文本情感文摘自动提取方法研究

为了快速获取网络文本中主题内容和情感信息,提出了文本情感文摘的概念,同时提出了一种基于条件随机场模型的情感文摘提取方法.首先提取文本中的句子长度、提示词以及情感词语作为基本特征,同时应用浅层狄利赫雷分配的主题模型,分析文本潜在主题信息,提取主题特征,将这两类特征同时应用到条件随机场模型中,从而获取文本的情感文摘.实验结果表明,该方法细腻刻画了文本的主题信息,同时考虑了文本主题的情感色彩,文摘提取效果较理想,能满足用户的实际需要.     

A two stream convolutional neural network with bi-directional GRU model to classify dynamic hand gesture

Dynamic hand gesture recognition is still an interesting topic for the computer vision community. A set of feature vectors can represent any hand gesture. A Recurrent Neural Network (RNN) can recognize these feature vectors as a hand gesture that analyzes the temporal and contextual information of the gesture sequence. Thus, we proposed a hybrid deep learning framework to recognize dynamic hand gestures. In the Hybrid model GoogleNet is pipelined with a Bidirectional GRU unit to recognize the dynamic hand gesture. Dynamic hand gestures consist of many frames, and features of each frame need to be extracted to get the temporal and dynamic information of the performed gesture. As RNN takes input as a sequence of feature vectors, we extract features from videos using pretrained GoogleNet. As Gated Recurrent Unit is one of the variants of RNN to classify the sequential data, we created a feature vector that corresponds to each video and passed it to the bidirectional GRU (BGRU) network to classify the gestures. We evaluate our model on four publicly available hand gesture datasets. The proposed method performs well and is comparable with the existing methods. For instance, we achieved 98.6% accuracy on Northwestern University Hand Gesture(NWUHG), 99.6% on SKIG, 99.4% on Cambridge Hand Gesture (CHG) datasets respectively. We performed our experiments on DHG14/28 dataset and achieved an accuracy of 97.8% with 14-gesture classes and 92.1% on 28-gesture classes. DHG14/28 dataset contains skeleton and depth data, and our proposed model used depth data and achieved comparable accuracy.     

基于词汇语义和句法依存的情感关键句识别

门户网站、博客和论坛中的新闻性文章往往都带有自己的情感倾向性,而情感关键句的识别对判断文章的情感倾向、了解社会动态和舆情状况有着非常重要的作用。传统方法主要基于词汇特征,未能充分利用潜在的句法和语义信息。本文提出了一种基于词汇语义和句法依存的情感关键句识别方法。该方法首先通过构建情感词典和关键词词典获取词汇语义信息,然后利用一种新颖的面向情感关键句提取算法获取句法依存信息,最后把情感关键句的识别问题看成一个是否为情感关键句的二分类问题加以解决。在COAE2014公开评测数据集上进行的实验表明本文方法的准确率和召回率均显著优于其他方法。     

基于TSCM模型的网络短文本情感挖掘

针对网络短文本情感挖掘问题,提出一种新的基于LDA和互联网短评行为理论的主题情感混合模型TSCM,TSCM模型中的整篇评论中每个句子的主题分布是不同的,TSCM产生词的流程是先确定词的情感极性,再确定词的主题,TSCM考虑了词与词之间的联系.真实数据集Movie与Amazon上的大量实验表明,与代表性算法JST、S-LDA、D-PLDA和SAS相比较,TSCM模型能对用户真实情感与讨论主题进行更加有效的分析建模.     

Multi-level fusion with deep neural networks for multimodal sentiment classification

The task of multimodal sentiment classification aims to associate multimodal information, such as images and texts with appropriate sentiment polarities. There are various levels that can affect human sentiment in visual and textual modalities. However, most existing methods treat various levels of features independently without having effective method for feature fusion. In this paper, we propose a multi-level fusion classification (MFC) model to predict the sentiment polarity based on the fusing features from different levels by exploiting the dependency among them. The proposed architecture leverages convolutional neural networks ( CNNs) with multiple layers to extract levels of features in image and text modalities. Considering the dependencies within the low-level and high-level features, a bi-directional (Bi) recurrent neural network (RNN) is adopted to integrate the learned features from different layers in CNNs. In addition, a conflict detection module is incorporated to address the conflict between modalities. Experiments on the Flickr dataset demonstrate that the MFC method achieves comparable performance compared with strong baseline methods.     

基于Transformer编码器的金融文本情感分析方法

目前针对文本情感分析的研究大多集中在商品评论和微博的情感分析领域,对金融文本的情感分析研究较少。针对该问题,文中提出一种基于Transformer编码器的金融文本情感分析方法。Transformer编码器是一种基于自注意力机制的特征抽取单元,在处理文本序列信息时可以把句中任意两个单词联系起来不受距离限制,克服了长程依赖问题。文中所提方法使用Transformer编码器构建情感分析网络。Transformer编码器采用多头注意力机制,对同一句子进行多次计算以捕获更多的隐含在上下文中的语义特征。文中在以金融新闻为基础构建的平衡语料数据集上进行实验,并与以卷积神经网络和循环神经网络为基础构建的模型进行对比。实验结果表明,文中提出的基于Transformer编码器的方法在金融文本情感分析领域效果最好。     

基于句式结构的评论倾向性识别方法

网络评论的倾向性识别是一种重要的网络舆情分析技术。针对网络评论的特点,提出了基于句式结构的网络评论倾向性识别方法。该识别方法主要思想是先对已进行倾向性识别的网络评论集进行分词并提取句式结构,形成各种句式结构的基准句集;再对待判别的网络评论提取句式结构,并比较其与相应句式结构的基准句集的相似度,从而判别出待识别的评论的褒贬倾向性。实验表明：该方法的准确率为80%以上,具有良好的实用性。     

Research on calculation method of text similarity based on smooth inverse frequency

In order to improve the accuracy of text similarity calculation, this paper presents a text similarity function part of speech and word order-smooth inverse frequency (PO-SIF) based on sentence vector, which optimizes the classical SIF calculation method in two aspects: part of speech and word order. The classical SIF algorithm is to calculate sentence similarity by getting a sentence vector through weighting and reducing noise. However, the different methods of weighting or reducing noise would affect the efficiency and the accuracy of similarity calculation. In our proposed PO-SIF, the weight parameters of the SIF sentence vector are first updated by the part of speech subtraction factor, to determine the most crucial words. Furthermore, PO-SIF calculates the sentence vector similarity taking into the account of word order, which overcomes the drawback of similarity analysis that is mostly based on the word frequency. The experimental results validate the performance of our proposed PO-SIF on improving the accuracy of text similarity calculation.     

多交互图卷积网络用于方面情感分析

方面情感分析旨在识别句子中特定方面的情感极性,是一项细粒度情感分析任务。传统基于注意力机制方法,仅在单词之间进行单一的语义交互,没有建立方面词与文本词的语法信息交互,导致方面词错误地关注到与其语法无关的文本词信息。此外,单词的位置距离特征和语法距离特征,分别体现其在句子线性形式中和句子语法依存树中的位置关系,而基于图卷积网络处理语法信息的方法却忽略距离特征,使距方面词较远的无关信息对其情感分析造成干扰。针对上述问题,该文提出多交互图卷积网络(MIGCN),首先将文本词位置距离特征馈入到每层图卷积网络,同时利用依存树中文本词的语法距离特征对图卷积网络的邻接矩阵加权,最后,设计语义交互和语法交互分别处理单词之间语义和语法信息。实验结果表明,在公共数据集上,准确率和宏F1值均优于基准模型。     

用反馈式语音识别理解方案进行汉语短语的识别理解

汉语语音理解系统的任务之一是把语音识别系统获得的汉语单音节转换成正确的汉字、词,乃至汉语的短语、语句,与语音识别系统一起,完成一个语音到文本(speech to text)的转换系统。本文利用一个闭环反馈方式汉语语音识别理解方案,在汉语词识别理解的基础上,进一步实现对汉语结构性短语的识别理解,获得了预期的结果。最后本文对实验结果和反馈式语音识别理解方案进行了讨论。     

Deep Neural Architecture for Recovering Dropped Pronouns in Korean

Pronouns are frequently dropped in Korean sentences, especially in text messages in the mobile phone environment. Restoring dropped pronouns can be a beneficial preprocessing task for machine translation, information extraction, spoken dialog systems, and many other applications. In this work, we address the problem of dropped pronoun recovery by resolving two simultaneous subtasks: detecting zero-pronoun sentences and determining the type of dropped pronouns. The problems are statistically modeled by encoding the sentence and classifying types of dropped pronouns using a recurrent neural network (RNN) architecture. Various RNN-based encoding architectures were investigated, and the stacked RNN was shown to be the best model for Korean zero-pronoun recovery. The proposed method does not require any manual features to be implemented; nevertheless, it shows good performance.     

一种基于多级空间视觉词典集体的图像分类方法

针对单一特征时存在提取的信息量不足,对图像内容描述比较片面,提出将传统的SIFT特征与KDES-G特征进行串行融合,生成一个联合向量作为新的特征向量.针对传统的视觉词典构造方法缺乏考虑视觉词汇在空间的分布特点,本文引入图像空间信息,提出了一种空间视觉词典的构造方法,先对图像进行空间金字塔划分,再把空间各子区域内的特征分别聚类,构建属于对应子空间区域的空间视觉词典.在图像表示阶段,图像各子区域内的特征基于其对应的空间视觉词典进行LLC稀疏编码,根据各子区域对图像贡献程度的不同,把编码后各子区域的特征向量赋予不同的权重加权处理,再连接形成最终的图像描述.最后,利用线性SVM进行图像分类,实验结果表明了本文方法的有效性和鲁棒性.     

