In P2P (Peer-to-Peer) video streaming systems using unstructured mesh, data scheduling is an important factor on system performance. An optimal data scheduling scheme should achieve two objectives ideally. The first objective is to optimize the perceived video quality of peers. The second objective is to maximize the network throughput, i.e., utilize the upload bandwidth of peers maximally. However, the optimized perceived video quality may not bring a maximized network throughput, and vice versa. In the paper, to better achieve the two objectives simultaneously, we formulate the data scheduling problem as a multi-objective optimization problem. To solve the multi-objective optimization problem, we propose a multi-objective particle swarm optimization data scheduling algorithm by encoding the peers' neighbors as the locations of the particles. Through simulations, we demonstrate the proposed algorithm outperforms other algorithms in terms of the perceived video quality and the utilization of peers' upload capacity.
{"title":"A multi-objective particle swarm optimization data scheduling algorithm for peer-to-peer video streaming","authors":"Pingshan Liu, Xiaoyi Xiong, Guimin Huang, Yimin Wen","doi":"10.1109/FSKD.2017.8393220","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393220","url":null,"abstract":"In P2P (Peer-to-Peer) video streaming systems using unstructured mesh, data scheduling is an important factor on system performance. An optimal data scheduling scheme should achieve two objectives ideally. The first objective is to optimize the perceived video quality of peers. The second objective is to maximize the network throughput, i.e., utilize the upload bandwidth of peers maximally. However, the optimized perceived video quality may not bring a maximized network throughput, and vice versa. In the paper, to better achieve the two objectives simultaneously, we formulate the data scheduling problem as a multi-objective optimization problem. To solve the multi-objective optimization problem, we propose a multi-objective particle swarm optimization data scheduling algorithm by encoding the peers' neighbors as the locations of the particles. Through simulations, we demonstrate the proposed algorithm outperforms other algorithms in terms of the perceived video quality and the utilization of peers' upload capacity.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"27 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":"131268687","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8392927
Min Li, Zhikang Xiang, Limei Zhang, Z. Lian, Liang Xiao
Segmentation of brain magnetic resonance imaging (MRI) images is greatly significant in neuroscience field. We propose a novel FCM method for segmentation of brain MRI images that makes full use of both the image intensity and spatial feature information. The proposed method can handle images having intensity inhomogeneity and noises by using the regularization that does not only consider the bias field but also takes neighborhood influence into account. Experiment indicates that the novel FCM method achieves more accurate and robust results in segmentation of brain MRI images compared to the expectation-maximization (EM) method and the conventional FCM method.
{"title":"Robust segmentation of brain MRI images using a novel fuzzy c-means clustering method","authors":"Min Li, Zhikang Xiang, Limei Zhang, Z. Lian, Liang Xiao","doi":"10.1109/FSKD.2017.8392927","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8392927","url":null,"abstract":"Segmentation of brain magnetic resonance imaging (MRI) images is greatly significant in neuroscience field. We propose a novel FCM method for segmentation of brain MRI images that makes full use of both the image intensity and spatial feature information. The proposed method can handle images having intensity inhomogeneity and noises by using the regularization that does not only consider the bias field but also takes neighborhood influence into account. Experiment indicates that the novel FCM method achieves more accurate and robust results in segmentation of brain MRI images compared to the expectation-maximization (EM) method and the conventional FCM method.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"13 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":"123884247","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8393296
Xiaofang Xu, Joao Amaro, Sam Caulfield, G. Falcão, D. Moloney
With the recent surge in popularity of Convolutional Neural Networks (CNNs), motivated by their significant performance in many classification and related tasks, a new challenge now needs to be addressed: how to accommodate CNNs in mobile devices, such as drones, smartphones, and similar low-power devices? In order to tackle this challenge we exploit the Vision Processing Unit (VPU) that combines dedicated CNN hardware blocks and very high power efficiency. The lack of readily available training data and memory requirements are two of the factors hindering the training and accuracy performance of 3D CNNs. In this paper, we propose a method for generating synthetic 3D point-clouds from realistic CAD scene models (based on the ModelNet10 dataset), in order to enrich the training process for volumetric CNNs. Furthermore, an efficient 3D volumetric object representation (VOLA) is employed. VOLA (Volumetric Accelerator) is a sexaquaternary (power-of-four subdivision) tree-based representation which allows for significant memory saving for volumetric data. Multiple CNN models were trained and the top performing model was ported to the Fathom Neural Compute Stick (NCS). Among the trained CNN models, the maximum test accuracy achieved is 91.3%. After deployment on the Fathom NCS, it takes 11ms (∼ 90 frames per second) to perform inference on each input volume, with a reported power requirement of 1.2W which leads to 75.75 inference per second per Watt.
