Social Networks and Multimedia Content Analysis: Mădălina-Steliana Deaconu

Economics, Management, and Financial Markets
Volume 5(4), 2010, pp. 208–213, ISSN 1842-3191
SOCIAL NETWORKS AND

MULTIMEDIA CONTENT ANALYSIS
MĂDĂLINA-STELIANA DEACONU
[email protected]
Titu Maiorescu University
ABSTRACT. We seek to contribute to a wider understanding of contextual

and cultural influences on human satisfaction with search for knowledge and
information in multimedia content, the advantages and disadvantages of
social networks as seen by users or society, using Web mining to extract
social network information, and the effectiveness of computers as social
actors. Considerable research attention has focused on a sound knowledge of
writing styles for on-screen reading, artificially-intelligent algorithms, the
concept of journalism in the digital age, and new global standards for digital
identities.
JEL: L63, L82, Z13
Keywords: social network, multimedia content, on-screen reading
1. Introduction
The variety of applications in which automatic, computer-based

multi-media content analysis, today’s multi-media vast journalism
content, and the analytical tools of data mining and artificial intel-
ligence can be utilized has dramatically increased interest in this area.
This has resulted in the development of an extensive body of liter-
ature concerning the importance of the interactions and the relevance
of individual persons in the network, the connection between the
users’ digital identities and their personalities, and the development
and improvement of management techniques of digital identities.
These findings highlight the importance of examining artificial social
intelligence in computing and signal processing communities, and
targeting journalistic content via digital identities.
208
2. The Development and Improvement of
Management Techniques of Digital Identities
Producing high-quality usable, credible digital text is a professional

skill.1 Distance learning permits the possibility of economies of scale.
Economy of scale is less apparent in online discussion settings. 2
Vinciarelli et al. state that the nonverbal behaviour plays a major role
in shaping the perception of social situations. Postures are typically
assumed unconsciously. Context plays a crucial role in understanding
of human behavioural signals. Automated analyzers of human social
signals and social behaviours should be multimodal (social signals
are intrinsically ambiguous). Humans display audio and visual ex-
pressions in a complementary and redundant manner, and do not use
the behavioral cues actually at their disposition. Nonverbal behaviours
are not unequivocally associated to a specific meaning. The use of
real-world data shows the actual effects of goals and motivations that
typically drive human behaviour. The analysis of group interactions
involves the need of observing multiple people involved in a large
number of one-to-one interactions.3
Vinciarelli points out that humans appear to be literally wired for
social interaction, while computers are the natural means for a wide
spectrum of new, inherently social activities. Computers must deal
effectively with spontaneous social behaviors of their users. An im-
portant facet of social intelligence is about understanding of socially
relevant behavioral patterns. Turn-taking patterns can be analyzed and
interpreted in terms of roles that people play (they are key evidence
of social interaction processes). Vinciarelli maintains that roles actually
bring order and predictability in turn-taking. Role related turn-taking
patterns are evident enough to achieve satisfactory performances.
Several social phenomena leave physical, machine detectable traces
in terms of predictable behavioral patterns. The integration of social
psychology into automatic approaches is effective and leads to a
form of artificial social intelligence. Vinciarelli posits that people
display social signals in terms of aggregates of nonverbal behavioral
cues. SSP (Social Signal Processing) investigates how dynamics of
human-human interaction can be applied to human-machine interaction.
Automatic approaches are capable of correctly understanding social
phenomena without integrating the findings of human sciences. In
most cases, social interactions analysis is reliable only if several
behavioral cues are analyzed jointly. Vinciarelli reasons that using
209
multiple cues is the only way to improve robustness of understanding
approaches. Integration between human sciences and technology is a
key towards success.4
3. Today’s Multi-media Vast Journalism Content
Latar and Nordfors focus on “journalistic content behavior targeting”

