基于注意力模型的多传感器人类活动识别

深度循环神经网络适用于处理时间序列数据, 然而循环神经网络特征提取能力差, 时间依赖关系挖掘不足。针对此问题, 提出了3种注意力机制和长短时记忆(LSTM)神经网络结合的模型用于人类活动识别问题, 并研究了3种注意力机制在不同数据集上单独及配合使用时对模型精度的影响。对于UCI_HAR数据集, 3种注意力LSTM模型准确率分别为94.13%、95.15%和94.81%,高于LSTM模型识别准确率93.2%。此外, 针对人类活动识别的传感器时间序列数据的标签特点, 提出将时间段分类任务转化为分割任务, 设计了2个基于分割任务的注意力门控循环单元(GRU)神经网络模型, Bahdanau注意力GRU模型在Skoda数据集和机会(Oppor)数据集准确率为84.61%和89.54%, 高于基准UNet模型的70.40%和88.51%。     

基于改进双流卷积递归神经网络的RGB-D物体识别方法

为了提高基于图像的物体识别准确率,提出一种改进双流卷积递归神经网络的RGB-D物体识别算法(Re-CRNN).将RGB图像与深度光学信息结合,基于残差学习对双流卷积神经网络(CNN)进行改进:增加顶层特征融合单元,在RGB图像和深度图像中学习联合特征,将提取的RGB和深度图像的高层次特征进行跨通道信息融合,继而使用So...     

基于GRU神经网络的结构异常监测数据修复方法EI北大核心CSCD

结构健康监测系统中通常存在大量的异常监测数据,为保证数据的完整性和可用性,有必要对异常监测数据进行修复。大多数基于深度学习对异常数据进行修复的研究通常使用单输入维度和单向预测的方法搭建模型。提出一种基于门控循环(gated recurrent unit,GRU)神经网络的结构异常监测数据修复方法,该方法充分利用深度学习神经网络适合处理复杂非线性映射问题的优势,并对GRU神经网络进行了优化与重构。利用结构温度、时序先后相关性优化神经网络的输入和输出构造,并提出了利用异常数据前后时间段的信息进行双向序列预测的方法提升数据预测和修复精度。最后,利用某古城墙的应变、裂缝与温度监测数据进行方法验证,采用重构后的GRU神经网络模型对异常数据序列进行修复,并与长短时记忆(long and short-term memory,LSTM)神经网络和反向传播(back propagation,BP)神经网络的修复精度进行比较。结果表明,相比单输入维度、单向预测的网络模型,重构后的GRU神经网络的预测精度大幅提高,且显著优于LSTM神经网络和BP神经网络。异常数据序列修复后,应变和裂缝宽度等结构响应与结构温度的线性相关性大幅增强。该方法对具有温度相关性的结构监测数据具有良好的修复能力。     

基于改进1DCNN-GRU的滚动轴承故障诊断

针对传统滚动轴承故障诊断模型无法充分利用信号的空间及时间特征，需要大量专业知识等问题，提出一种改进一维卷积神经网络(1DCNN)与门控递归神经网络(GRU)结合的故障诊断方法。首先，利用具有不同卷积核的卷积层最大化提取信号的空间特征信息；其次，提出改进的线性修正单元(IReLU)增强网络的特征提取能力；然后，引入堆叠GRU模块进一步提取1DCNN模块输出数据的时间特征，完成空间及时间特征融合；最后，对融合后的特征进行识别。实验表明所提方法故障识别率高达99.96%,对不同负载下的数据均具有较高的识别率及较强的泛化性能。     

全卷积循环神经网络的语音情感识别

语音情感识别是人机交互的热门研究领域之一。然而,由于缺乏对语音中时频相关信息的研究,导致情感信息挖掘深度不够。为了更好地挖掘语音中的时频相关信息,提出了一种全卷积循环神经网络模型,采用并行多输入的方式组合不同模型,同时从两个模块中提取不同功能的特征。利用全卷积神经网络（Fully Convolutional Network,FCN）学习语音谱图特征中的时频相关信息,同时,利用长短期记忆（Long Short-Term Memory,LSTM）神经网络来学习语音的帧级特征,以补充模型在FCN学习过程中缺失的时间相关信息,最后,将特征融合后使用分类器进行分类,在两个公开的情感数据集上的测试验证了所提算法的优越性。     

卷积门控循环残差网络在轴承故障诊断中的研究

针对滚动轴承故障特征较难提取及许多深度学习方法因模型简单而导致准确率偏低的问题,提出一种基于残差网络的门控循环网络(GRU),该算法可以减少时序信息的丢失及解决由于网络较深而出现性能下降的问题。该模型包含2个卷积层、2个GRU层、1个残差块以及1个输出层,其先利用具有强大特征提取能力的卷积神经网络(CNN)提取轴承振动信号中的信息,然后将提取到的信息输入GRU中以保证时序信息不丢失,再通过残差模块解决神经网络深度较深问题,最后通过输出层输出结果。结果表明：该方法可以一次性诊断多种轴承的不同位置及不同尺寸的故障,且对比其他深度学习网络,该算法具有更高的准确性。     

基于AI深度学习的新闻文本分类法研究

针对传统机器学习未能考虑局部信息对新闻文本分类的影响以及无法及时捕获长距离特征信息等问题,本文提出了一种基于AI深度学习的新闻文本分类方法。为进一步验证新闻文本分类效果,该方法将径向基函数模型与多层感知层模型结合,并进行深度学习训练,通过样本标记迭代试验进行测试。结果表明,AI深度学习加快了模型训练过程,提升了模型在迭代过程的正确率,降低了漏报率,显著增强了分类效果。     

双通道特征融合CNN-GRU齿轮箱故障诊断

旋转部件是否发生局部故障,关键是判断其振动信号在空间上是否出现周期性冲击以及周期大小。卷积神经网络(CNN)善于挖掘数据空间上的局部重要的信息特征,具有"端对端"的优势,从而克服了人工提取特征的缺陷;由于振动信号在时间维度上也蕴含着丰富的信息,而长短时记忆网络(LSTM)善于从动态变化的序列数据中学习到时间上的关联性;门控递归单元(GRU)属于LSTM的变种,但相对于LSTM结构更加简洁,参数的数量更少,因此将CNN的空间处理能力和GRU时序处理能力的优势结合,提出一种双通道特征融合CNN-GRU齿轮箱故障诊断方法;即采用并列式结构令CNN与GRU双通道同时提取齿轮箱原始振动信号的故障特征,然后将双通道提取的特征向量合并成一个融合特征向量,输入到SoftMax进行故障分类。该方法可以直接从原始振动信号自适应提取到空间和时序的融合特征,实现了"端对端"的故障诊断。用齿轮实测数据和西储大学轴承数据进行验证,试验结果表明,所提方法识别准确率较高,具有实用性和可行性。     

基于词向量和CNN的书籍评论情感分析方法

针对在线书籍评论的情感分析问题,基于词向量和深度学习原理,提出了一种基于词向量和卷积神经网络(Convolutional Neural Networks, CNN)的书籍评论情感分类方法.实验结果表明,本方法对在线书籍评论的情感分类准确率达到92.99%,同时,此方法对于大量文本集的情感分析有很好的适用性.此外还研究了不同语料库、书籍评论词向量的维度、书籍评论数据集的大小等对情感分析结果的影响,实验结果显示,构建针对性的语料库,基于词向量和卷积神经网络的书籍评论情感分析方法是一种简单而有效的文本情感分析方法,该方法具有扩展性和对不同评论文本的适用性.     

用于声音分类的Deep LightGBM算法

在医学诊断、场景分析、语音识别、生态环境分析等方面语音分类都有着广泛的应用价值。传统的语音分类器采用的是神经网络。但是在精确度,模型设置,参数调整和资料的预处理等方面,有较大的缺陷。在这一基础上,文章提出了一种以“深度森林”为基础的改进方法——LightGBM的深度学习模型（Deep LightGBM模型）。它能够在保证模型简洁的前提下,提高分类精度和泛化能力。该算法有效降低了参数依赖性。在UrbanSound8K这一数据集中,采用向量方法进行语音特征的提取,其分类精确度达95.84%。将卷积神经网络（Convolutional Neural Net‐work,CNN）抽取的特征和向量法获取的特征进行融合,并利用新的模型进行训练,其准确率可达97.67%。实验证明,此算法采用的特征提取方式与Deep LightGBM配合获得的模型参数调整容易,精度高,不会产生过度拟合,并且泛化能力好。     

Investigation on the Chinese Text Sentiment Analysis Based on Convolutional Neural Networks in Deep Learning

