Skip to main content
main-content

Über dieses Buch

This timely text/reference presents the latest advances in various aspects of social media modeling and social media computing research. Gathering together superb research from a range of established international conferences and workshops, the editors coherently organize and present each of the topics in relation to the basic principles and practices of social media modeling and computing. Individual chapters can be also be used as self-contained references on the material covered. Topics and features: presents contributions from an international selection of preeminent experts in the field; discusses topics on social-media content analysis; examines social-media system design and analysis, and visual analytic tools for event analysis; investigates access control for privacy and security issues in social networks; describes emerging applications of social media, for music recommendation, automatic image annotation, and the analysis and improvement of photo-books.

Inhaltsverzeichnis

Frontmatter

Social Media Content Analysis

Frontmatter

Quantifying Visual-Representativeness of Social Image Tags Using Image Tag Clarity

Abstract
Tags associated with images in various social media sharing web sites are valuable information source for superior image retrieval experiences. Due to the nature of tagging, many tags associated with images are not visually descriptive. In this chapter, we propose Image Tag Clarity to evaluate the effectiveness of a tag in describing the visual content of its annotated images, which is also known as the image tag visual-representativeness. It is measured by computing the zero-mean normalized distance between the tag language model estimated from the images annotated by the tag and the collection language model. The tag/collection language models are derived from the bag of visual-word local content features of the images. The visual-representative tags that are commonly used to annotate visually similar images are given high tag clarity scores. Evaluated on a large real-world dataset containing more than 269K images and their associated tags, we show that the image tag clarity score can effectively identify the visual-representative tags from all tags contributed by users. Based on the tag clarity scores, we have made a few interesting observations that could be used to support many tag-based applications.
Aixin Sun, Sourav S. Bhowmick

Tag-Based Social Image Search: Toward Relevant and Diverse Results

Abstract
Recent years have witnessed a great success of social media websites. Tag-based image search is an important approach to access the image content of interest on these websites. However, the existing ranking methods for tag-based image search frequently return results that are irrelevant or lack of diversity. This chapter presents a diverse relevance ranking scheme which simultaneously takes relevance and diversity into account by exploring the content of images and their associated tags. First, it estimates the relevance scores of images with respect to the query term based on both visual information of images and semantic information of associated tags. Then semantic similarities of social images are estimated based on their tags. Based on the relevance scores and the similarities, the ranking list is generated by a greedy ordering algorithm which optimizes Average Diverse Precision (ADP), a novel measure that is extended from the conventional Average Precision (AP). Comprehensive experiments and user studies demonstrate the effectiveness of the approach.
Kuiyuan Yang, Meng Wang, Xian-Sheng Hua, Hong-Jiang Zhang

Social Image Tag Ranking by Two-View Learning

Abstract
Tags play a central role in text-based social image retrieval and browsing. However, the tags annotated by web users could be noisy, irrelevant, and often incomplete for describing the image contents, which may severely deteriorate the performance of text-based image retrieval models. In order to solve this problem, researchers have proposed techniques to rank the annotated tags of a social image according to their relevance to the visual content of the image. In this paper, we aim to overcome the challenge of social image tag ranking for a corpus of social images with rich user-generated tags by proposing a novel two-view learning approach. It can effectively exploit both textual and visual contents of social images to discover the complicated relationship between tags and images. Unlike the conventional learning approaches that usually assumes some parametric models, our method is completely data-driven and makes no assumption about the underlying models, making the proposed solution practically more effective. We formulate our method as an optimization task and present an efficient algorithm to solve it. To evaluate the efficacy of our method, we conducted an extensive set of experiments by applying our technique to both text-based social image retrieval and automatic image annotation tasks. Our empirical results showed that the proposed method can be more effective than the conventional approaches.
Jinfeng Zhuang, Steven C. H. Hoi

Combining Multi-modal Features for Social Media Analysis

Abstract
In this chapter we discuss methods for efficiently modeling the diverse information carried by social media. The problem is viewed as a multi-modal analysis process where specialized techniques are used to overcome the obstacles arising from the heterogeneity of data. Focusing at the optimal combination of low-level features (i.e., early fusion), we present a bio-inspired algorithm for feature selection that weights the features based on their appropriateness to represent a resource. Under the same objective of optimal feature combination we also examine the use of pLSA-based aspect models, as the means to define a latent semantic space where heterogeneous types of information can be effectively combined. Tagged images taken from social sites have been used in the characteristic scenarios of image clustering and retrieval, to demonstrate the benefits of multi-modal analysis in social media.
Spiros Nikolopoulos, Eirini Giannakidou, Ioannis Kompatsiaris, Ioannis Patras, Athena Vakali

