基于迁移学习的唐诗宋词情感分析

随着计算社会学的兴起,利用数据挖掘分析社会情感是近期的研究重点.当前的研究主要针对现代文本,对于古代诗歌这类短文本的情感分析相对较少.本文提出了一个基于短文本特征扩展的迁移学习模型CATL-PCO,通过分析诗歌情感对当时社会及文化进行进一步了解.该模型首先基于频繁词对对古文特征向量进行扩展,再通过迁移学习方式,建立三个分类器并投票得出最后的情感分析结果.CATL-PCO模型首先能够解决古文短文本特征稀疏的问题,在此基础上进一步解决由于现代译文信息匮乏所导致的古代诗歌情感分析困难问题,从而准确的分析古诗词情感倾向,从计算社会学的角度,增进对中国历史的认识.实验表明,当训练集为中国唐诗时,本文提出方法能够准确的对唐代诗歌进行情感分类,并能应用于唐代和宋代各个时期情感分析及代表流派分析.     

基于情感标签的极性分类

情感极性分析是文本挖掘中一种非常重要的技术.然而在不同领域中,很多情感极性分类系统存在分类精度低和缺少大量标注数据的缺陷.针对这些问题,提出了一种基于情感标签的极性分类方法.首先通过所有文本建立Sentiment-Topic模型,抽取出文本的情感标签;然后利用情感标签将文本划分为两个子文本,并通过Co-training算法对子文本进行分类;最后合并两个子文本的分类结果,并确定文本的情感极性.实验结果表明该方法具有较高的分类精度,而且不需要大量的分类样本.     

中文文本情感倾向性五元模型研究

目前,情感倾向性判断正成为文本信息服务技术研究的热点和难点之一,而＂中文文本情感倾向性观点＂表示模型是文本情感倾向性判断的基础。情感倾向性五元模型从情感倾向性观点的持有者、倾向性的来源、倾向性的指向、倾向性的立场和倾向性的种类五个方面刻画中文文本情感倾向性的概念,丰富了情感倾向性的表示方法,提高了文本情感倾向性判断的精度,并以此为基础给出了句子级文本情感倾向性判断的定义和过程。     

大数据环境下的网络舆情管理系统设计

随着网络的快速发展,越来越多的人们通过网络发表个人观点及看法,网络舆情成为社会舆情中的重点对象和主要方式.本文通过对大数据环境下网络舆情及其特点的阐述、分析,结合数据挖掘、文本情感分析等技术,初步构建出了网络舆情管理系统的模型.     

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.     

面向顾客点评数据的属性层次观点挖掘研究

This paper focuses on how to improve aspect-level opinion mining for online customer reviews. We first propose a novel generative topic model, the Joint Aspect/Sen-timent (JAS) model, to jointly extract aspects and aspect-dependent sentiment lexicons from online customer reviews. An as-pect-dependent sentiment lexicon refers to the aspect-specific opinion words along with their aspect-aware sentiment polarities with respect to a specific aspect. We then apply the extracted aspect- dependent sentiment lexi-cons to a series of aspect-level opinion mining tasks, including implicit aspect identification, aspect-based extractive opinion summarization, and aspect-level sentiment classification. Experimental results demonstrate the effectiveness of the JAS model in learning aspect- dependent sentiment lexicons and the practical values of the extracted lexicons when applied to these practical tasks.     

Analytic‐based product opinion detection algorithm for twitter microblogging network

Twitter, with an ever‐increasing user base, has greatly influenced the opinion and purchase habits of the common masses. This has in turn forced the product firms to get involved with sentiment analysis which enables them to mine the actual opinion about their product and make business decisions accordingly. Even though a majority of the existing methods detect sentiment of the tweet with a reasonable accuracy, few ignore emoticons while others consider them as stop words. Emoticons have enabled the users to express their emotion more accurately which eliminates the ambiguity that can arise with usage of words. The trending popularity of emoticons among the users combined with its ease of usage makes it highly lucrative in sentiment analysis. Hence, mining the product opinion without considering the emoticons will severely undermine the accuracy and reliability of the opinion. Moreover, sarcasm detection is still an uncharted territory in opinion mining and is exceedingly difficult to factor it in. Sarcastic tweets when left undetected will affect the accuracy of the opinion. Therefore, the polarity of the individual words and emoticons of the tweets are computed using linguistic analysis. The sarcastic tweets are then classified and eliminated based on their anomalous polarity. By placing a higher emphasis on emoticons, the proposed emoticon‐based linguistic opinion algorithm yields satisfactory results when compared with other traditional and state of the art approaches.     

一种基于Opinosis图和马尔科夫随机游走模型的多文本情感摘要框架

针对在线文本情感摘要生成问题,本文提出了一种基于Opinosis图和马尔科夫随机游走模型的情感摘要框架.首先,该框架将原始文本转化为Opinosis图,并利用其挖掘出文本中的特征词,这些特征词可以用来对原始文本的句子进行分类;其次本文在基于聚类的条件马尔科夫随机游走模型的基础上增加了情感层,改进后的模型可以判断同一聚类中各句子的情感倾向是否具有代表性并结合情感和聚类信息对句子进行排序.实验结果表明,本文提出的方法与基准算法相比在ROUGE（Recall-Oriented Understudy for Gisting Evaluation）值上具有明显提高.     

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.     

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.     