Nowadays, the amount of wed data is increasing at a rapid speed, which presents a serious challenge to the web monitoring. Text sentiment analysis, an important research topic in the area of natural language processing, is a crucial task in the web monitoring area. The accuracy of traditional text sentiment analysis methods might be degraded in dealing with mass data. Deep learning is a hot research topic of the artificial intelligence in the recent years. By now, several research groups have studied the sentiment analysis of English texts using deep learning methods. In contrary, relatively few works have so far considered the Chinese text sentiment analysis toward this direction. In this paper, a method for analyzing the Chinese text sentiment is proposed based on the convolutional neural network (CNN) in deep learning in order to improve the analysis accuracy. The feature values of the CNN after the training process are nonuniformly distributed. In order to overcome this problem, a method for normalizing the feature values is proposed. Moreover, the dimensions of the text features are optimized through simulations. Finally, a method for updating the learning rate in the training process of the CNN is presented in order to achieve better performances. Experiment results on the typical datasets indicate that the accuracy of the proposed method can be improved compared with that of the traditional supervised machine learning methods, e.g., the support vector machine method.     

A Knowledge-Enriched and Span-Based Network for Joint Entity and Relation Extraction

The joint extraction of entities and their relations from certain texts plays a significant role in most natural language processes. For entity and relation extraction in a specific domain, we propose a hybrid neural framework consisting of two parts: a span-based model and a graph-based model. The span-based model can tackle overlapping problems compared with BILOU methods, whereas the graph-based model treats relation prediction as graph classification. Our main contribution is to incorporate external lexical and syntactic knowledge of a specific domain, such as domain dictionaries and dependency structures from texts, into end-to-end neural models. We conducted extensive experiments on a Chinese military entity and relation extraction corpus. The results show that the proposed framework outperforms the baselines with better performance in terms of entity and relation prediction. The proposed method provides insight into problems with the joint extraction of entities and their relations.     

基于多分辨率时频特征融合的声学场景分类

声学场景分类是计算机听觉中最难的任务之一,在单一特征条件下采用基本的卷积神经网络相对于传统的分类方法精度已经有所提升,但是效果依然不够理想。针对这一问题,在卷积神经网络框架下,提出了一种基于时频特征融合的声学场景分类方案。在分类模型构建方面,提出一种多分辨率卷积池化方案,构造多分辨率卷积神经网络,以更好地适应提取特征的时频结构;在特征选取方面,融合低层次包络特征对数——Mel子带能量和高层次结构特征——非负矩阵分解系数矩阵,把两种二维特征堆叠为三维特征送入分类模型。在2017年和2018年声学场景分类和事件检测挑战赛的开发数据集上进行了训练和测试。实验结果表明,文中提出方案比基线系统的分类精度分别提高7.5%和10.3%,可有效改善分类效果。     

基于深度学习的舰船辐射噪声多特征融合分类

为了改善分类系统的性能,进一步提高舰船辐射噪声分类的正确率,该文提出了一种基于深度神经网络的多特征融合分类方法。该方法首先提取舰船辐射噪声几种不同的特征,将提取的特征同时用于训练具有多个输入分支的深度神经网络,使网络直接在多种特征参数上进行联合学习,通过神经网络的输入分支和连接层实现特征融合,再对舰船辐射噪声进行分类。为了特征深度学习提取了舰船辐射噪声的频谱特征、梅尔倒谱系数和功率谱特征,并将多特征融合分类方法与在一种特征上进行深度学习分类方法的正确率进行对比。实验结果表明,基于深度学习的多特征融合分类方法可以有效地提高舰船辐射噪声分类的正确率,是一种可行的分类方法。     

Fitness adaptive deer hunting-based region growing and recurrent neural network for melanoma skin cancer detection

This proposal aims to enhance the accuracy of a dermoscopic skin cancer diagnosis with the aid of novel deep learning architecture. The proposed skin cancer detection model involves four main steps: (a) preprocessing, (b) segmentation, (c) feature extraction, and (d) classification. The dermoscopic images initially subjected to a preprocessing step that includes image enhancement and hair removal. After preprocessing, the segmentation of lesion is deployed by an optimized region growing algorithm. In the feature extraction phase, local features, color morphology features, and morphological transformation-based features are extracted. Moreover, the classification phase uses a modified deep learning algorithm by merging the optimization concept into recurrent neural network (RNN). As the main contribution, the region growing and RNN improved by the modified deer hunting optimization algorithm (DHOA) termed as Fitness Adaptive DHOA (FA-DHOA). Finally, the analysis has been performed to verify the effectiveness of the proposed method.     