Multi-label Image Annotation by Structural Grouping Sparsity

Abstract
We can obtain high-dimensional heterogeneous features from real-world images on photo-sharing website, for an example Flickr. Those features are implemented to describe their various aspects of visual characteristics, such as color, texture and shape etc. The heterogeneous features are often over-complete to describe certain semantic. Therefore, the selection of limited discriminative features for certain semantics is hence crucial to make the image understanding more interpretable. This chapter introduces one approach for multi-label image annotation with a regularized penalty. We call it Multi-label Image Boosting by the selection of heterogeneous features with structural Grouping Sparsity (MtBGS). MtBGS induces a (structural) sparse selection model to identify subgroups of homogeneous features for predicting a certain label. Moreover, the correlations among multiple tags are utilized in MtBGS to boost the performance of multi-label annotation. Extensive experiments on public image datasets show that the proposed approach has better multi-label image annotation performance and leads to a quite interpretable model for image understanding.
Yahong Han, Fei Wu, Yueting Zhuang

Social Media System Design and Analysis

Frontmatter

Mechanism Design for Incentivizing Social Media Contributions

Abstract
Despite recent advancements in user-driven social media platforms, tools for studying user behavior patterns and motivations remain primitive. We highlight the voluntary nature of user contributions and that users can choose when (and when not) to contribute to the common media pool. A Game theoretic framework is proposed to study the dynamics of social media networks where contribution costs are individual but gains are common. We model users as rational selfish agents, and consider domain attributes like voluntary participation, virtual reward structure, network effect, and public-sharing to model the dynamics of this interaction. The created model describes the most appropriate contribution strategy from each user’s perspective and also highlights issues like ‘free-rider’ problem and individual rationality leading to irrational (i.e. sub-optimal) group behavior. We also consider the perspective of the system designer who is interested in finding the best incentive mechanisms to influence the selfish end-users so that the overall system utility is maximized. We propose and compare multiple mechanisms (based on optimal bonus payment, social incentive leveraging, and second price auction) to study how a system designer can exploit the selfishness of its users, to design incentive mechanisms which improve the overall task-completion probability and system performance, while possibly still benefiting the individual users.
Vivek K. Singh, Ramesh Jain, Mohan Kankanhalli

Efficient Access Control in Multimedia Social Networks

Abstract
Multimedia social networks (MMSNs) have provided a convenient way to share multimedia contents such as images, videos, blogs, etc. Contents shared by a person can be easily accessed by anybody else over the Internet. However, due to various privacy, security, and legal concerns people often want to selectively share the contents only with their friends, family, colleagues, etc. Access control mechanisms play an important role in this situation. With access control mechanisms one can decide the persons who can access a shared content and who cannot. But continuously growing content uploads and accesses, fine grained access control requirements (e.g. different access control parameters for different parts in a picture), and specific access control requirements for multimedia contents can make the time complexity of access control to be very large. So, it is important to study an efficient access control mechanism suitable for MMSNs. In this chapter we present an efficient bit-vector transform based access control mechanism for MMSNs. The proposed approach is also compatible with other requirements of MMSNs, such as access rights modification, content deletion, etc. Mathematical analysis and experimental results show the effectiveness and efficiency of our proposed approach.
Amit Sachan, Sabu Emmanuel

Call Me Guru: User Categories and Large-Scale Behavior in YouTube

Abstract
While existing studies on YouTube’s massive user-generated video content have mostly focused on the analysis of videos, their characteristics, and network properties, little attention has been paid to the analysis of users’ long-term behavior as it relates to the roles they self-define and (explicitly or not) play in the site. In this chapter, we present a statistical analysis of aggregated user behavior in YouTube from the perspective of user categories, a feature that allows people to ascribe to popular roles and to potentially reach certain communities. Using a sample of 270,000 users, we found that a high level of interaction and participation is concentrated on a relatively small, yet significant, group of users, following recognizable patterns of personal and social involvement. Based on our analysis, we also show that by using simple behavioral features from user profiles, people can be automatically classified according to their category with accuracy rates of up to 73%.
Joan-Isaac Biel, Daniel Gatica-Perez

