Brazilian Portuguese corpus named PropBank-br, which was built with texts from Brazilian newspapers. Such a corpus represents a challenging classification problem as it suffers with the scarcity of annotated data and very imbalanced distributions, like the majority of non-English corpus. Experiments were performed considering the argument classification task over the whole corpus and, specifically, over the most frequent verbs. Results in the verb-specific scenario revealed that the high-level system is able to obtain a considerable gain in terms of predictive performance, even over a state-of-the-art algorithm for SRL.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114487045","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

A data fusion equipment monitoring method based on fuzzy set and improved D-S evidence theory 一种基于模糊集和改进D-S证据理论的数据融合设备监测方法

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8392912

Han Ding, Ruichun Hou, Xiangqian Ding

{"title":"A data fusion equipment monitoring method based on fuzzy set and improved D-S evidence theory","authors":"Han Ding, Ruichun Hou, Xiangqian Ding","doi":"10.1109/FSKD.2017.8392912","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8392912","url":null,"abstract":"In order to solve data problems with redundant, conflict and uncertainty in monitoring large mechanical equipment, a data fusion equipment monitoring method is proposed through the combination of fuzzy set and improved D-S evidence theory. Firstly, a recognition framework is built based on the actual situation of the equipment. Then, the likelihood of the attributes is calculated according to the fuzzy set membership function and the sensor's observation function, and the likelihood is used to determine the basic belief assignment function value of the attributes. Finally, the data fusion is carried out using the weight-based D-S's combination rule, and the state of equipment can be derived from the data fusion results. A simulation of monitoring method with application to the ozone generator is carried out using the proposed method, the results show that the accuracy of the proposed method is proved, and the uncertainty of the results is obviously reduced comparing with classic analyzing methods, which concludes that the proposed method has a practical significance in monitoring the state of equipment.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114813228","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

A data selection framework for k-means algorithm to mine high precision clusters 基于k均值算法的高精度聚类数据选择框架

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393013

Zhengzheng Lou, Chaoyang Zhang

{"title":"A data selection framework for k-means algorithm to mine high precision clusters","authors":"Zhengzheng Lou, Chaoyang Zhang","doi":"10.1109/FSKD.2017.8393013","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393013","url":null,"abstract":"Traditional clustering algorithms employ all the data items to learn the cluster patterns. However, in real-world applications, some data show clear coherent behaviour and can be summarized well, while some data present weak tendencies to be assigned to any particular pattern. For such situation, this paper presents a data selection framework for K-Means algorithm to get high precision clusters from the data collection. It differs from traditional k-means-type algorithms in three respects. First, in the cluster learning process, we take the changed value of cluster's Bregman Information, which is generated by merging one data item into the potential clusters, as the measure of data item's clustering tendency. Second, only data items with strong clustering tendencies, that is the changed value of cluster's Bregman Information is less than the predefined radius, are selected to learn the cluster patterns, while the remaining data points are ignored and belong to no cluster. The clustering is non-exhaustive. Third, the radius of the clusters can be changed in the learning process. It is a dynamic learning framework. Experiments on synthetic, document and image data show the effectiveness of the proposed algorithm.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"334 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115879939","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

Risk index assessment for urban natural gas pipeline leakage based on artificial neural network 基于人工神经网络的城市天然气管道泄漏风险指标评价

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8392945

Yang Zhou, Zhengwei Wu

引用次数: 5

A multidimensional time-series association rules algorithm based on spark 基于spark的多维时间序列关联规则算法

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393066

Dongyue Liu, Bin Wu, Chao Gu, Yan Ma, Bai Wang

{"title":"A multidimensional time-series association rules algorithm based on spark","authors":"Dongyue Liu, Bin Wu, Chao Gu, Yan Ma, Bai Wang","doi":"10.1109/FSKD.2017.8393066","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393066","url":null,"abstract":"Fault prediction of industrial systems has been a hot research orientation in recent years, which allows the maintainer to know the operation conditions and the fault to be occurred in advance so as to reduce the risk of fault and the economic loss. In general, association rules learning is one of the most effective methods in fault prediction of industrial systems, however, traditional methods based on association rules are not suitable for sparse time-series data that are common in industrial systems (e.g. transmission line data). Although some methods based on clustering to reduce the dimension of data have been proposed, these methods may lose some of the key rules from the dataset and reduce the effectiveness of the results. In order to solve the problem, we propose a novel algorithm called Multidimensional Time-series Association Rules(MTAR) in this paper, which can fully utilize the information and find out more valuable rules from multidimensional time-series data. Meanwhile, we implement the parallelization of the algorithm based on the parallel computing framework Spark, which can improve the performance of the algorithm greatly. Experiments are conducted on the transmission line dataset in Power Grid System to show the effectiveness and the efficiency of the proposed approach.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125346576","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

A feature optimization algorithm of concept similarity based on Chinese wikipedia 基于中文维基百科的概念相似度特征优化算法

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393108

Xiaofei Chang, Lei Liu, Mengtao Sun, Yalu Jia, Chunxia Zhang

引用次数: 1

A parallel clustering and test partitioning techniques based mining trajectory algorithm for moving objects 一种基于并行聚类和测试划分技术的运动目标轨迹挖掘算法

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393312

Qian He, Yiting Chen, Qinghe Dong, Dongsheng Cheng

引用次数: 1

Information consistency in information-centric networking 信息中心组网中的信息一致性

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393228

Lan Shi, Peng Yin, Jianhui Lv, Ying Zhao

引用次数: 0

Multi-level image thresholding based on improved fireworks algorithm 基于改进烟花算法的多级图像阈值分割

2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD) Pub Date : 2017-07-29 DOI: 10.1109/FSKD.2017.8393414

Miao Ma, Weige Zheng, Jie Wu, Kaifang Yang, Min Guo

引用次数: 0