基于频繁模式图的多维关联规则挖掘算法研究

关联规则挖掘是数据挖掘领域中重要的研究分支,频繁项集或频繁谓词集的计算是其中的关键问题.本文针对包括多值属性的关系数据库,以多维关联规则挖掘为目标,研究频繁谓词集的计算方法,提出了MPG算法及IMPG增量算法.MPG算法通过构建频繁模式图MP-graph,按照深度优先搜索方法,动态挖掘频繁谓词集,只需扫描数据库一次.此外,该方法至多增加一次数据库扫描,就能扩展为IMPG算法,进行增量关联规则挖掘.文章分析了算法时间和空间性能,用实验说明了算法的有效性.     

Mining Mobile Sequential Patterns in a Mobile Commerce Environment

In this paper, we explore a new data mining capability for a mobile commerce environment. To better reflect the customer usage patterns in the mobile commerce environment, we propose an innovative mining model, called mining mobile sequential patterns, which takes both the moving patterns and purchase patterns of customers into consideration. How to strike a compromise among the use of various knowledge to solve the mining on mobile sequential patterns is a challenging issue. We devise three algorithms (algorithm TJ_LS, algorithm TJ_PT, and algorithm TJ_PF) for determining the frequent sequential patterns, which are termed large sequential patterns in this paper, from the mobile transaction sequences. Algorithm TJ_LS is devised in light of the concept of association rules and is used as the basic scheme. Algorithm TJ_PT is devised by taking both the concepts of association rules and path traversal patterns into consideration and gains performance improvement by path trimming. Algorithm TJ_PF is devised by utilizing the pattern family technique which is developed to exploit the relationship between moving and purchase behaviors, and thus is able to generate the large sequential patterns very efficiently. A simulation model for the mobile commerce environment is developed, and a synthetic workload is generated for performance studies. In mining mobile sequential patterns, it is shown by our experimental results that algorithm TJ_PF significantly outperforms others in both execution efficiency and memory saving, indicating the usefulness of the pattern family technique devised in this paper. It is shown by our results that by taking both moving and purchase patterns into consideration, one can have a better model for a mobile commerce system and is thus able to exploit the intrinsic relationship between these two important factors for the efficient mining of mobile sequential patterns     

分布式存储结构的频繁闭合模式挖掘并行算法

研究分布式存储结构下频繁闭合模式挖掘的并行化问题,针对频繁闭合模式的特点,提出了两阶段并行判断频繁模式闭合性的方法,基于串行算法FPclose和两种FP-tree的并行构造方式,分别给出了两个频繁闭合模式挖掘并行算法DP-FP和DL-FP,性能分析表明,这两个算法具有较大的并行化,较小的I/O开销与良好的负载平衡。     

基于多维特征分析的社交网络意见领袖挖掘

在社交网络中进行意见领袖的挖掘对信息传播与演化的深度分析、舆情监控和引导具有重要意义,本文综合结构特征、行为特征和用户的情感特征对意见领袖节点挖掘问题进行研究.本文首先对微博真实文本数据进行话题识别得到主题社区,在主题社区中基于用户节点之间的关注关系构建交互网络拓扑.然后分别从结构、行为和情感三个维度对用户的影响力进行度量.最后,分析用户在主题社区中的影响力分布与传播规律,提出意见领袖识别算法MFP(Multi-Feature PageRank).实验表明,该算法可有效地挖掘潜在的意见领袖节点,能够获得较高的支持率.     

基于HASH表的多谓词约束下频繁项集挖掘

针对在交易数据库中挖掘出指定顾客相关属性的频繁项集这一问题,提出了基于维约束进行求解的构想.采用模式增长的挖掘方法,但与传统的模式树不同的是将原先每一节点频繁计数值设为在所有可能的谓词约束下该项的计数形成的向量,并利用HASH表进行向量值及项所在层的位置映射,因此,在不同的约束组合下的频繁项集挖掘将不再需要扫描数据库.仿真实验表明该挖掘算法的完备性,通过与先筛选再挖掘的算法进行比较,证明该挖掘算法具有更高的效率.     

基于云模型的高校网络舆情监控系统

设计了一种基于云模型的高校网络舆情监控系统,该系统采用基于立即价值和未来价值综合评价的方式指导网页爬行策略,采用云模型指导爬行方向,同时通过正向和逆向云模型对主题网页进行聚类,并通过查询扩展技术提高网页查全率,取得了很好的实际应用效果.     

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。     

Traffic Characteristics Based Dynamic Radio Resource Management in Heterogeneous Wireless Networks

The traffic with tidal phenomenon in Heterogeneous Wireless Networks （HWNs） has radically increased the complexity of radio resource management and its performance analysis. In this paper, a Simplified Dynamic Hierarchy Resource Management （SDHRM） algorithm exploiting the resources dynami- cally and intelligently is proposed with the consideration of tidal traffic. In network-level resource allocation, the proposed algorithm first adopts wavelet neural network to forecast the traffic of each sub-area and then allocates the resources to those sub-areas to maximise the network utility. In connection-level net- work selection, based on the above resource allocation and the pre-defined QoS require- ment, three typical network selection policies are provided to assign traffic flow to the most appropriate network. Furthermore, based on multidimensional Markov model, we analyse the performance of SDHRM in HWNs with heavy tailed traffic. Numerical results show that our theoretical values coincide with the simulation results and the SDHRM can im- prove the resource utilization.