2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)最新文献_第4页

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660437

Yu Xiong, Hao Jin, Tao Feng, R. Jia, Qing Zhang, C. Zhao

{"title":"Content Popularity Prediction Based on Integrated Features and Federated Learning","authors":"Yu Xiong, Hao Jin, Tao Feng, R. Jia, Qing Zhang, C. Zhao","doi":"10.1109/IC-NIDC54101.2021.9660437","DOIUrl":"https://doi.org/10.1109/IC-NIDC54101.2021.9660437","url":null,"abstract":"Mobile content service has been experiencing an explosive traffic growth in radio access networks. Most of data traffic is contributed by duplicated data transmission due to frequent download of popular contents requested by multiple mobile users. Proactive content caching has been an effective approach to alleviate traffic burden and improve user experience. Content popularity is an important factor that affects proactive caching. However, content popularity is usually unknown in advance. Therefore, predicting content popularity has become an important challenge on MEC oriented content management and orchestration. In this paper, in the networking scenario with one MBS and several SBSs, content popularity prediction is investigated based on integrated features of user and content. Considering user privacy and reducing transmission cost of uploading data for learning, a content popularity prediction algorithm is proposed based on integrated features and federated learning (PPFUC-FL). The proposed algorithm is evaluated with MovieLens dataset. Simulation results indicate that PPFUC-FL has good performance on precision accuracy compared with content popularity obtained from the real dataset.","PeriodicalId":264468,"journal":{"name":"2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126105966","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 Temporal Model for Small-Footprint Keyword Spotting 一个有效的时间模型用于小足迹关键字识别

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660544

Shuo Zhang, Tianhao Zhang, Songlu Chen, Feng Chen, Xu-Cheng Yin

{"title":"An Efficient Temporal Model for Small-Footprint Keyword Spotting","authors":"Shuo Zhang, Tianhao Zhang, Songlu Chen, Feng Chen, Xu-Cheng Yin","doi":"10.1109/IC-NIDC54101.2021.9660544","DOIUrl":"https://doi.org/10.1109/IC-NIDC54101.2021.9660544","url":null,"abstract":"Keyword spotting (KWS), as an essential part of human-computer interaction, is widely used in mobile device terminals. However, the hardware resources of these devices are usually limited, so running on these devices requires a small memory footprint. However, previous works still need massive parameters to achieve high performance. In this work, we propose a context-dependent and compact network for small-footprint KWS. Firstly, to reduce the running time, we apply a sub-sampling technique in which hidden activation values are calculated in a few time steps based on time delay neural network (TDNN). Secondly, to take full advantage of the global context information of the feature maps, we utilize a squeeze-and-excitation block to emphasize the most discriminating area and distinguish the speech and non-speech regions. Finally, we conduct extensive experiments with the publicly available Google Speech Commands dataset and the private Biaobei Chinese Speech Commands dataset. The experimental results on the public dataset verify that the classification error rate of our method reaches 3.56% with only 11K parameters and 322K multiplications, which achieves state-of-the-art performance with the fewest parameters and multiplications.","PeriodicalId":264468,"journal":{"name":"2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)","volume":"2016 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127484692","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

Research on Trajectory Tracking Algorithm Based on LSTM-UKF 基于LSTM-UKF的轨迹跟踪算法研究

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660592

Jing Zhang, Yingnian Wu, S. Jiao

引用次数: 1

An Online Solution for Secured Deep Learning Models Based on Crowd Sourced SGX 基于众包SGX的安全深度学习模型在线解决方案

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660566

Xuaner Wu, Konglin Zhu, Yuyang Peng, Lin Zhang

引用次数: 0

Voice Source Tracking Technology Based on Perceptual Hash of Fingerprint Image 基于指纹图像感知哈希的语音源跟踪技术

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660452

Lin Liu, Wenbo Guo, Peng Jia

引用次数: 0

Lightweight and Privacy-preserving Search over Encryption Blockchain 加密区块链上的轻量级和隐私保护搜索

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660565

Hongao Zhang, Zhen Yang, Haiyang Yu

引用次数: 1

Time Series Data Processing Algorithm in Deep Water Drilling 深水钻井时间序列数据处理算法

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660524

Ruidong Zhao, Zhiming Yin, Yonghua Li

{"title":"Time Series Data Processing Algorithm in Deep Water Drilling","authors":"Ruidong Zhao, Zhiming Yin, Yonghua Li","doi":"10.1109/IC-NIDC54101.2021.9660524","DOIUrl":"https://doi.org/10.1109/IC-NIDC54101.2021.9660524","url":null,"abstract":"The machine learning technology can be applied to the drilling data of deep water. The drilling data is an internally associated multi -dimensional time series data set, which contains a certain amount of abnormal data that greatly affect the data mining process. Because of the high dimension and large scale in drilling data set, existing detection algorithms perform poorly in drilling data sets. To achieve more effective outlier detection, we propose a data processing algorithm based on fusion outlier detection method. Firstly, Isolation Forest, Elliptic Envelope and Local Outlier Factor are used to detect outliers, judge the abnormal data in different conditions, which are weighted to judge and remove the outliers. Secondly, Savitzky-Golay(SG) filter is used to smooth the data, which eliminates the burrs in the data and get clean time series data. Finally, the proposed algorithm is tested in the real drilling data sets. Compared with existing methods, the experiments show that the proposed algorithm can achieve better performance, the RMSE and MAE values are 0.454 and 0.361, respectively.","PeriodicalId":264468,"journal":{"name":"2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)","volume":"38 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120917212","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