vs. Digital Identities (DI). Survival of the world economy in the age
of information depends on introducing automatic decision-making
procedures to maintain competitiveness. Journalistic content comprises
multimedia-format news of information and cultural products. Letting
digital identities filter content automatically will engender serious social
consequences. Journalistic content is undergoing major changes due
to the new, technological interactive platforms. The latest generation
of producers of journalism focuses on producing content and pub-
lishing on-line. On the Internet the new media companies can offer
their customers hard data on which user looked at what, where, when
and for how long. Latar and Nordfors assert that the new media
companies sell actual attention. Data mining is a process of obtaining
new knowledge by analyzing digital databases, randomly constructed
using smart algorithms. Constructing the digital identity is a dynamic
process (digital identity management is developing rapidly). Every
surfer has a unique, dynamic way of surfing rendered to automatic
cognitive diagnosis through the artificial-intelligence algorithms. Most
major social networks ask joiners to define their personal profiles
and in return offer applications. Some major networks see their pur-
pose today in providing services, information and products adapted
to members’ digital identities. At the heart of the social network is
the graph of contacts between network members. Social networks
include complete mapping of surfers’ social and professional con-
nections. Latar and Nordfors emphasize that functioning in virtual
worlds influence peoples’ consciousness and actions in reality. Through
a person’s activity in virtual worlds you can learn about behavior,
personality and opinions in the real world. There is no longer a dif-
ference between infrastructure used for one-to-one or one-to-many
communication. Journalism interacts with augmented intelligence of
individuals, the collective intelligence of society, and the artificial
intelligence of machines. The integration between the automatic com-
puter content analysis and the personal digital identities will allow
for a meaningful, personally adjusted content. New media platforms
210
make the competition for consumer attention fierce and complex. In
Latar and Nordfors’ view, context within which content is delivered
is becoming of prime importance. Authentic transfer of information
is essential for automated decision-making processes. The consumers’
digital identities directly affect the type of services and information
they will receive in their lifetime. Producing a successful story is
equal to solving a dynamic equation involving the journalist, the
audience and the business model. The interaction between digital
identities may improve the outcome for all parties involved.5
Zahariadis et al. explain that the emergence and popularity of
online social networks has led to a more collaborative environment.
Social networks provide a powerful refl ection of the structure and
dynamics of the society. Users both produce and consume significant
quantities of multimedia content. Graph analysis and social network
analysis are valuable tools for studying the web and human behav-
iours of the web users.6
4. Semantic Computing and

Knowledge Models of Social Networks
Sheu reports that semantic computing addresses the derivation and

matching of the semantics of computational content. Semantic analysis
provides the information resource for semantic integration and semantic
services. Semantic computing aims to provide more powerful com-
puting services for all kinds of users. Sometime semantic services
are limited or insufficient for applications requiring several services
working together. Natural language interface, multimodal interface,
and visual interface are becoming increasingly important.7 Deng et
al. hold that many computer-based speech recognizers do not try to
interpret, or use the meaning in, the recognition process. the under-
lying paradigm for the technology is based on the information chan-
nel model of speech communication. 8 Desikan et al. contend that
large-scale social dynamics on social networking websites can be
studied using link analysis. Compact structures for capturing infor-
mation entities can be efficiently represented as directed or undirected
graphs. Semantic Web captures the knowledge models of social net-
works: they provide a useful tool to manage social information and
beliefs of people and their interactions.9
211
Jung and Euzenat focus on using the structure of knowledge used
by people in order to extract meaningful relations at the social level,
proposing a three-layered architecture for constructing semantic social
network (a social layer, an ontology layer, and a concept layer).
Semantic social network analysis helps people to find other peers
with similar interests, and helps peers to find the best company for
starting designing consensus ontologies. In social networks it is normal
to be connected to several authorities. The number of ontologies is
playing an important role in social affinity.10 Deuze considers the
highly individualized, informally networked and contingent context
of media work, stressing that our world is a completely mediated
public space where media underpin and overarch the experiences of
everyday life. The ways in which media professionals make sense of
their work on an individual level are becoming part of the lives of
media users. Media industries produce content and invest in platforms
for connectivity. Deuze argues that the roles played by advertising
creatives, media producers, and content consumers are intertwined.
The networked form of enterprise is typical for media work. Media
organizations brace themselves for inter- and interinstitutional collab-
orations and cross-media production.11
5. Conclusions
The results of the current study converge with prior research on the
interactive nature of screen texts, the construction of personality-
linked databases, the formation of digital identities, the infrastruc-
tures of the new content-neutral media entities, and the reaction of
users to social signals exhibited by computer characters. In sum, the
results of the current study provide useful insights on spreading jour-
nalistic content based on profiling, meaningful automatic journalistic
multimedia content, the effectiveness of detecting social signals, the
rapid growth in popularity of social networks, and the way social
phenomena shape turn-taking. In addition, this paper provides im-
plications for practice and research to further explore the dramatic
growth of social multimedia and user generated content, the advance
of the interactive nature of new media, the importance of social
signals in everyday life, social networking sites as the primary chan-
nel of communicating ideas and sharing media, and integrating human
feedback and involvement in the automatic multi-model content
analysis.
212
REFERENCES
1. Jiménez-Crespo, M.A. (2010), “Localization and Writing for a New

