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.     

Agent-mediated shared conceptualizations in tagging services

Some of the most remarkable innovative technologies from the Web 2.0 are the collaborative tagging systems. They allow the use of folksonomies as a useful structure for a number of tasks in the social web, such as navigation and knowledge organization. One of the main deficiencies comes from the tagging behaviour of different users which causes semantic heterogeneity in tagging. As a consequence a user cannot benefit from the adequate tagging of others. In order to solve the problem, an agent-based reconciliation knowledge system, based on Formal Concept Analysis, is applied to facilitate the semantic interoperability between personomies. This article describes experiments that focus on conceptual structures produced by the system when it is applied to a collaborative tagging service, Delicious. Results will show the prevalence of shared tags in the sharing of common resources in the reconciliation process.     

Incorporating user motivations to design for video tagging

User video tagging can enhance the indexing of large collections of videos, or can provide the basis for personalizing output. However, before the benefits of tagging can be reaped, users must be motivated to provide videos with tags. This article describes a two-stage study that aimed at collecting the most important motivations for users to tag video material. First, focus groups with internet users were held to elicit all possible motivations to tag videos on the internet. Next, 125 persons ranked these motivations for two cases via an online survey and responded to statements that assessed their acceptance of personalized output, based on their tags. Motivations related to indexing appear to be far more important for people than motivations related to socializing or communication. Furthermore, people were moderately positive about personalized output, based on their tags. Finally, important user barriers to tagging are discussed.     

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。     

Folksonomy‐Based Recommender Systems: A State‐of‐the‐Art Review

Collaborative tagging systems, also known as folksonomies, have grown in popularity over the Web on account of their simplicity to organize several types of content (e.g., Web pages, pictures, and video) using open‐ended tags. The rapid adoption of these systems has led to an increasing amount of users providing information about themselves and, at the same time, a growing and rich corpus of social knowledge that can be exploited by recommendation technologies. In this context, tripartite relationships between users, resources, and tags contained in folksonomies set new challenges for knowledge discovery approaches to be applied for the purposes of assisting users through recommendation systems. This review aims at providing a comprehensive overview of the literature in the field of folksonomy‐based recommender systems. Current recommendation approaches stemming from fields such as user modeling, collaborative filtering, content, and link‐analysis are reviewed and discussed to provide a starting point for researchers in the field as well as explore future research lines.     

基于标签聚类与项目主题的协同过滤推荐算法

传统基于项目的协同过滤算法在计算项目相似度时仅依靠评分数据,未考虑项目的自身特征。社会化标注的出现使得标签能在一定程度上反映项目特征,但标签具有语义模糊的特点,因此直接将标签纳入协同过滤算法存在一定问题。为解决上述问题,提出一种改进的基于项目的协同过滤推荐算法。该算法对标签进行聚类并生成主题标签簇,根据项目标注情况计算项目与主题间的相关度并生成项目-主题相关度矩阵,同时将其与项目-评分矩阵相结合来计算项目间的相似度,采用协同过滤完成对目标项目的评分预测,以实现个性化推荐。在Movielens数据集上的实验结果表明,该算法能够解决标签的语义模糊问题并提升推荐质量。     

基于标签、得分和偏好时效性的项目推荐方法

网络信息的爆炸式增长使得推荐系统成为一项研究的热点。现存的推荐系统在实际运营中存在各自的缺陷。在web2.0环境下,标签、项目得分以及用户标注项目的时间均包含暗示用户偏好的重要信息,这些信息对提高推荐系统准确度是十分重要的。在借鉴协同过滤思想的基础上,提出综合考虑标签、项目得分和用户偏好时效性的项目推荐模型,并对此模型的体系结构及应用前景进行了分析。     

Integrating ontology technology with folksonomies for personalized social tag recommendation

Social tagging systems leverage social interoperability by facilitating the searching, sharing, and exchanging of tagging resources. A major drawback of existing social tagging systems is that social tags are used as keywords in keyword-based search. They focus on keywords and human interpretability rather than on computer interpretable semantic knowledge. Therefore, social tags are useful for information sharing and organizing, but they lack the computer-interpretability needed to facilitate a personalized social tag recommendation. An interesting issue is how to automatically generate a personalized social tag recommendation list to users when a resource is accessed by users. The novel solution proposed in this study is a hybrid approach based on semantic tag-based resource profile and user preference to provide personalized social tag recommendation. Experiments show that the Precision and Recall of the proposed hybrid approach effectively improves the accuracy of social tag recommendation.     

