基于评论质量的多文档文本情感摘要

文本情感摘要任务旨在对带有情感的文本数据进行浓缩、提炼进而产生文本所表达的关于情感意见的摘要,用以帮助用户更好地阅读、理解情感文本的内容。该文主要研究多文档的文本情感摘要问题,重点针对网络上存在的同一个产品的多个评论进行摘要抽取。在情感文本中,情感相关性是一个重要的特点,该文将充分考虑情感信息对文本情感摘要的重要影响。同时,对于评论语料,质量高的评论或者说可信度高的评论可以帮助用户更好的了解评论中所评价的对象。因此,该文将充分考虑评论质量对文本情感摘要的影响。并且为了进行关于文本情感摘要的研究,该文收集并标注了一个基于产品评论的英文多文档文本情感摘要语料库。实验证明,情感信息和评论质量能够帮助多文档文本情感摘要,提高摘要效果。     

弹幕信息协助下的视频多标签分类

文中探究了弹幕信息协助下的视频多标签分类任务.多标签视频分类任务根据视频内容从不同角度赋予视频多个标签,与视频推荐等应用紧密相关.多标签视频数据集的高标注成本和对视频内容的多角度理解是该研究领域面临的主要问题.弹幕是一种新近出现的用户评论形式,受到了众多用户的欢迎.由于用户参与度高,弹幕视频网站的视频拥有大量用户自发添...     

基于弹幕的网络舆情文本挖掘与情感分析

针对传统评论方式依赖整体感知且相对滞后的问题,以弹幕这一新兴短信息表达方式为研究对象,采用文本挖掘与情感分析的方式研究弹幕与网络舆情之间的潜在联系。采用网络爬虫技术采集网络舆情弹幕数据,使用Jieba库实现分词、去停用词及高频词统计,基于WordCloud库绘制词云图,实现可视化,并使用SnowNLP库计算网络舆情弹幕的情感得分,运用隐含狄利克雷分布(Latent Dirichlet Allocation, LDA)模型进行主题词提取,实现对网络舆情弹幕的情感分类和主题分析。实验结果表明,该方法可多维展现网民的情感倾向与关注焦点,是对传统评论文本研究的有效补充。     

基于弹幕情感分析和主题模型的视频推荐算法

针对互联网上大量自制视频缺少用户评分、推荐准确率不高的问题,提出一种融合弹幕情感分析和主题模型的视频推荐算法（VRDSA）。首先,对视频的弹幕评论进行情感分析,得到视频的情感向量,之后基于情感向量计算视频之间的情感相似度;同时,基于视频的标签建立主题模型来得到视频标签的主题分布,并使用主题分布计算视频之间的主题相似度;接着,对视频的情感相似度和主题相似度进行融合得到视频间的综合相似度;然后,结合视频间的综合相似度和用户的历史记录得到用户对视频的偏好度;同时通过视频的点赞量、弹幕量、收藏数等用户互动指标对视频的大众认可度进行量化,并结合用户历史记录计算出视频的综合认可度;最后,基于用户对视频的偏好度和视频的综合认可度预测用户对视频的认可度,并生成个性化推荐列表来完成视频的推荐。实验结果表明,与融合协同过滤和主题模型的弹幕视频推荐算法（DRCFT）以及嵌入LDA主题模型的协同过滤算法（ULR-itemCF）相比,所提算法推荐的准确率平均提高了17.1%,召回率平均提高了22.9%,F值平均提高了22.2%。所提算法对弹幕进行情感分析,并融合主题模型,以此来完成对视频的推荐,并且充分挖掘了弹幕数据的情感性,使得推荐结果更加准确。     

基于弹幕文本挖掘的社交媒体KOL研究

社交媒体关键意见领袖KOL为品牌营销带来更多机会,所以备受广告主青睐,但KOL行业的低门槛进入和数据造假行为,导致广告主无法快速找到与自身品牌匹配的KOL。基于以上背景,对KOL发布在社交平台的视频进行研究,对视频中的弹幕文本进行动态主题分析,刻画弹幕主题随时间的变化,同时使用卷积神经网络模型对含有广告的视频弹幕文本进行情感分析,进一步分析观众对于KOL推广行为的情感极性。实验结果表明,本文提出的KOL分析方法更加全面具体地评估了KOL的商业价值,能够帮助广告主高效找到合适的KOL。     

基于视频弹幕的特征发现及情感分析研究

随着数字媒体技术的快速发展,弹幕在电子产品评测视频中的使用频率逐年增高,越来越多的用户倾向于通过这种方式了解产品的信息并发表自己的见解.该类弹幕评论除了具有短小、实时性强等特点外,还包含着大量弹幕用户的情感倾向信息.这些信息对于网站和商家都具有重要意义.针对这个问题,对爬取的华为P30手机评测视频共9万2千余条视频弹幕...     