{"title":"Classify 3D voxel based point-cloud using convolutional neural network on a neural compute stick","authors":"Xiaofang Xu, Joao Amaro, Sam Caulfield, G. Falcão, D. Moloney","doi":"10.1109/FSKD.2017.8393296","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393296","url":null,"abstract":"With the recent surge in popularity of Convolutional Neural Networks (CNNs), motivated by their significant performance in many classification and related tasks, a new challenge now needs to be addressed: how to accommodate CNNs in mobile devices, such as drones, smartphones, and similar low-power devices? In order to tackle this challenge we exploit the Vision Processing Unit (VPU) that combines dedicated CNN hardware blocks and very high power efficiency. The lack of readily available training data and memory requirements are two of the factors hindering the training and accuracy performance of 3D CNNs. In this paper, we propose a method for generating synthetic 3D point-clouds from realistic CAD scene models (based on the ModelNet10 dataset), in order to enrich the training process for volumetric CNNs. Furthermore, an efficient 3D volumetric object representation (VOLA) is employed. VOLA (Volumetric Accelerator) is a sexaquaternary (power-of-four subdivision) tree-based representation which allows for significant memory saving for volumetric data. Multiple CNN models were trained and the top performing model was ported to the Fathom Neural Compute Stick (NCS). Among the trained CNN models, the maximum test accuracy achieved is 91.3%. After deployment on the Fathom NCS, it takes 11ms (∼ 90 frames per second) to perform inference on each input volume, with a reported power requirement of 1.2W which leads to 75.75 inference per second per Watt.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"140 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":"123246836","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8392972
Bruno M. P. Moura, G. Schneider, A. Yamin, R. Reiser, M. Pilla
Scheduling tasks is a known NP-Hard problem. As grow the number of variables such as computational power and network metrics, even heuristic-based schedulers start to become overwhelmed by the underlying complexity. Computational Grids (CGs) are known for their heterogeneity of resources and interconnections, and as these resources may be deployed throughout the world, it is not possible to have a single, centralized, precise view of the system at any given moment. This paper provides a new approach with Fuzzy Type-2 logics to treat uncertainties and dynamic behavior for scheduling tasks in grid environments, named Int-fGrid. The scheduler was validated through simulations in the SimGrid framework with a model of the GridRS architecture. Our results show that the Fuzzy Type-2 approach provides makespans up to 18.5 times better than the best alternative tested scheduler XSufferage.
{"title":"Int-fGrid: A type-2 fuzzy approach for scheduling tasks of computational grids","authors":"Bruno M. P. Moura, G. Schneider, A. Yamin, R. Reiser, M. Pilla","doi":"10.1109/FSKD.2017.8392972","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8392972","url":null,"abstract":"Scheduling tasks is a known NP-Hard problem. As grow the number of variables such as computational power and network metrics, even heuristic-based schedulers start to become overwhelmed by the underlying complexity. Computational Grids (CGs) are known for their heterogeneity of resources and interconnections, and as these resources may be deployed throughout the world, it is not possible to have a single, centralized, precise view of the system at any given moment. This paper provides a new approach with Fuzzy Type-2 logics to treat uncertainties and dynamic behavior for scheduling tasks in grid environments, named Int-fGrid. The scheduler was validated through simulations in the SimGrid framework with a model of the GridRS architecture. Our results show that the Fuzzy Type-2 approach provides makespans up to 18.5 times better than the best alternative tested scheduler XSufferage.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"60 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":"115072938","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8393387
Ying Gao, Yanhai Gan, Junyu Dong, Lin Qi, Huiyu Zhou
The majority of studies on texture analysis focus on classification and generation, and few works concern perceptual similarity between textures, which is one of the fundamental problems in the field of texture analysis. Previous methods for perceptual similarity learning were mainly assisted by psychophysical experiments and computational feature extraction. However, the calculated similarity matrix is always seriously biased from human observation. In this paper, we propose a novel method for similarity prediction, which is based on convolutional neural networks (CNNs) and stacked sparse auto-encoder (SSAE). The experimental results show that the predicted similarity matrixes are more perceptually consistent with psychophysical experiments compared to other predicting methods.