Personalized topic-based tag recommendation

More and more content on the Web is generated by users. To organize this information and make it accessible via current search technology, tagging systems have gained tremendous popularity. Especially for multimedia content they allow to annotate resources with keywords (tags) which opens the door for classic text-based information retrieval. To support the user in choosing the right keywords, tag recommendation algorithms have emerged. In this setting, not only the content is decisive for recommending relevant tags but also the user's preferences.In this paper we introduce an approach to personalized tag recommendation that combines a probabilistic model of tags from the resource with tags from the user. As models we investigate simple language models as well as Latent Dirichlet Allocation. Extensive experiments on a real world dataset crawled from a big tagging system show that personalization improves tag recommendation, and our approach significantly outperforms state-of-the-art approaches.     

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.     

基于LDA的隐式标签协同过滤推荐算法

固定标签协同过滤推荐算法,未充分考虑标签因子的多样化,主要依靠人工标记,扩展性不强,主观因素多。本文从用户的喜好特征因素角度出发,在固定标签协同过滤推荐算法的基础上,提出一种隐式标签协同过滤推荐算法。该算法利用LDA主题模型生成项目文本的隐式标签,得到项目-标签特征权重,根据算法性能优化的要求选择标签数量,将项目-标签矩阵与用户评分矩阵结合得到用户对标签的偏好矩阵,最后通过协同过滤算法产生推荐。实验结果表明,本文提出的基于LDA的隐式标签协同过滤推荐算法缓解了数据稀疏性问题,项目推荐的召回率、准确度和F1值有较大提升。     

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。     

SHIATSU: tagging and retrieving videos without worries

The dramatic growth of video content over modern media channels (such as the Internet and mobile phone platforms) directs the interest of media broadcasters towards the topics of video retrieval and content browsing. Several video retrieval systems benefit from the use of semantic indexing based on content, since it allows an intuitive categorization of videos. However, indexing is usually performed through manual annotation, thus introducing potential problems such as ambiguity, lack of information, and non-relevance of index terms. In this paper, we present SHIATSU, a complete system for video retrieval which is based on the (semi-)automatic hierarchical semantic annotation of videos exploiting the analysis of visual content; videos can then be searched by means of attached tags and/or visual features. We experimentally evaluate the performance of SHIATSU on two different real video benchmarks, proving its accuracy and efficiency.     

Exploring sharing patterns for video recommendation on YouTube-like social media

YouTube-like video sharing sites (VSSes) have gained increasing popularity in recent years. Meanwhile, Face-book-like online social networks (OSNs) have seen their tremendous success in connecting people of common interests. These two new generation of networked services are now bridged in that many users of OSNs share video contents originating from VSSes with their friends, and it has been shown that a significant portion of views of VSS videos are attributed to this sharing scheme of social networks. To understand how the video sharing behavior, which is largely based on social relationship, impacts users’ viewing pattern, we have conducted a long-term measurement with RenRen and YouKu, the largest online social network and the largest video sharing site in China, respectively. We show that social friends have higher common interest and their sharing behaviors provide guidance to enhance recommended video lists. In this paper, we take a first step toward learning OSN video sharing patterns for video recommendation. An autoencoder model is developed to learn the social similarity of different videos in terms of their sharing in OSNs. We, therefore, propose a similarity-based strategy to enhance video recommendation for YouTube-like social media. Evaluation results demonstrate that this strategy can remarkably improve the precision and recall of recommendations, as compared to other widely adopted strategies without social information.     

面向视频的细粒度多模态实体链接

随着互联网和大数据的飞速发展,数据规模越来越大,种类也越来越多.视频作为其中重要的一种信息方式,随着近期短视频的发展,占比越来越大.如何对这些大规模视频进行理解分析,成为学界关注的热点.实体链接作为一种背景知识补全方式,可以提供丰富的外部知识.视频上的实体链接可以有效地帮助理解视频内容,从而实现对视频内容的分类、检索、推荐等.但是现有的视频链接数据集和方法的粒度过粗,因此提出面向视频的细粒度实体链接,并立足于直播场景,构建了细粒度视频实体链接数据集.此外,依据细粒度视频链接任务的难点,提出利用大模型抽取视频中的实体及其属性,并利用对比学习得到视频和对应实体的更好表示.实验结果表明,该方法能够有效地处理视频上的细粒度实体链接任务.     

A framework for tag-aware recommender systems

In social tagging system, a user annotates a tag to an item. The tagging information is utilized in recommendation process. In this paper, we propose a hybrid item recommendation method to mitigate limitations of existing approaches and propose a recommendation framework for social tagging systems. The proposed framework consists of tag and item recommendations. Tag recommendation helps users annotate tags and enriches the dataset of a social tagging system. Item recommendation utilizes tags to recommend relevant items to users. We investigate association rule, bigram, tag expansion, and implicit trust relationship for providing tag and item recommendations on the framework. The experimental results show that the proposed hybrid item recommendation method generates more appropriate items than existing research studies on a real-world social tagging dataset.     

WESONet: Applying semantic web technologies and collaborative tagging to multimedia web information systems

The publication of different media types, like images, audio and video in the World Wide Web is getting more importance each day. However, searching and locating content in multimedia sites is challenging. In this paper, we propose a platform for the development of multimedia web information systems. Our approach is based on the combination between semantic web technologies and collaborative tagging. Producers can add meta-data to multimedia content associating it with different domain-specific ontologies. At the same time, users can tag the content in a collaborative way. The proposed system uses a search engine that combines both kinds of meta-data to locate the desired content. It will also provide browsing capabilities through the ontology concepts and the developed tags.     

基于协作式标注图像数据的垃圾标签检测方法

由于用户标签的不准确和语义模糊使得协作式标注图像检索正确率低,而现有垃圾标签过滤方法往往关注标签本身,忽略了协作式标签与图像的关联性。本文在分析协作式标注图像视觉内容与标签的关联性的基础上,提出一种基于协作式标注图像视觉内容的垃圾标签检测方法。该方法分析同一标签下图像视觉内容,设计不同的核函数用于颜色和SIFT(Scale invariant feature transform)特征子集,同时将2种低维特征映射到高维多模特征空间形成混合核函数,对同一标签下的图像进行基于混合核的最大最小距离聚类,少数群体的标签说明与图像内容关联性小则为用户标注错误的标签,从而检测垃圾标签。实验结果表明,该方法能够提高协作式图像垃圾标签检测的正确性。     

Personalized Tag Recommendation Using Social Influence

Tag recommendation encourages users to add more tags in bridging the semantic gap between human concept and the features of media object,which provides a feasible solution for content-based multimedia information retrieval.In this paper,we study personalized tag recommendation in a popular online photo sharing site - Flickr.Social relationship information of users is collected to generate an online social network.From the perspective of network topology,we propose node topological potential to characterize user’s social influence.With this metric,we distinguish different social relations between users and find out those who really have influence on the target users.Tag recommendations are based on tagging history and the latent personalized preference learned from those who have most influence in user’s social network.We evaluate our method on large scale real-world data.The experimental results demonstrate that our method can outperform the non-personalized global co-occurrence method and other two state-of-the-art personalized approaches using social networks.We also analyze the further usage of our approach for the cold-start problem of tag recommendation.