arXiv subjects

Fabrício Benevenuto

Publications and source records attributed to Fabrício Benevenuto.

27 records · Page 2Linked to original sources

arXiv2017

The Emergence of Crowdsourcing among Pokémon Go Players

Since its launching, Pok{é}mon Go has been pointed as the largest gaming phenomenon of the smartphone age. As the game requires the user to walk in the real world to see and capture Pok{é}mons, a new wave of crowdsourcing apps have emerged to allow users to collaborate with each other, sharing where and when Pok{é}mons were found. In this paper we characterize one of such initiatives, called PokeCrew. Our analyses uncover a set of aspects of user behavior and system usage in such emerging crowdsourcing task, helping unveil some problems and benefits. We hope our effort can inspire the design of new crowdsourcing systems.

cs.SI↗

arXiv2016

SentiBench - a benchmark comparison of state-of-the-practice sentiment analysis methods

In the last few years thousands of scientific papers have investigated sentiment analysis, several startups that measure opinions on real data have emerged and a number of innovative products related to this theme have been developed. There are multiple methods for measuring sentiments, including lexical-based and supervised machine learning methods. Despite the vast interest on the theme and wide popularity of some methods, it is unclear which one is better for identifying the polarity (i.e., positive or negative) of a message. Accordingly, there is a strong need to conduct a thorough apple-to-apple comparison of sentiment analysis methods, \textit{as they are used in practice}, across multiple datasets originated from different data sources. Such a comparison is key for understanding the potential limitations, advantages, and disadvantages of popular methods. This article aims at filling this gap by presenting a benchmark comparison of twenty-four popular sentiment analysis methods (which we call the state-of-the-practice methods). Our evaluation is based on a benchmark of eighteen labeled datasets, covering messages posted on social networks, movie and product reviews, as well as opinions and comments in news articles. Our results highlight the extent to which the prediction performance of these methods varies considerably across datasets. Aiming at boosting the development of this research area, we open the methods' codes and datasets used in this article, deploying them in a benchmark system, which provides an open API for accessing and comparing sentence-level sentiment analysis methods.

cs.CL↗

arXiv2016

Fusing Audio, Textual and Visual Features for Sentiment Analysis of News Videos

This paper presents a novel approach to perform sentiment analysis of news videos, based on the fusion of audio, textual and visual clues extracted from their contents. The proposed approach aims at contributing to the semiodiscoursive study regarding the construction of the ethos (identity) of this media universe, which has become a central part of the modern-day lives of millions of people. To achieve this goal, we apply state-of-the-art computational methods for (1) automatic emotion recognition from facial expressions, (2) extraction of modulations in the participants' speeches and (3) sentiment analysis from the closed caption associated to the videos of interest. More specifically, we compute features, such as, visual intensities of recognized emotions, field sizes of participants, voicing probability, sound loudness, speech fundamental frequencies and the sentiment scores (polarities) from text sentences in the closed caption. Experimental results with a dataset containing 520 annotated news videos from three Brazilian and one American popular TV newscasts show that our approach achieves an accuracy of up to 84% in the sentiments (tension levels) classification task, thus demonstrating its high potential to be used by media analysts in several applications, especially, in the journalistic domain.

cs.CL↗

arXiv2016

TrendLearner: Early Prediction of Popularity Trends of User Generated Content

We here focus on the problem of predicting the popularity trend of user generated content (UGC) as early as possible. Taking YouTube videos as case study, we propose a novel two-step learning approach that: (1) extracts popularity trends from previously uploaded objects, and (2) predicts trends for new content. Unlike previous work, our solution explicitly addresses the inherent tradeoff between prediction accuracy and remaining interest in the content after prediction, solving it on a per-object basis. Our experimental results show great improvements of our solution over alternatives, and its applicability to improve the accuracy of state-of-the-art popularity prediction methods.

cs.SI↗

arXiv2015

The H-index Paradox: Your Coauthors Have a Higher H-index than You Do

One interesting phenomenon that emerges from the typical structure of social networks is the friendship paradox. It states that your friends have on average more friends than you do. Recent efforts have explored variations of it, with numerous implications for the dynamics of social networks. However, the friendship paradox and its variations consider only the topological structure of the networks and neglect many other characteristics that are correlated with node degree. In this article, we take the case of scientific collaborations to investigate whether a similar paradox also arises in terms of a researcher's scientific productivity as measured by her H-index. The H-index is a widely used metric in academia to capture both the quality and the quantity of a researcher's scientific output. It is likely that a researcher may use her coauthors' H-indexes as a way to infer whether her own H-index is adequate in her research area. Nevertheless, in this article, we show that the average H-index of a researcher's coauthors is usually higher than her own H-index. We present empirical evidence of this paradox and discuss some of its potential consequences.

cs.SI↗

arXiv2014

On the Dynamics of Social Media Popularity: A YouTube Case Study

Understanding the factors that impact the popularity dynamics of social media can drive the design of effective information services, besides providing valuable insights to content generators and online advertisers. Taking YouTube as case study, we analyze how video popularity evolves since upload, extracting popularity trends that characterize groups of videos. We also analyze the referrers that lead users to videos, correlating them, features of the video and early popularity measures with the popularity trend and total observed popularity the video will experience. Our findings provide fundamental knowledge about popularity dynamics and its implications for services such as advertising and search.

cs.SI↗

arXiv2014

Comparing and Combining Sentiment Analysis Methods

Several messages express opinions about events, products, and services, political views or even their author's emotional state and mood. Sentiment analysis has been used in several applications including analysis of the repercussions of events in social networks, analysis of opinions about products and services, and simply to better understand aspects of social communication in Online Social Networks (OSNs). There are multiple methods for measuring sentiments, including lexical-based approaches and supervised machine learning methods. Despite the wide use and popularity of some methods, it is unclear which method is better for identifying the polarity (i.e., positive or negative) of a message as the current literature does not provide a method of comparison among existing methods. Such a comparison is crucial for understanding the potential limitations, advantages, and disadvantages of popular methods in analyzing the content of OSNs messages. Our study aims at filling this gap by presenting comparisons of eight popular sentiment analysis methods in terms of coverage (i.e., the fraction of messages whose sentiment is identified) and agreement (i.e., the fraction of identified sentiments that are in tune with ground truth). We develop a new method that combines existing approaches, providing the best coverage results and competitive agreement. We also present a free Web service called iFeel, which provides an open API for accessing and comparing results across different sentiment methods for a given text.

cs.CL↗

arXiv2014

Reverse Engineering Socialbot Infiltration Strategies in Twitter

Data extracted from social networks like Twitter are increasingly being used to build applications and services that mine and summarize public reactions to events, such as traffic monitoring platforms, identification of epidemic outbreaks, and public perception about people and brands. However, such services are vulnerable to attacks from socialbots $-$ automated accounts that mimic real users $-$ seeking to tamper statistics by posting messages generated automatically and interacting with legitimate users. Potentially, if created in large scale, socialbots could be used to bias or even invalidate many existing services, by infiltrating the social networks and acquiring trust of other users with time. This study aims at understanding infiltration strategies of socialbots in the Twitter microblogging platform. To this end, we create 120 socialbot accounts with different characteristics and strategies (e.g., gender specified in the profile, how active they are, the method used to generate their tweets, and the group of users they interact with), and investigate the extent to which these bots are able to infiltrate the Twitter social network. Our results show that even socialbots employing simple automated mechanisms are able to successfully infiltrate the network. Additionally, using a $2^k$ factorial design, we quantify infiltration effectiveness of different bot strategies. Our analysis unveils findings that are key for the design of detection and counter measurements approaches.

cs.SI↗

arXiv2013

PANAS-t: A Pychometric Scale for Measuring Sentiments on Twitter

Online social networks have become a major communication platform, where people share their thoughts and opinions about any topic real-time. The short text updates people post in these network contain emotions and moods, which when measured collectively can unveil the public mood at population level and have exciting implications for businesses, governments, and societies. Therefore, there is an urgent need for developing solid methods for accurately measuring moods from large-scale social media data. In this paper, we propose PANAS-t, which measures sentiments from short text updates in Twitter based on a well-established psychometric scale, PANAS (Positive and Negative Affect Schedule). We test the efficacy of PANAS-t over 10 real notable events drawn from 1.8 billion tweets and demonstrate that it can efficiently capture the expected sentiments of a wide variety of issues spanning tragedies, technology releases, political debates, and healthcare.

cs.SI↗