Medium: A Review of Digital Style Guides,” Revista Tradumàtica 8: online.
2. Cook, D.A. (2007), “Web-based Learning: Pros, Cons and Controver-
sies,” Clinical Medicine 7(1): 37–42.
3. Vinciarelli, A. et al. (2009), “Social Signal Processing: Survey of an
Emerging Domain,” Image and Vision Computing Journal 27(12): 1743–1759.
4. Vinciarelli, A. (2009), “Capturing Order in Social Interactions,”
IEEE Signal Processing Magazine 26(12): 133–137.
5. Latar, N.L. and Nordfors, D. (2009), Digital Identities and Journalism
Content: How Artificial Intelligence and Journalism May Co-Develop and
Why Society Should Care,” Innovation Journalism 6(7): online.
6. Zahariadis, Th. et al. (2010), Social Networks Overview: Current
Trends and Research Challenges. Luxembourg: Publications Office of the
European Union.
7. Sheu, P. (2010), “Semantic Computing,” in Sheu, P. et al. (eds.),
Semantic Computing. Hoboken, NJ: John Wiley & Sons, 1–10.
8. Li Deng et al. (2010), “A Semantic and Detection-Based Approach to
Speech and Language Processing,” [7], 49–68.
9. Desikan, P. et al. (2010), “Link Analysis in Web Mining: Techniques
and Applications,” [7], 69–86.
10. Jung, J. and Euzenat, J. (2007), “Towards Semantic Social Networks,”
Proceedings of the 4th European Semantic Web Conference, Innsbruck, June,
Lecture Notes in Artificial Intelligence 4519. Dordrecht: Springer, 267–280.
11. Deuze, M. (2009), “The Media Logic of Media Work,” Journal of
Media Sociology 1(1/2): 22–40.
© Mădălina-Steliana Deaconu
213
Copyright of Economics, Management & Financial Markets is the property of Addleton Academic Publishers
and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for individual use.

Social Networks and Multimedia Content Analysis: Mădălina-Steliana Deaconu

Uploaded by

Copyright:

Available Formats

Social Networks and Multimedia Content Analysis: Mădălina-Steliana Deaconu

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Social Networks and Multimedia Content Analysis: Mădălina-Steliana Deaconu

Uploaded by

Copyright:

Available Formats

Economics, Management, and Financial Markets

Volume 5(4), 2010, pp. 208–213, ISSN 1842-3191

SOCIAL NETWORKS AND

ABSTRACT. We seek to contribute to a wider understanding of contextual

Keywords: social network, multimedia content, on-screen reading

The variety of applications in which automatic, computer-based

Producing high-quality usable, credible digital text is a professional

3. Today’s Multi-media Vast Journalism Content

Latar and Nordfors focus on “journalistic content behavior targeting”

4. Semantic Computing and

Sheu reports that semantic computing addresses the derivation and

1. Jiménez-Crespo, M.A. (2010), “Localization and Writing for a New

You might also like