基于弹幕分析的在线直播平台用户理解

弹幕评论是网络直播平台与用户交互的主要方式之一,借助弹幕行为的分析可以更有效地实现对网络直播平台的用户理解.通过采集和利用3大热门直播平台(斗鱼、熊猫与战旗)的弹幕相关数据,本文以假设验证的方式从用户属性与用户行为两个角度对在线直播平台用户进行分析与理解,并建立基于用户行为特征时间序列的用户活跃模型对用户互动活跃度进行量化评估.研究表明,平台在线人数具有周期性变化的时间规律,观众地域具有沿海发达城市集中分布的空间取向,所提出的用户活跃模型能够对网络直播平台用户的行为活跃趋势做出合理的预测分析.     

基于注意力与门控机制相结合的细粒度情感分析

细粒度情感分析(fine-grained sentiment analysis)是自然语言处理领域的关键问题之一,其通过学习文本的上下文信息来进行特定方面的情感分析,可以帮助用户和商家更好地了解用户评论特定方面的情感。针对基于用户评论的方面级别细粒度情感分析任务,提出了BiGRU-Attention与门控机制(gated mechanisms)相结合的文本情感分类模型。首先,通过整合现有的情感资源,将HOWNET评价情感词典作为种子情感词典,利用SO-PMI算法扩充用户评论情感词典,结合否定词典以及词性信息扩充用户评论情感知识,将用户评价情感知识作为用户评论情感特征信息;其次,引入字词特征与情感特征信息,将它们联合作为模型输入,使用BiGRU对文本进行深层次的特征提取;然后,结合门控机制以及注意力机制,根据获取的方面词信息进一步提取与方面词相关的上下文情感特征信息;最后,在输出层进行文本情感分析,经过softmax获得最终的情感极性。在AIchallenger2018细粒度情感分析中文数据集上,所提模型的Macro_F1_score值达到了0.7...     

基于情感分析的个性化推荐算法研究

对基于情感分析的个性化推荐算法进行研究。为了推荐用户可能感兴趣的产品，该算法研究了以前的评级数据和用户文本评论中的情感数据，并将其与推荐算法相结合。使用情感词典和情感分类算法对文本评论进行聚类分析，并将情感得分作为评分数据的补充，然后使用基于邻域的协同过滤算法来为用户推荐物品。使用京东评论数据集进行了实验，并与其他基于协同过滤算法进行了比较。实验结果表明，该算法能够显著提高推荐准确度和用户满意度。     

视频网站评论数据处理及分析——以B站为例

视频网站的用户评论、弹幕逐渐成为视频内容之外的一个引起热议的话题。视频评论作为新的文化产品和文化形式,引起了社会的关注。文章以B站为例,通过网络爬虫收集数据,利用大数据相关技术,对评论数据进行处理分析,并以词云图及散点图的方式呈现用户的关注点及语言表达的差异性,直观地为相关研究提供参考。     

Watching a Movie Alone yet Together: Understanding Reasons for Watching Danmaku Videos

By overlaying timeline-synchronized user comments on videos, Danmaku commenting creates a unique co-viewing experience of online videos. This study aims to understand the reasons for watching or not watching Danmaku videos. From a review of the literature and a pilot study, an initial pool of motivations and hindrances to Danmaku video viewing was gathered. Then, a survey involving 248 participants to identify the underlying factor structures of motivations and hindrances was conducted. Their influences on users’ attitude and behaviors with Danmaku videos were also examined. The results showed that people viewed Danmaku videos to obtain information, entertainment, and social connectedness. Introverted young men with high openness to new experience are more likely to view Danmaku videos. Infrequent viewers refused to watch Danmaku videos mainly because of the visual clutter that resulted from Danmaku comments.     

基于弹幕情感分析的视频片段推荐模型

针对传统的视频情感分析方法计算效率较低且结果不易解释等问题,提出一种基于弹幕文本的视频片段情感识别算法,并以此作为视频片段的推荐依据。首先对基于情感分析的视频片段推荐问题提出形式化描述。其次,通过构建基于隐含狄利克雷分布（LDA）的弹幕词语分类,评估弹幕词语在视频片段中的多维情感向量,同时,根据视频片段之间的情感依赖关系推荐视频的情感片段。所提方法的推荐准确度比基于词频-逆文档频率（TF-IDF）的推荐算法提高了28.9%,相对于传统LDA模型提高了43.8%。实验结果表明所提模型可有效应用于信息繁杂的不规则文本情感分析。     

Affective video content analysis based on multimodal data fusion in heterogeneous networks

In heterogeneous networks, different modalities are coexisting. For example, video sources with certain lengths usually have abundant time-varying audiovisual data. From the users’ perspective, different video segments will trigger different kinds of emotions. In order to better interact with users in heterogeneous networks and improve their user experiences, affective video content analysis to predict users’ emotions is essential. Academically, users’ emotions can be evaluated by arousal and valence values, and fear degree, which provides an approach to quantize the prediction accuracy of the reaction of the audience and users towards videos. In this paper, we propose the multimodal data fusion method for integrating the visual and audio data in order to perform the affective video content analysis. Specifically, to align the visual and audio data, the temporal attention filters are proposed to obtain the time-span features of the entire video segments. Then, by using the two-branch network structure, matched visual and audio features are integrated in the common space. At last, the fused audiovisual feature is employed for the regression and classification subtasks in order to measure the emotional responses of users. Simulation results show that the proposed method can accurately predict the subjective feelings of users towards the video contents, which provides a way to predict users’ preferences and recommend videos according to their own demand.     

融合评论分析和隐语义模型的视频推荐算法

针对网络视频元数据信息缺失严重和多媒体数据本身特征难以提取等问题,提出了融合评论分析和隐语义模型的网络视频推荐算法.从视频评论入手,通过分析用户对不同视频的评论内容以判断其情感倾向并加以量化,继而构建用户对项目的虚拟评分矩阵,弥补了显式评分数据稀疏性问题.考虑到网络视频的多元性和高维度特性,为了深度挖掘用户对网络视频的潜在兴趣,针对虚拟评分矩阵采用隐语义模型(LFM)对网络视频分类,在传统的用户—项目二元推荐系统基础之上添加虚拟类目信息以进一步发掘用户—类目—项目关联关系.实验在多重标准下进行,对YouTube评论集的实验表明,所提推荐方法获得了较高的推荐精度.     

Using objective ground-truth labels created by multiple annotators for improved video classification: A comparative study

We address the problem of predicting category labels for unlabeled videos in a large video dataset by using a ground-truth set of objectively labeled videos that we have created. Large video databases like YouTube require that a user uploading a new video assign to it a category label from a prescribed set of labels. Such category labeling is likely to be corrupted by the subjective biases of the uploader. Despite their noisy nature, these subjective labels are frequently used as gold standard in algorithms for multimedia classification and retrieval. Our goal in this paper is NOT to propose yet another algorithm that predicts labels for unseen videos based on the subjective ground-truth. On the other hand, our goal is to demonstrate that the video classification performance can be improved if instead of using subjective labels, we first create an objectively labeled ground-truth set of videos and then train a classifier based on such a ground-truth so as to predict objective labels for the set of unlabeled videos.     

Storyboard-based accurate automatic summary video editing system

The recent popularity of smart mobile devices has led to a significant increase in the needs of multimedia services. Finding new more efficient methods for automatic classification and retrieval of a large number of multimedia files will significantly reduce manpower costs. However, most current video content analysis methods adopt low-level features to analyze video frame by frame, and need to improve high-level semantic analysis on a number of issues. Hence, this study presents a storyboard-based accurate automatic summary video editing system that uses storyboard information, such as character dialogue, narration, caption, background music and shot changes, to enable accurate video content retrieval and automatic render summary videos. The proposed system can be applied to the course video trailer and the commercial video trailer for quick preview video content or suitable viewing configuration for smart mobile devices. Consequently, the audience can quickly understand the whole video story and the video editors can substantially reduce the time taken to publish videos.     