An Optimized Approach to Vehicle-Type Classification Using a Convolutional Neural Network

Vehicle type classification is considered a central part of an intelligent traffic system. In recent years, deep learning had a vital role in object detection in many computer vision tasks. To learn high-level deep features and semantics, deep learning offers powerful tools to address problems in traditional architectures of handcrafted feature-extraction techniques. Unlike other algorithms using handcrated visual features, convolutional neural network is able to automatically learn good features of vehicle type classification. This study develops an optimized automatic surveillance and auditing system to detect and classify vehicles of different categories. Transfer learning is used to quickly learn the features by recording a small number of training images from vehicle frontal view images. The proposed system employs extensive data-augmentation techniques for effective training while avoiding the problem of data shortage. In order to capture rich and discriminative information of vehicles, the convolutional neural network is fine-tuned for the classification of vehicle types using the augmented data. The network extracts the feature maps from the entire dataset and generates a label for each object (vehicle) in an image, which can help in vehicle-type detection and classification. Experimental results on a public dataset and our own dataset demonstrated that the proposed method is quite effective in detection and classification of different types of vehicles. The experimental results show that the proposed model achieves 96.04% accuracy on vehicle type classification.     

Deep learning based pipeline with multichannel inputs for patent classification

This work introduces a deep learning pipeline for automatic patent classification with multichannel inputs based on LSTM and word vector embeddings. Sophisticated text mining methods are used to extract the most important segments from patent texts, and a domain-specific pre-trained word embeddings model for the patent domain is developed; it was trained on a very large dataset of more than five million patents. The deep learning pipeline is using multiple parallel LSTM networks that read the source patent document using different input dimensions namely embeddings of different segments of patent texts, and sparse linear input of different metadata. Classifying patents into corresponding technical fields is selected as a use case. In this use case, a series of patent classification experiments are conducted on different patent datasets, and the experimental results indicate that using the segments of patent texts as well as the metadata as multichannel inputs for a deep neural network model, achieves better performance than one input channel.     

基于注意力机制的包装命名实体识别

目的 为了解决包装行业相关文本命名实体识别困难问题,提出在BiLSTM（Bidirectional Long Short-Term Memory）神经网络中加入注意力机制（Attention）和字词联合特征,构建一种基于注意力机制的BiLSTM深度学习模型（简称Attention-BiLSTM）,以识别包装命名实体。方法 首先构建包装领域词典匹配包装语料中词语的类别特征,同时将包装语料转换为字特征和词特征联合的向量特征,并且在过程中加入POS（词性）信息。然后将以上特征联合馈送到BiLSTM网络,以获取文本的全局特征,并利用注意力机制获取局部特征。最后根据文本的全局特征和局部特征使用CRF（Conditional Random Field）解码整个句子的最优标注序列。结果 通过对《中国包装网》新闻数据集的实验,获得了85.6%的F值。结论 所提方法在包装命名实体识别中优于传统方法。     

A Novel Combinational Convolutional Neural Network for Automatic Food-Ingredient Classification

With the development of deep learning and Convolutional Neural Networks (CNNs), the accuracy of automatic food recognition based on visual data have significantly improved. Some research studies have shown that the deeper the model is, the higher the accuracy is. However, very deep neural networks would be affected by the overfitting problem and also consume huge computing resources. In this paper, a new classification scheme is proposed for automatic food-ingredient recognition based on deep learning. We construct an up-to-date combinational convolutional neural network (CBNet) with a subnet merging technique. Firstly, two different neural networks are utilized for learning interested features. Then, a well-designed feature fusion component aggregates the features from subnetworks, further extracting richer and more precise features for image classification. In order to learn more complementary features, the corresponding fusion strategies are also proposed, including auxiliary classifiers and hyperparameters setting. Finally, CBNet based on the well-known VGGNet, ResNet and DenseNet is evaluated on a dataset including 41 major categories of food ingredients and 100 images for each category. Theoretical analysis and experimental results demonstrate that CBNet achieves promising accuracy for multi-class classification and improves the performance of convolutional neural networks.