CHAPTER 2 BEYOND SENTIMENT: HOW SOCIAL NETWORK ANALYTICS CAN ENHANCE OPINION MINING AND SENTIMENT ANALYSIS F. Pallavicini a , P. Cipresso b , F. Mantovani a University of Milano-Bicocca, Milan, Italy a Applied Technology for Neuro-Psychology Lab, IRCCS Istituto Auxologico Italiano, Milan, Italy b 1 INTRODUCTION The exponential growth in the use of digital devices, together with ubiquitous online access, provides unprecedented ground for the constant connectivity of people and offers tremendous capabilities for publicly expressing opinions, attitudes, or reactions regarding many aspects of everyday human activities [1]. Social media, such as blogs, forums, and social network platforms (eg, Facebook, LinkedIn, Twitter, Instagram, YouTube) are quickly becoming an integral part of people’s lives, the virtual spaces where daily individuals share opinions and information and maintain and/or expand their relational network. The massive use of online social networks and the abundance of data collected through them has raised exponentially the attention of the scientific and business community toward them [2–4]. Nowadays, the constant refinement of analytical tools is offering a richer array of opportunities to analyze these data for many different purposes [5]. Differences in features and characteristics of online social networks are reflected in the huge amount of different statistics and metrics that it is possible to track and analyze. The most adopted metrics are numeric, relatively easy to obtain, and freely available, such as engagement and influence metrics [6]. However, metrics of this types are often defined as “vanity metrics,” since they do not interpret or contextualize the data collected. 1 , 2 For this reason, other types of methods of analysis has been introduced. Among them, one of the most used is sentiment analysis (SA) [7], which is the analysis of the feelings (ie, opinions, emotions and attitudes) behind the words using natural language processing tools. SA is considered a quality metric, which looks behind numbers to understand how information about emotion and attitudes is conveyed in language [7]. Given the rising interest in the application of SA to data from online social networks, the research in this area has acknowledged the limitations 1 http://www.socialmediatoday.com/social-business/2015-04-09/social-vanity-metrics-top-4-worst-offenders 2 https://www.socialmediaexplorer.com/social-media-measurement/in-praise-of-vanity-metrics/ Sentiment Analysis in Social Networks. http://dx.doi.org/10.1016/B978-0-12-804412-4.00002-4 Copyright © 2017 Elsevier Inc. All rights reserved. 13