针对用户兴趣的视频精彩片段提取

目的 视频精彩片段提取是视频内容标注、基于内容的视频检索等领域的热点研究问题。视频精彩片段提取主要根据视频底层特征进行精彩片段的提取,忽略了用户兴趣对于提取结果的影响,导致提取结果可能与用户期望不相符。另一方面,基于用户兴趣的语义建模需要大量的标注视频训练样本才能获得较为鲁棒的语义分类器,而对于大量训练样本的标注费时费力。考虑到互联网中包含内容丰富且易于获取的图像,将互联网图像中的知识迁移到视频片段的语义模型中可以减少大量的视频数据标注工作。因此,提出利用互联网图像的用户兴趣的视频精彩片段提取框架。方法 利用大量互联网图像对用户兴趣语义进行建模,考虑到从互联网中获取的知识变化多样且有噪声,如果不加选择盲目地使用会影响视频片段提取效果,因此,将图像根据语义近似性进行分组,将语义相似但使用不同关键词检索得到的图像称为近义图像组。在此基础上,提出使用近义语义联合组权重模型权衡,根据图像组与视频的语义相关性为不同图像组分配不同的权重。首先,根据用户兴趣从互联网图像搜索引擎中检索与该兴趣语义相关的图像集,作为用户兴趣精彩片段提取的知识来源;然后,通过对近义语义图像组的联合组权重学习,将图像中习得的知识迁移到视频中;最后,使用图像集中习得的语义模型对待提取片段进行精彩片段提取。结果 本文使用CCV数据库中的视频对本文提出的方法进行验证,同时与多种已有的视频关键帧提取算法进行比较,实验结果显示本文算法的平均准确率达到46.54,较其他算法相比提高了21.6%,同时算法耗时并无增加。此外,为探究优化过程中不同平衡参数对最终结果的影响,进一步验证本文方法的有效性,本文在实验过程中通过移除算法中的正则项来验证每一项对于算法框架的影响。实验结果显示,在移除任何一项后算法的准确率明显降低,这表明本文方法所提出的联合组权重模型对提取用户感兴趣视频片段的有效性。结论 本文提出了一种针对用户兴趣语义的视频精彩片段提取方法,根据用户关注点的不同,为不同用户提取其感兴趣的视频片段。     

Exploring Online Learners' Interactive Dynamics by Visually Analyzing Their Time‐anchored Comments

MOOCs (Massive Open Online Courses) are increasingly prevalent as an online educational resource open to everyone and have attracted hundreds of thousands learners enrolling these online courses. At such scale, there is potentially rich information of learners' behaviors embedded in the interactions between learners and videos that may help instructors and content producers adjust the instructions and refine the online courses. However, the lack of tools to visualize information from interactive data, including messages left to the videos at particular timestamps as well as the temporal variations of learners' online participation and perceived experience, has prevented people from gaining more insights from video‐watching logs. In this paper, we focus on extracting and visualizing useful information from time‐anchored comments that learners left to specific time points of the videos when watching them. Timestamps as a kind of metadata of messages can be useful to recover the interactive dynamics of learners occurring around the videos. Therefore, we present a visualization system to analyze and categorize time‐anchored comments based on topics and content types. Our system integrates visualization methods of temporal text data, namely ToPIN and ThemeRiver, which can help people understand the quality and quantity of online learners' feedback and their states of learning. To evaluate the proposed system, we visualized time‐anchored commenting data from two online course videos, and conducted two user studies participated by course instructors and third‐party educational evaluators. The results validate the usefulness of the approach and show how the quantitative and qualitative visualizations can be used to gain interesting insights around learners' online learning behaviors.     

Improving performance of X-rated video classification with the optimized repeated curve-like spectrum feature and the skip-and-analysis processing

This paper addresses the automatic classification of X-rated videos by analyzing its obscene sounds. In this paper, we propose the optimized repeated curve-like spectrum feature for classifying obscene sounds and the skip-and-analysis processing for classifying videos. The optimized repeated curve-like spectrum feature uses the longer frame size for stationary frequency region based on the fact that most of obscene sounds, such as sexual moans and screams, consist of mostly vowels and the variation of syllables occurs slowly compared to general speech. It also uses the customized mel-scaled bandpass filter for the valid frequency regions of obscene sounds with the frequency contents mainly under 5 kHz. The skip-and-analysis processing is based on the video playback characteristics that a harmful or normal scene continues to be played at least for certain duration of time during a playback. When the skip-and-analysis processing is applied, clips to be analyzed are selected by skip interval values and only these selected clips are used to classify videos. The processing performances of the optimized repeated curve-like spectrum feature have improvements from 21 % to 25.6 % compared to the repeated curve-like spectrum feature without degradation of classification performance in clip-level classification. Furthermore, when the skip-and-analysis processing is applied, the processing performance of classifying is improved significantly by from 82.59 % to 95.03 % maintaining the classification performance of more than 90 % at F1-score.