结合全局词向量特征的循环神经网络语言模型

 针对循环神经网络语言模型对长距离历史信息学习能力不足的问题,本文提出了结合全局词向量特征的循环神经网络语言模型。首先利用GloVe(Global Word Vectors)算法训练出全局词向量,然后将其作为特征向量输入到引入特征层的循环神经网络中进行训练。相对于局部词向量方法,全局词向量能够利用全局统计信息来训练出含有更加丰富的语义和句法信息词向量。为了验证新方法的性能,本文在Penn Treebank和Wall Street Journal语料库上分别进行困惑度和连续语音识别实验。实验结果表明结合全局词向量的循环神经网络语言模型的困惑度相比传统的循环神经网络语言模型降低了202%,同时语音识别系统的词错误率降低了183%。      

基于词向量的SVM集成学习社交网络抑郁倾向检测方法

目前抑郁症的诊断手段单一、诊断率低,为此,文章提出一种基于词向量的SVM集成学习社交网络抑郁倾向检测方法。人工标注和专家校验获得训练数据,使用词向量进行文本向量化。以SVM为基分类器进行Boosting集成学习。实验结果表明,文章提出的模型可以用于抑郁倾向的检测。     

基于句法分析和答案分类的中文问答系统

本文根据疑问词和谓语的距离信息对问句进行细致的句型分析,然后对答句进行浅层句法分析,在此基础上,抽取出问题特征集、答句特征集和组合特征集作为分类特征,引入最大熵模型和支持向量机训练答案抽取分类器.基于不同特征组合训练得到的分类器在五类事实性问题上进行了测试,其F值分别达到70.87%和85.75%.     

中国科普网站的特征向量研究

在中国,识别科普网站的内容长期以来主要是依靠专家判断来进行。这种主观判断不仅费时费力,效果也并不好。这其中最主要的一个原因是网站内容比较丰富,人工浏览效率低下,在一定的时间只能处理有限的内容,对于整个网站的判断会存在不全面的地方,也具有主观性。对此问题的解决需要提出一个基于人工智能的可以进行快速定量计算的方法。本文提出的科普网站特征向量就是讲网站内容通过计算机进行处理抽象出来的一个向量空间模型,它能比较好的表现网站的文字内容和意思,可以最终实现机器自动判断网站内容是否含有科普成分以及什么性质的科普内容。     

基于胶囊网络的中长微博情感分析

针对通过微博文本获取用户情感倾向,以提高舆情监控效率的问题。利用深度学习的方法实现微博语料的情感分类,构建符合近年文本长度分布特点的高质量微博情感分类数据集,分析微博文本长度对情感分类的影响。由于中长语料主观性强、句子关联度弱,其检测准确率偏低。针对此问题,本文提出一种基于胶囊网络的中长微博情感分析模型。采用注意力机制,在融合局部特征与全局特征的基础上,利用胶囊向量实现深层情感特征提取,提高中长语料的检测效果。利用本文搜集的数据集进行实验,结果表明,相较于多种深度学习算法,本文模型性能更佳。在不同文本长度语料的对比实验中,伴随着文本长度的增加,分类准确率逐渐降低。相较于传统的LSTM算法,本文模型随文本长度增加效果提升,证明了该模型针对中长微博文本情感分类的可行性。      

基于改进Transformer模型的文本摘要生成方法

传统的文本摘要方法,如基于循环神经网络和Encoder-Decoder框架构建的摘要生成模型等,在生成文本摘要时存在并行能力不足或长期依赖的性能缺陷,以及文本摘要生成的准确率和流畅度的问题。对此,提出了一种动态词嵌入摘要生成方法。该方法基于改进的Transformer模型,在文本预处理阶段引入先验知识,将ELMo(Embeddings from Language Models)动态词向量作为训练文本的词表征,结合此词对应当句的文本句向量拼接生成输入文本矩阵,将文本矩阵输入到Encoder生成固定长度的文本向量表达,然后通过Decoder将此向量表达解码生成目标文本摘要。实验采用Rouge值作为摘要的评测指标,与其他方法进行的对比实验结果表明,所提方法所生成的文本摘要的准确率和流畅度更高。     

A multistage motion vector processing method for motion-compensated frame interpolation.

In this paper, a novel, low-complexity motion vector processing algorithm at the decoder is proposed for motion-compensated frame interpolation or frame rate up-conversion. We address the problems of having broken edges and deformed structures in an interpolated frame by hierarchically refining motion vectors on different block sizes. Our method explicitly considers the reliability of each received motion vector and has the capability of preserving the structure information. This is achieved by analyzing the distribution of residual energies and effectively merging blocks that have unreliable motion vectors. The motion vector reliability information is also used as a prior knowledge in motion vector refinement using a constrained vector median filter to avoid choosing identical unreliable one. We also propose using chrominance information in our method. Experimental results show that the proposed scheme has better visual quality and is also robust, even in video sequences with complex scenes and fast motion.