{"title":"Perceptual texture similarity learning using deep neural networks","authors":"Ying Gao, Yanhai Gan, Junyu Dong, Lin Qi, Huiyu Zhou","doi":"10.1109/FSKD.2017.8393387","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393387","url":null,"abstract":"The majority of studies on texture analysis focus on classification and generation, and few works concern perceptual similarity between textures, which is one of the fundamental problems in the field of texture analysis. Previous methods for perceptual similarity learning were mainly assisted by psychophysical experiments and computational feature extraction. However, the calculated similarity matrix is always seriously biased from human observation. In this paper, we propose a novel method for similarity prediction, which is based on convolutional neural networks (CNNs) and stacked sparse auto-encoder (SSAE). The experimental results show that the predicted similarity matrixes are more perceptually consistent with psychophysical experiments compared to other predicting methods.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"37 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":"115352115","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8393071
Liang Bao, Shanshan Wu, Weizhao Chen, Zisheng Zhu, Fan Yi
Outlier detection has been a popular data mining task. However, there is a lack of serious study on outlier detection for trajectory data. In this paper, a trajectory outlier detection based on local outlier fraction algorithm (TODLOF) is proposed to detect outliers in the trajectory dataset based on the partition-and-detection framework. When partitioning a trajectory, a minimum description length principle (MDL) based method is adopted. The local outlier factor (LOF) is used as the basis for judging the outlier in the detection stage, which improve the accuracy of the anomaly detection. Finally, experiments were carried out, with hurricane trajectory data and animal migration data as inputs, to prove that this algorithm can detect anomaly trajectories efficiently. And an online version of this algorithm is also presented to meet the requirements of real-time application.
{"title":"Trajectory outlier detection based on partition-and-detection framework","authors":"Liang Bao, Shanshan Wu, Weizhao Chen, Zisheng Zhu, Fan Yi","doi":"10.1109/FSKD.2017.8393071","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393071","url":null,"abstract":"Outlier detection has been a popular data mining task. However, there is a lack of serious study on outlier detection for trajectory data. In this paper, a trajectory outlier detection based on local outlier fraction algorithm (TODLOF) is proposed to detect outliers in the trajectory dataset based on the partition-and-detection framework. When partitioning a trajectory, a minimum description length principle (MDL) based method is adopted. The local outlier factor (LOF) is used as the basis for judging the outlier in the detection stage, which improve the accuracy of the anomaly detection. Finally, experiments were carried out, with hurricane trajectory data and animal migration data as inputs, to prove that this algorithm can detect anomaly trajectories efficiently. And an online version of this algorithm is also presented to meet the requirements of real-time application.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"7 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":"121572257","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8392943
Fatigue Life Durability, Fatigue Behaviour
For composites static strength, fatigue damage and durability demonstrate a scatter factor of results larger than for isotropic materials. To characterize it the fuzzy set approach is proposed. Two different mechanical descriptions of fatigue life are used in order to describe the uncertainty and randomness of parameters characterizing the fatigue damage and finally the fatigue durability. The theoretical predictions representing the lower and upper bounds of a fatigue life are compared with experimental data. In general, the present analysis shows that the fuzzy set description allows us to take into account much more parameters than classical deterministic or statistical methods.
{"title":"Fuzzy approach to fatigue problems in composite materials and structures","authors":"Fatigue Life Durability, Fatigue Behaviour","doi":"10.1109/FSKD.2017.8392943","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8392943","url":null,"abstract":"For composites static strength, fatigue damage and durability demonstrate a scatter factor of results larger than for isotropic materials. To characterize it the fuzzy set approach is proposed. Two different mechanical descriptions of fatigue life are used in order to describe the uncertainty and randomness of parameters characterizing the fatigue damage and finally the fatigue durability. The theoretical predictions representing the lower and upper bounds of a fatigue life are compared with experimental data. In general, the present analysis shows that the fuzzy set description allows us to take into account much more parameters than classical deterministic or statistical methods.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"29 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":"114733054","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8393008
Lietao Fang, Hong Jiang, Shuqi Cui
As a classical data mining algorithm, decision tree has a wide range of application areas. Most of the researches on decision tree are based on ID3 and its derivative algorithms, which are all based on information entropy. In this paper, as the most important key point of the decision tree, the metric of the split attribute is studied. The mutual information is introduced into decision tree classification. The results show that the decision tree classification model based on mutual information is a better classifier. Compared with the ID3 classifier based on information entropy, it is verified that the accuracy of the decision tree algorithm based on mutual information has been greatly improved, and the construction of the classifier is more rapid.
