Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Graph-based social media analysis
Pitas I., Chapman & Hall/CRC, Boca Raton, FL, 2016. 442 pp. Type: Book (978-1-498719-04-9)
Date Reviewed: Feb 14 2017

The proliferation of data from social media is presenting new opportunities for discovering and analyzing social networks. Much of this data, however, is unstructured, presenting challenges in both constructing and inferring information from these networks. Graph analytics is a tool that has proven to be valuable in solving these problems. Graph-based media analysis is focused on this confluence of network theory and graph analytics.

The book consists of 12 chapters. The first three chapters give a broad overview of the foundational material that includes the dominant social media platforms, extracting data from these platforms using their corresponding application programming interfaces (APIs), representing this data as networks using graph structures, computational complexity of graph-based social media analysis methods, their memory requirements, and processing using distributed computing platforms. Basic linear algebra and algebraic graph analysis techniques that can be used for graph clustering, community detection, graph matching, and graph anomaly detection are discussed.

Chapter 4 gives an overview of information retrieval methods and looks into searching and ranking web pages most relevant to a user’s interest. Ranking using centrality measures that utilize the hyperlink structure and network topology of web pages are discussed along with topic-sensitive ranking and ranking in heterogeneous networks.

Multimedia objects are tagged by labels that describe their semantic content. These labels are, however, not always provided and, when present, may not be accurate. Manually reviewing and labeling vast amounts of data is also not practical. The next several chapters look at solutions to many of these interesting aspects. Chapter 5 investigates graph-based label propagation algorithms, a class of semi-supervised classifiers, that automatically disseminate labels from a small set of labeled objects to the larger set of data. Once labeled, processing large volumes of data and the metadata that characterizes them can be challenging. To improve performance, chapters 6 and 7 look at dimensionality reduction techniques that derive low-dimensional data retaining properties of interest from a high-dimensional dataset.

Information retrieval based solely on media content often yields unsatisfactory results. Multimedia social search, a topic of chapter 8, takes into account not the social media content, but its related social context such as ownership, tags, geolocation, and so on when conducting a search. Together these relationships form a hypergraph, a representation that possesses strong mathematical foundations allowing one to cast the multimedia social search problem into a clustering and ranking problem. Chapter 11 compliments chapter 8 by exploring other aspects of multimedia social search, such as temporal evolution, latent model adaptation, incremental spectral clustering, incremental hypermatrix factorization, and scalability.

Chapters 9 and 10 look at applications of digital signal processing to graph analysis. Signal processing techniques can be used to overcome some of the challenges such as high dimensionality and massive scale associated with social media data. Compressing, storing, processing, and visualizing data at this massive scale is the focus of chapter 12.

The book addresses technical and scientific challenges that emerge from the marriage of network theory and graph analytics using techniques from linear algebra, machine learning, big data analysis, and signal processing. It can serve as a great resource for undergraduate and graduate courses on social media analysis. It is also an excellent reference for those with research interests in this field.

Reviewer:  Raghvinder Sangwan Review #: CR145064 (1705-0250)
Bookmark and Share
  Reviewer Selected
Featured Reviewer
 
 
Data Mining (H.2.8 ... )
 
 
Web-Based Interaction (H.5.3 ... )
 
 
World Wide Web (WWW) (H.3.4 ... )
 
 
Group And Organization Interfaces (H.5.3 )
 
 
Systems And Software (H.3.4 )
 
Would you recommend this review?
yes
no
Other reviews under "Data Mining": Date
Feature selection and effective classifiers
Deogun J. (ed), Choubey S., Raghavan V. (ed), Sever H. (ed) Journal of the American Society for Information Science 49(5): 423-434, 1998. Type: Article
May 1 1999
Rule induction with extension matrices
Wu X. (ed) Journal of the American Society for Information Science 49(5): 435-454, 1998. Type: Article
Jul 1 1998
Predictive data mining
Weiss S., Indurkhya N., Morgan Kaufmann Publishers Inc., San Francisco, CA, 1998. Type: Book (9781558604032)
Feb 1 1999
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy