Catching a viral video

The sharing and re-sharing of videos on social sites, blogs e-mail, and other means has given rise to the phenomenon of viral videos—videos that become popular through internet sharing. In this paper we seek to better understand viral videos on YouTube by analyzing sharing and its relationship to video popularity using millions of YouTube videos. The socialness of a video is quantified by classifying the referrer sources for video views as social (e.g. an emailed link, Facebook referral) or non-social (e.g. a link from related videos). We find that viewership patterns of highly social videos are very different from less social videos. For example, the highly social videos rise to, and fall from, their peak popularity more quickly than less social videos. We also find that not all highly social videos become popular, and not all popular videos are highly social. By using our insights on viral videos we are able develop a method for ranking blogs and websites on their ability to spread viral videos.     

With a View to the Side: YouTube’s Sidebar and YouTuber’s Linguistic Style as Hints for Trust-Related Evaluations

Based on a heuristic approach to information processing, thematic reference of YouTube’s sidebars and YouTuber’s linguistic style are viewed as cues that should impact viewers’ evaluation of videos. In this study, a 2 × 2 online experiment was conducted wherein these factors were varied systematically for a video about nutrition myths. 147 participants assessed the credibility of information, YouTuber’s trustworthiness, and the self-reported learning gain. Results showed that a sidebar referring to similar (vs. unrelated) videos increased participants’ perceived trustworthiness of a YouTuber who used a YouTube-typical (vs. formal) language. Moreover, participants judged the learning gain to be higher when YouTube’s sidebar referred to similar videos. However, thematic reference of sidebar and linguistic style did not impact participants’ credibility judgments. Since people seem to recognize YouTube’s sidebars when evaluating videos, YouTube’s recommendation criteria might not only mediate videos, but also influence people’s judgments of YouTubers and videos.     

Characterizing and modelling popularity of user-generated videos

This paper develops a framework for studying the popularity dynamics of user-generated videos, presents a characterization of the popularity dynamics, and proposes a model that captures the key properties of these dynamics. We illustrate the biases that may be introduced in the analysis for some choices of the sampling technique used for collecting data; however, sampling from recently-uploaded videos provides a dataset that is seemingly unbiased. Using a dataset that tracks the views to a sample of recently-uploaded YouTube videos over the first eight months of their lifetime, we study the popularity dynamics. We find that the relative popularities of the videos within our dataset are highly non-stationary, owing primarily to large differences in the required time since upload until peak popularity is finally achieved, and secondly to popularity oscillation. We propose a model that can accurately capture the popularity dynamics of collections of recently-uploaded videos as they age, including key measures such as hot set churn statistics, and the evolution of the viewing rate and total views distributions over time.     

基于聚类层次模型的视频推荐算法

目前推荐系统存在评论数据稀疏、冷启动和用户体验度低等问题,为了提高推荐系统的性能和进一步改善用户体验,提出基于聚类层次模型的视频推荐算法。首先,从相关用户方面着手,通过近邻传播（AP）聚类分析得到相似用户,从而收集相似用户中的历史网络视频数据,进而形成视频推荐集合;其次,利用用户行为的历史数据计算出用户对视频的喜好值,再把视频的喜好值转换成视频的标签权重;最后,通过层次分析模型算出视频推荐集合中用户喜好视频的排序,产生推荐列表。基于MovieLens Latest Dataset和YouTube视频评论文本数据集,实验结果表明所提算法在均方根误差和决策精度方面均表现出良好的性能。     

Effect of metacognitive strategies and verbal-imagery cognitive style on biology-based video search and learning performance

Videos have diverse content that can assist students in learning. However, because videos are linear media, video users may take a longer time than readers of text to evaluate the context. Therefore, the process of video search may vary from one user to another depending on the users' individual characteristics, and the effectiveness of video learning may also vary across individuals. This study evaluated 100 Taiwanese fifth graders searching for videos related to “understanding animals” on YouTube and examined the effects of the students' metacognitive strategies (planning, monitoring, and evaluating) and verbal-imagery cognitive style on their video searches. The observable indicators were quantitative (search behaviors, search performance, and learning performance) and qualitative (search process observations and interviews). The study concludes that metacognitive strategy is the primary influencer of video search. Students with better metacognitive skills used fewer keywords, browsed fewer videos, and spent less time evaluating videos, but they achieved higher learning performance. They reviewed the video metadata information on the user interface and did not attempt to watch videos on the video recommendation lists, particularly videos that were irrelevant to the task requirements. During the course of the searches, keyword usage had a significant influence on the students' search performance and learning performance. The fewer keywords the students used, the better search and learning performance they were able to achieve. Our results are different from those of previous studies on text, image, and map searches. Accordingly, users must adopt different search strategies when using various types of search engines.     