Robust Augmentations for Small Object Detection of Aerial Images 航空图像小目标检测的鲁棒增强方法

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660458

Weiyu Xiong, Zhanyu Ma, Yi-Zhe Song

{"title":"Robust Augmentations for Small Object Detection of Aerial Images","authors":"Weiyu Xiong, Zhanyu Ma, Yi-Zhe Song","doi":"10.1109/IC-NIDC54101.2021.9660458","DOIUrl":"https://doi.org/10.1109/IC-NIDC54101.2021.9660458","url":null,"abstract":"Object detection is one of the most fundamental but important computer vision tasks. However, small object detection remains an unsolved challenge due to insufficient detailed appearances and additional noises. Meanwhile, aerial images and intelligent transportation systems are under the restriction of difficulties such as dense object arrangement, a large number of small objects, and different perspectives, compared with natural images. To deal with these problems, an adversarial-like data augmentation training is proposed in this paper to narrow the data gap between remote sensing images and natural ones. The difficulty of the remote sensing object detection is verified and analyzed firstly by the classic single-stage anchor-based detector RetinaNet. Then, the multi-scale and data augmentations are introduced to alleviate the mismatch between general detector training and aerial images based on the anchor-free state-of-the-art (SOTA) model FCOS. Experiments on the remote sensing dataset, NWPU VHR10, demonstrate the quasi-antagonism data augmentation method improves the both anchor-based and anchor-free SOTA detectors for natural images with significant margins and shows the effectiveness on aerial images, especially small objects.","PeriodicalId":264468,"journal":{"name":"2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129910142","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

Multi-Scene Safety Helmet Detection with Multi-Scale Spatial Attention Feature 基于多尺度空间注意特征的多场景安全帽检测

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660519

Xinbo Ai, Cheng Chen, Yingjian Wang, Yanjun Guo

{"title":"Multi-Scene Safety Helmet Detection with Multi-Scale Spatial Attention Feature","authors":"Xinbo Ai, Cheng Chen, Yingjian Wang, Yanjun Guo","doi":"10.1109/IC-NIDC54101.2021.9660519","DOIUrl":"https://doi.org/10.1109/IC-NIDC54101.2021.9660519","url":null,"abstract":"The safety helmet detection system based on video surveillance has appeared in many smart construction sites. However, existing safety helmet detection algorithms have difficulty in detecting overlapping and small objects owing to the influence of complex environmental, and the features of safety helmet contain noise unrelated to the detection object which resulting poor detection performance. To address this problem, in this paper, a layer feature weighted module (LFWM) is added after different scale feature maps and getting the score matrix of the same size as the feature map. Finally, point-wise multiply is applied between score matrix and feature map for filtering the irrelevant noise. This method can highlight the local features of safety helmets in different feature maps and suppress the noise features that are not related to the detection object. Experiments show the proposed method can improve the safety helmet detection performance in different scenarios, and the (mean Average Precision) mAP improved by 4.19% compared with the original RetinaNet (ResNet50).","PeriodicalId":264468,"journal":{"name":"2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134013932","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 Multi-feature Emotion Classification Model Based on LDA Subject Target Words 基于LDA主题目标词的多特征情感分类模型

2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC) Pub Date : 2021-11-17 DOI: 10.1109/IC-NIDC54101.2021.9660505

Shike Shao, Cui Ding, Lei Li

{"title":"A Multi-feature Emotion Classification Model Based on LDA Subject Target Words","authors":"Shike Shao, Cui Ding, Lei Li","doi":"10.1109/IC-NIDC54101.2021.9660505","DOIUrl":"https://doi.org/10.1109/IC-NIDC54101.2021.9660505","url":null,"abstract":"In recent years, a variety of sentiment classification models based on deep neural networks have emerged. Most of the existing models are trained based on word embedding, or rely on expensive word-level annotation, or use sentence-level annotation only. However, some important linguistic phenomena and resources have not been fully studied. Aiming at the linguistic phenomenon that a sentence may have multiple sentiments and different target words may have different sentiments, this thesis proposes a multi-feature sentiments classification model based on LDA. The model automatically extracts the subject target words through LDA, screens the global sentiment features of sentences, extracts the local sentiment features of sentences with the external sentiment vocabulary, and integrates various features with the sentiments classification model. A series of experiments on three datasets show that the multi-feature model is effective. The introduction of LDA can not only reduce the demand for labeled target words, improve the accuracy of sentiments classification, but also more accurately analyze the internal emotional trend of public opinion events.","PeriodicalId":264468,"journal":{"name":"2021 7th IEEE International Conference on Network Intelligence and Digital Content (IC-NIDC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132386628","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