Social Media Visual Analytics for Events

Abstract
For large-scale multimedia events such as televised debates and speeches, the amount of content on social media channels such as Facebook or Twitter can easily become overwhelming, yet still contain information that may aid and augment understanding of the multimedia content via individual social media items, or aggregate information from the crowd’s response. In this work we discuss this opportunity in the context of a social media visual analytic tool, Vox Civitas, designed to help journalists, media professionals, or other researchers make sense of large-scale aggregations of social media content around multimedia broadcast events. We discuss the design of the tool, present and evaluate the text analysis techniques used to enable the presentation, and detail the visual and interaction design. We provide an exploratory evaluation based on a user study in which journalists interacted with the system to analyze and report on a dataset of over one 100 000 Twitter messages collected during the broadcast of the U.S. State of the Union presidential address in 2010.
Nicholas Diakopoulos, Mor Naaman, Tayebeh Yazdani, Funda Kivran-Swaine

Social Media Applications

Frontmatter

Using Rich Social Media Information for Music Recommendation via Hypergraph Model

Abstract
There are various kinds of social media information, including different types of objects and relations among these objects, in music social communities such as Last.fm and Pandora. This information is valuable for music recommendation. However, there are two main challenges to exploit this rich social media information: (a) There are many different types of objects and relations in music social communities, which makes it difficult to develop a unified framework taking into account all objects and relations. (b) In these communities, some relations are much more sophisticated than pairwise relation, and thus cannot be simply modeled by a graph. We propose a novel music recommendation algorithm by using both multiple kinds of social media information and music acoustic-based content. Instead of graph, we use hypergraph to model the various objects and relations, and consider music recommendation as a ranking problem on this hypergraph. While an edge of an ordinary graph connects only two objects, a hyperedge represents a set of objects. In this way, hypergraph can be naturally used to model high-order relations.
Shulong Tan, Jiajun Bu, Chun Chen, Xiaofei He

Using Geotags to Derive Rich Tag-Clouds for Image Annotation

Abstract
Geotagging has become popular for many multimedia applications. In this chapter, we present an integrated and intuitive system for location-driven tag suggestion, in the form of tag-clouds, for geotagged photos. Potential tags from multiple sources are extracted and weighted. Sources include points of interest (POI) tags from a public Geographic Names Information System (GNIS) database, community tags from Flickr® pictures, and personal tags shared through users’ own, family, and friends’ photo collections. To increase the effectiveness of GNIS POI tags, bags of place-name tags are first retrieved, clustered, and then re-ranked using a combined tf-idf and spatial distance criteria. The community tags from photos taken in the vicinity of the input geotagged photo are ranked according to distance and visual similarity to the input photo. Personal tags from other personally related photos inherently carry a significant weight due more to their high relevance than to both the generic place-name tags and community tags, and are ranked by weights that decay over time and distance differences. Finally, a rich set of the most relevant location-driven tags is presented to the user in the form of individual tag clouds under the three mentioned source categories. The tag clouds act as intuitive suggestions for tagging an input image. We also discuss quantitative and qualitative findings from a user study that we conducted. Evaluation has revealed the respective benefits of the three categories toward the effectiveness of the integrated tag suggestion system.
Dhiraj Joshi, Jiebo Luo, Jie Yu, Phoury Lei, Andrew Gallagher

Social Aspects of Photobooks: Improving Photobook Authoring from Large-Scale Multimedia Analysis

Abstract
With photo albums we aim to capture personal events such as weddings, vacations, and parties of family and friends. By arranging photo prints, captions and paper souvenirs such as tickets over the pages of a photobook we tell a story to capture and share our memories. The photo memories captured in such a photobook tell us much about the content and the relevance of the photos for the user. The way in which we select photos and arrange them in the photo album reveal a lot about the events, persons and places on the photos: captions describe content, closeness and arrangement of photos express relations between photos and their content and especially about the social relations of the author and the persons present in the album. Nowadays the process of photo album authoring has become digital, photos and texts can be arranged and laid out with the help of authoring tools in a digital photo album which can be printed as a physical photobook. In this chapter we present results of the analysis of a large repository of digitally mastered photobooks to learn about their social aspects. We explore to which degree a social aspect can be identified and how expressive and vivid different classes of photobooks are. The photobooks are anonymized, real world photobooks from customers of our industry partner CeWe Color. The knowledge gained from this social photobook analysis is meant both to better understand how people author their photobooks and to improve the automatic selection of and layout of photobooks.
Philipp Sandhaus, Susanne Boll

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise