|
Social Media Analytics: the Value Proposition - Rohini Srihari
There has been a meteoric rise in the amount of content on the web
generated by ordinary users, particularly through mobile devices. This
includes social media sites such as Facebook, Twitter, and YouTube, as
well as blogs, discussion forums, and reader responses to articles on
traditional news sites. Such data can be mined for many purposes
including business-related competitive insight, e-commerce, as well as
citizen response to current issues. This talk will survey commercial
applications exploiting social media data, the business models
driving these, and vendors providing the solutions. Computational
techniques being used for extracting such information and assimilating
it into actionable intelligence will also be briefly discussed. The
talk will also touch on applications being pursued by the DoD and
intelligence community, where the value proposition (benefits, costs
and value) is different, but equally compelling.
|
|
A tempest: Or, On the flood of interest in sentiment analysis, opinion
mining, and the computational treatment of subjective language. - Lillian Lee
"What do other people think?" has always been an important
consideration to most of us when making decisions. Long before the
World Wide Web, we asked our friends who they were planning to vote
for and consulted Consumer Reports to decide which dishwasher to
buy. But the Internet has (among other things) made it possible to
learn about the opinions and experiences of those in the vast pool of
people that are neither our personal acquaintances nor well-known
professional critics --- that is, people we have never heard of. Enter
sentiment analysis, a flourishing research area devoted to the
computational treatment of subjective and opinion-oriented
language. Sample phenomena to contend with range from sarcasm in blog
postings to the interpretation of political speeches. This talk will
cover some of the motivations, challenges, and approaches in this
broad and exciting field.
|
|
Online Shopping and Social Media - Natalie S. Glance
Community generated content, or social media, has become
increasingly important over the past several years. Social media sites
such as blogs, twitter and online discussion boards have been
recognized as valuable sources of market intelligence for companies
wishing to keep abreast of their customers' attitudes expressed
online. There has been little focus, however, on providing a similar
service to shoppers themselves. In fact, shoppers perform research
prior to making a purchase and tap into many kinds of online
information; in particular they may seek out editorial or user reviews
of specific products, buying guides for categories of products or
informal conversational product discussion such as those found in
message boards.
In this talk, I will discuss our recent results in aiding consumers
with their shopping research by providing access to community
generated content, focusing on reviews and online forums. Reviews of
products and merchants can have a large impact on how well a product
sells. Given a set of reviews of products or merchants from a wide
range of authors and several reviews websites, how can we measure the
true quality of the product or merchant? Likewise, discussion forums,
are an especially good place to find product comparisons within a
category of items, to find expert opinions, and to find first-hand
product experiences. I'll present a solution for pulling online forum
results from the web into the user interaction flow of the shopping
site.
|
|
Identifying Peer Influence in Massive Social Networks - Sinan Aral
The talk will report on and discuss 3 papers:
Distinguishing Influence Based Contagion from Homophily Driven Diffusion in Dynamic Networks
(Published in the Proceedings of the National Academy of Sciences, 2009, vol. 106, no.51.)
Node characteristics and behaviors are often correlated with the structure of social networks over time. While evidence of this type of assortative mixing and temporal clustering of behaviors amongst linked nodes is used to support claims of peer influence and social contagion in networks, homophily may also explain such evidence. Here we develop a dynamic matched sample estimation framework to distinguish influence and homophily effects in dynamic networks, and apply this framework to a global instant messaging network of 27.4 million users, using data on the day-by-day adoption of a mobile service application and users' longitudinal behavioral, demographic and geographic data. We find that previous methods overestimate peer influence in product adoption decisions in this network by up to 700% and that homophily explains over 50% of the perceived behavioral contagion. These findings and methods are essential to both our understanding the mechanisms that drive contagions in networks and our knowledge of how to propagate or combat them in domains as diverse as epidemiology, marketing, development economics and public health.
Creating Social Contagion through Viral Product Design: A Randomized Trial of Peer Influence in Networks
We examine how firms can create word of mouth peer influence and social contagion by incorporating viral features into the design of their products. Evaluating the effects of such product design decisions on social contagion is difficult because econometric identification of peer influence is non-trivial. Although several approaches have been proposed, it is widely believed that the most effective way to obtain unbiased estimates of peer effects is to conduct large-scale randomized trials of peer-to-peer communications intended to influence particular economic decisions, such as the decision to adopt a product. We therefore designed and conducted a randomized field experiment testing the effectiveness of passive-broadcast and active-personalized viral messaging capabilities in creating peer influence and social contagion among the 1.4 million friends of 9,687 experimental users of Facebook.com. The experiment utilizes a customized commercial Facebook application to observe user behavior, communications traffic and the peer influence effects of randomly enabled viral messaging capabilities on application diffusion in the local networks of experimental and control population users. Results show that viral product design features generate econometrically identifiable peer influence and social contagion effects. Features that require more activity on the part of the user and are more personalized to recipients create greater marginal increases in the likelihood of adoption per message, but generate fewer total messages creating countervailing effects on peer influence. On average, passive-broadcast viral messaging capabilities, which are less personalized but also require less user effort, generate a 246% increase in local peer influence and contagion effects over a baseline model in which viral messaging is disabled. Adding active-personalized viral messaging capabilities, which are more personalized but require more user effort, generates an additional 98% increase in local peer influence and contagion effects over the passive-broadcast model. Analysis shows that initial peer adoptions in users' local networks drive a viral feedback loop that accelerates contagion. These results shed light on how viral products can be designed to generate social contagion and how randomized trials can be used to identify peer influence effects in social networks.
Identifying Social Influence: A Comment on Opinion Leadership and Social
Contagion in New Product Diffusion (Forthcoming in Marketing Science)
I sketch five broad questions that could, if appropriately addressed, dramatically improve how we conceptualize and manage social contagions in a variety of domains: 1) What exactly is (causal) social influence? 2) How do product characteristics affect peer influence and contagion? 3) What is the role of sustained product use in creating sustainable contagions? 4) How do the distributions of individual characteristics over network nodes affect contagion? 5) Are there 'systems' of complementary contagion management strategies?
|