Workload generation for YouTube

This paper introduces a workload characterization study of the most popular short video sharing service of Web 2.0, YouTube. Based on a vast amount of data gathered in a five-month period, we analyzed characteristics of around 250,000 YouTube popular and regular videos. In particular, we collected lists of related videos for each video clip recursively and analyzed their statistical behavior. Understanding YouTube traffic and similar Web 2.0 video sharing sites is crucial to develop synthetic workload generators. Workload simulators are required for evaluating the methods addressing the problems of high bandwidth usage and scalability of Web 2.0 sites such as YouTube. The distribution models, in particular Zipf-like behavior of YouTube popular video files suggests proxy caching of YouTube popular videos can reduce network traffic and increase scalability of YouTube Web site. YouTube workload characteristics provided in this work enabled us to develop a workload generator to evaluate the effectiveness of this approach.     

Video recommendation over multiple information sources

Video recommendation is an important tool to help people access interesting videos. In this paper, we propose a universal scheme to integrate rich information for personalized video recommendation. Our approach regards video recommendation as a ranking task. First, it generates multiple ranking lists by exploring different information sources. In particular, one novel source user’s relationship strength is inferred through the online social network and applied to recommend videos. Second, based on multiple ranking lists, a multi-task rank aggregation approach is proposed to integrate these ranking lists to generate a final result for video recommendation. It is shown that our scheme is flexible that can easily incorporate other methods by adding their generated ranking lists into our multi-task rank aggregation approach. We conduct experiments on a large dataset with 76 users and more than 11,000 videos. The experimental results demonstrate the feasibility and effectiveness of our approach.     

一种基于用户播放行为序列的个性化视频推荐策略

本文针对在线视频服务网站的个性化推荐问题,提出了一种基于用户播放行为序列的个性化推荐策略.该策略通过深度神经网络词向量模型分析用户播放视频行为数据,将视频映射成等维度的特征向量,提取视频的语义特征.聚类用户播放历史视频的特征向量,建模用户兴趣分布矩阵.结合用户兴趣偏好和用户观看历史序列生成推荐列表.在大规模的视频服务系统中进行了离线实验,相比随机算法、基于物品的协同过滤和基于用户的协同过滤传统推荐策略,本方法在用户观看视频的Top-N推荐精确率方面平均分别获得22.3%、30.7%和934%的相对提升,在召回率指标上分别获得52.8%、41%和1065%的相对提升.进一步地与矩阵分解算法SVD++、基于双向LSTM模型和注意力机制的Bi-LSTM+Attention算法和基于用户行为序列的深度兴趣网络DIN比较,Top-N推荐精确率和召回率也得到了明显提升.该推荐策略不仅获得了较高的精确率和召回率,还尝试解决传统推荐面临大规模工业数据集时的数据要求严苛、数据稀疏和数据噪声等问题.     

Boosting web video categorization with contextual information from social web

Web video categorization is a fundamental task for web video search. In this paper, we explore web video categorization from a new perspective, by integrating the model-based and data-driven approaches to boost the performance. The boosting comes from two aspects: one is the performance improvement for text classifiers through query expansion from related videos and user videos. The model-based classifiers are built based on the text features extracted from title and tags. Related videos and user videos act as external resources for compensating the shortcoming of the limited and noisy text features. Query expansion is adopted to reinforce the classification performance of text features through related videos and user videos. The other improvement is derived from the integration of model-based classification and data-driven majority voting from related videos and user videos. From the data-driven viewpoint, related videos and user videos are treated as sources for majority voting from the perspective of video relevance and user interest, respectively. Semantic meaning from text, video relevance from related videos, and user interest induced from user videos, are combined to robustly determine the video category. Their combination from semantics, relevance and interest further improves the performance of web video categorization. Experiments on YouTube videos demonstrate the significant improvement of the proposed approach compared to the traditional text based classifiers.     

Characteristics of YouTube network traffic at a campus network – Measurements,models, and implications

User-Generated Content has become very popular since new web services such as YouTube allow for the distribution of user-produced media content. YouTube-like services are different from existing traditional VoD services in that the service provider has only limited control over the creation of new content. We analyze how content distribution in YouTube is realized and then conduct a measurement study of YouTube traffic in a large university campus network. Based on these measurements, we analyzed the duration and the data rate of streaming sessions, the popularity of videos, and access patterns for video clips from the clients in the campus network. The analysis of the traffic shows that trace statistics are relatively stable over short-term periods while long-term trends can be observed. We demonstrate how synthetic traces can be generated from the measured traces and show how these synthetic traces can be used as inputs to trace-driven simulations. We also analyze the benefits of alternative distribution infrastructures to improve the performance of a YouTube-like VoD service. The results of these simulations show that P2P-based distribution and proxy caching can reduce network traffic significantly and allow for faster access to video clips.     

Real-Time Near-Duplicate Elimination for Web Video Search With Content and Context

With the exponential growth of social media, there exist huge numbers of near-duplicate web videos, ranging from simple formatting to complex mixture of different editing effects. In addition to the abundant video content, the social Web provides rich sets of context information associated with web videos, such as thumbnail image, time duration and so on. At the same time, the popularity of Web 2.0 demands for timely response to user queries. To balance the speed and accuracy aspects, in this paper, we combine the contextual information from time duration, number of views, and thumbnail images with the content analysis derived from color and local points to achieve real-time near-duplicate elimination. The results of 24 popular queries retrieved from YouTube show that the proposed approach integrating content and context can reach real-time novelty re-ranking of web videos with extremely high efficiency, where the majority of duplicates can be rapidly detected and removed from the top rankings. The speedup of the proposed approach can reach 164 times faster than the effective hierarchical method proposed in , with just a slight loss of performance.     

Lyapunov stability and performance of user-assisted Video-on-Demand services

Currently, most video on-demand services offered over the Internet do not exploit the idle resources available from end-users, including YouTube. We present a taxonomic analysis of user-assistance in video on-demand systems, where users are both clients and servers, helping with the task of video distribution. From a theoretical perspective, we develop a deterministic fluid model suitable for sequential systems. We mathematically prove the Peer-to-Peer Sequential Fluid Model is globally stable in the Lyapunov sense, no matter the network parameters of the cooperative system. We theoretically prove that cooperative systems always outperform non-cooperative solutions. From a practical point of view, a caching problem is proposed and discussed in order to tackle technological concerns to massively distribute popular videos on-demand. The goal is to distribute video items into repositories minimizing the waiting times of end-users. The caching problem is inside the class of NP-Complete computational problems, and heuristically solved with a GRASP methodology enriched with a path-relinking technique. Predictions inspired in a statistical analysis of real-life YouTube traces suggest the introduction of cooperation is both robust and economically attractive. These results highlight the harmony between our theoretical development and practice.     

Efficient indexing and retrieval of large-scale geo-tagged video databases

We are witnessing a significant growth in the number of smartphone users and advances in phone hardware and sensor technology. In conjunction with the popularity of video applications such as YouTube, an unprecedented number of user-generated videos (UGVs) are being generated and consumed by the public, which leads to a Big Data challenge in social media. In a very large video repository, it is difficult to index and search videos in their unstructured form. However, due to recent development, videos can be geo-tagged (e.g., locations from GPS receiver and viewing directions from digital compass) at the acquisition time, which can provide potential for efficient management of video data. Ideally, each video frame can be tagged by the spatial extent of its coverage area, termed Field-Of-View (FOV). This effectively converts a challenging video management problem into a spatial database problem. This paper attacks the challenges of large-scale video data management using spatial indexing and querying of FOVs, especially maximally harnessing the geographical properties of FOVs. Since FOVs are shaped similar to slices of pie and contain both location and orientation information, conventional spatial indexes, such as R-tree, cannot index them efficiently. The distribution of UGVs’ locations is non-uniform (e.g., more FOVs in popular locations). Consequently, even multilevel grid-based indexes, which can handle both location and orientation, have limitations in managing the skewed distribution. Additionally, since UGVs are usually captured in a casual way with diverse setups and movements, no a priori assumption can be made to condense them in an index structure. To overcome the challenges, we propose a class of new R-tree-based index structures that effectively harness FOVs’ camera locations, orientations and view-distances, in tandem, for both filtering and optimization. We also present novel search strategies and algorithms for efficient range and directional queries on our indexes. Our experiments using both real-world and large synthetic video datasets (over 30 years’ worth of videos) demonstrate the scalability and efficiency of our proposed indexes and search algorithms.     

移动网络YouTube视频码率及分辨率识别方法

YouTube作为世界最大的视频网站,在网络流量中的占比日益增加。大量的视频流量给互联网服务提供商带来了巨大的挑战。随着YouTube全站采用HTTPS加密,如何从HTTPS加密流量中获取视频QoE评估信息至关重要。在分析了YouTube App在Android和IOS平台上所采用的流媒体传输模式的基础上,提出了C4.5决策树与[K]-means聚类相结合的方法对视频片段进行分辨率识别。实验结果证明该方法能够准确地识别视频播放过程中每个视频片段的播放码率及分辨率。     

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.     

On the tag localization of web video

Nowadays, numerous social videos have pervaded on the web. Social web videos are characterized with the accompanying rich contextual information which describe the content of videos and thus greatly facilitate video search and browsing. Generally, those contextual data such as tags are provided at the whole video level, without temporal indication of when they actually appear in the video, let alone the spatial annotation of object related tags in the video frames. However, many tags only describe parts of the video content. Therefore, tag localization, the process of assigning tags to the underlying relevant video segments or frames even regions in frames is gaining increasing research interests and a benchmark dataset for the fair evaluation of tag localization algorithms is highly desirable. In this paper, we describe and release a dataset called DUT-WEBV, which contains about 4,000 videos collected from YouTube portal by issuing 50 concepts as queries. These concepts cover a wide range of semantic aspects including scenes like “mountain”, events like “flood”, objects like “cows”, sites like “gas station”, and activities like “handshaking”, offering great challenges to the tag (i.e., concept) localization task. For each video of a tag, we carefully annotate the time durations when the tag appears in the video and also label the spatial location of object with mask in frames for object related tag. Besides the video itself, the contextual information, such as thumbnail images, titles, and YouTube categories, is also provided. Together with this benchmark dataset, we present a baseline for tag localization using multiple instance learning approach. Finally, we discuss some open research issues for tag localization in web videos.     

A dynamic caching algorithm based on internal popularity distribution of streaming media

Most proxy caches for streaming videos do not cache the entire video but only a portion of it. This is partly due to the large size of video objects. Another reason is that the popularity of different parts of a video can be different, e.g., the prefix is generally more popular. Therefore, the development of efficient cache mechanisms requires an understanding of the internal popularity characteristics of streaming videos. This paper has two major contributions. Firstly, we analyze two 6-month long traces of RTSP video requests recorded at different streaming video servers of an entertainment video-on-demand provider, and show that the traces provide evidence that the internal popularity of the majority of the most popular videos obeys a k-transformed Zipf-like distribution. Secondly, we propose a caching algorithm which exploits this empirical internal popularity distribution. We find that this algorithm has similar performance compared with fine-grained caching but requires significantly less state information.     

一种面向多维特征分析过滤的视频推荐算法

近年来,抖音、快手、微视等短视频APP取得了巨大成功,用户拍摄并上传到APP平台上的视频数量暴增。在这种信息过载的环境下,为用户挖掘并推荐其感兴趣的视频成为了视频发布平台面临的难题,因此为这些平台设计高效的视频推荐算法显得尤其重要。文中针对媒体大数据挖掘和推荐领域的数据集稀疏性高和规模巨大的问题,提出一种面向多维特征分析过滤的视频推荐算法。首先,从用户行为和视频标签等多个维度对视频进行特征提取,然后进行相似性分析,加权计算视频相似度,从而获取相似视频候选集,并对相似视频候选集进行过滤,再通过排序选择评分最高的若干个视频推荐给用户。最后,基于MovieLens公开数据集,使用python3语言实现了文中提出的视频推荐算法。在数据集上进行的大量实验表明,相比传统的协同过滤算法,文中提出的面向多维特征分析过滤的视频推荐算法将推荐结果的准确率提升了6%,召回率提升了4%,覆盖率提升了18%。实验数据充分说明,从多个维度考虑视频之间的相似性,并配合大规模矩阵分解技术,在一定程度上缓解了数据集稀疏性高、数据量巨大的难题,从而有效地提高了推荐结果的准确性、召回率和覆盖率。     

Transfer Learning on Deep Neural Networks to Detect Pornography

While the internet has a lot of positive impact on society, there are negative components. Accessible to everyone through online platforms, pornography is, inducing psychological and health related issues among people of all ages. While a difficult task, detecting pornography can be the important step in determining the porn and adult content in a video. In this paper, an architecture is proposed which yielded high scores for both training and testing. This dataset was produced from 190 videos, yielding more than 19 h of videos. The main sources for the content were from YouTube, movies, torrent, and websites that hosts both pornographic and non-pornographic contents. The videos were from different ethnicities and skin color which ensures the models can detect any kind of video. A VGG16, Inception V3 and Resnet 50 models were initially trained to detect these pornographic images but failed to achieve a high testing accuracy with accuracies of 0.49, 0.49 and 0.78 respectively. Finally, utilizing transfer learning, a convolutional neural network was designed and yielded an accuracy of 0.98.