{"title":"An improved decision tree algorithm based on mutual information","authors":"Lietao Fang, Hong Jiang, Shuqi Cui","doi":"10.1109/FSKD.2017.8393008","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393008","url":null,"abstract":"As a classical data mining algorithm, decision tree has a wide range of application areas. Most of the researches on decision tree are based on ID3 and its derivative algorithms, which are all based on information entropy. In this paper, as the most important key point of the decision tree, the metric of the split attribute is studied. The mutual information is introduced into decision tree classification. The results show that the decision tree classification model based on mutual information is a better classifier. Compared with the ID3 classifier based on information entropy, it is verified that the accuracy of the decision tree algorithm based on mutual information has been greatly improved, and the construction of the classifier is more rapid.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"160 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":"121086457","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8392934
Rong Li, Yanpeng Qu, Ansheng Deng, Q. Shen, C. Shang
Feature selection offers a crucial way to reduce the irrelevant and misleading features for a given problem, while retaining the underlying semantics of selected features. Whilst maintaining the quality of problem-solving (e.g., classification), a superior feature selection process should be reduce the number of attributes as much as possible. In this paper, a non-unique decision value (NDV), which is defined as the number of attribute values that can lead to non-unique decision values, is proposed to rapidly capture the uncertainty in the boundary region of a granular space. Also, as an evaluator of the selected feature subset, an NDV-based differentiation entropy (NDE) is introduced to implement a novel feature selection process. The experimental results demonstrate that the selected features by the proposed approach outperform those attained by other state-of-the-art feature selection methods, in respect of both the size of reduction and the classification accuracy.
{"title":"A new approach to exploring rough set boundary region for feature selection","authors":"Rong Li, Yanpeng Qu, Ansheng Deng, Q. Shen, C. Shang","doi":"10.1109/FSKD.2017.8392934","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8392934","url":null,"abstract":"Feature selection offers a crucial way to reduce the irrelevant and misleading features for a given problem, while retaining the underlying semantics of selected features. Whilst maintaining the quality of problem-solving (e.g., classification), a superior feature selection process should be reduce the number of attributes as much as possible. In this paper, a non-unique decision value (NDV), which is defined as the number of attribute values that can lead to non-unique decision values, is proposed to rapidly capture the uncertainty in the boundary region of a granular space. Also, as an evaluator of the selected feature subset, an NDV-based differentiation entropy (NDE) is introduced to implement a novel feature selection process. The experimental results demonstrate that the selected features by the proposed approach outperform those attained by other state-of-the-art feature selection methods, in respect of both the size of reduction and the classification accuracy.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"3 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":"126004309","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}
Pub Date : 2017-07-29DOI: 10.1109/FSKD.2017.8393010
Md. Asimuzzaman, P. D. Nath, F. Hossain, Asif Hossain, R. Rahman
Sentiment Analysis — also called Opinion Mining is the process that collects opinions through text forms to determine if the opinion being expressed is positive, negative, neutral etc. Our research has been done on Bangla Sentiment Analysis. There are few achievements in this field for Bangla. We put together our paper in the context of Fuzzy Sentiment Analysis. The semantic relations and various grammatical structures of these text forms increased the difficulty of determining the polarity of sentences. In this paper, we have used Adaptive Neuro-Fuzzy Inference System to predict the polarity of Bangla tweets and used fuzzy rules to represent semantic rules that are simple but greatly influence the actual polarity of the sentences.
{"title":"Sentiment analysis of bangla microblogs using adaptive neuro fuzzy system","authors":"Md. Asimuzzaman, P. D. Nath, F. Hossain, Asif Hossain, R. Rahman","doi":"10.1109/FSKD.2017.8393010","DOIUrl":"https://doi.org/10.1109/FSKD.2017.8393010","url":null,"abstract":"Sentiment Analysis — also called Opinion Mining is the process that collects opinions through text forms to determine if the opinion being expressed is positive, negative, neutral etc. Our research has been done on Bangla Sentiment Analysis. There are few achievements in this field for Bangla. We put together our paper in the context of Fuzzy Sentiment Analysis. The semantic relations and various grammatical structures of these text forms increased the difficulty of determining the polarity of sentences. In this paper, we have used Adaptive Neuro-Fuzzy Inference System to predict the polarity of Bangla tweets and used fuzzy rules to represent semantic rules that are simple but greatly influence the actual polarity of the sentences.","PeriodicalId":236093,"journal":{"name":"2017 13th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)","volume":"27 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":"125857242","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：