The Rise And The Fall Of A Citizen Reporter

2y ago
27 Views
2 Downloads
3.17 MB
11 Pages
Last View : 1m ago
Last Download : 2m ago
Upload by : Camryn Boren
Transcription

Wellesley CollegeWellesley College Digital Scholarship and ArchiveComputer Science Faculty ScholarshipComputer Science5-2013The Rise and the Fall of a Citizen ReporterP. Takis MetaxasWellesley College, pmetaxas@wellesley.eduEni MustafarajWellesley College, emustafa@wellesley.eduFollow this and additional works at: ultyRecommended CitationMetaxas, Panagiotis and Mustafaraj, Eni (2013). "The Rise and the Fall of a Citizen Reporter." To be presented at WebSci13, May 2-4,2013, Paris, France.This Conference Proceeding is brought to you for free and open access by the Computer Science at Wellesley College Digital Scholarship and Archive.It has been accepted for inclusion in Computer Science Faculty Scholarship by an authorized administrator of Wellesley College Digital Scholarshipand Archive. For more information, please contact ir@wellesley.edu.

The Rise and the Fall of a Citizen ReporterPanagiotis MetaxasWellesley CollegeComputer Science Departmentpmetaxas@wellesley.eduABSTRACTRecently, research interest has been growing in thedevelopment of online communities sharing news andinformation curated by “citizen reporters”. Using “BigData” techniques researchers try to discover influencegroups and major events in the lives of such communities.However, the big picture may sometimes miss importantstories that are essential to the development and evolutionof online communities. In particular, how does one identifyand verify events when the important actors are operatinganonymously and without sufficient news coverage, as indrug war-torn Mexico? In this paper, we present sometechniques that allow us to make sense of the data collected,identify important dates of significant events therein, anddirect our limited resources to discover hidden stories that,in our case, affect the lives and safety of prominent citizenreporters. In particular, we describe how focused analysisenabled us to discover an important story in the life of thiscommunity involving the reputation of an anonymousleader, and how trust was built in order to verify the validityof that story.Author Keywordswebscience; social computing; social media; citizenreporters; civic media; crisis informatics; crowdsourcing;news; drug war; microblogging; narcotweets; Twitter;Mexico.ACM Classification KeywordsK.4.3 [Computers and Society]: Organizational Impacts–Computer-supported cooperative work. H.5.3 Groups &Organization Interfaces—collaborative computing,computer-supported cooperative work; K.4.2 Social IssuesINTRODUCTIONThe rise of the Social Web has created a new informationsource: citizen reporters. Social media and networkingplatforms, including Twitter and Facebook, allow everyonein the world to report what is happening in theirAuthor’s Final Copy.To be presented at theWebSci’13, May 2–4, 2013, Paris, France.Eni MustafarajWellesley CollegeComputer Science Departmentemustafa@wellesley.eduneighborhood or the city they live in, in real-time.Platforms specializing in organizing humanitarian responseto disasters, such as Ushahidi, rely on people on the groundto report on situations that need immediate attention [14].Anyone can be a reporter, but how do we assess thecredibility of citizen reporting? When we read news, weusually choose our information sources based on thereputation of the media organization. Even though in thepast there have been breaches of such trust, and all mediaorganizations have an embedded bias that affects what theychoose to report [3], we generally trust the newsorganizations and expect that their reporting is credible.But what about the credibility of citizen reporting? Unlikeestablished news organizations, it lacks the inherentstructures that help us evaluate credibility. However,sometimes citizen reporting might be the only source ofinformation we might have. How can we use technology tohelp us verify the credibility of such reports, especiallywhen our safety may depend on the information we receivefrom citizen reporters?In the last few years, Mexico has seen a significant amountof violence related to the struggle for control by powerfuland ruthless drug cartels [16]. The cities in the pathsbetween drug-producing Central and South America anddrug-consuming USA have seen an increase in cartelorchestrated violence and gruesome scenes aiming tointimidate rival groups. Reports bring the number of drugrelated deaths to over 60,000 people [23], [25].The widely held opinion that different branches of thepolice force were infiltrated and corrupted by cartels forcedthe government of former President Calderon to bringArmy and Marine forces into cities and towns [26]. Whilethe armed forces are trusted by the citizens and have hadconsiderable success in their efforts, the cartels haveincreased their terrorizing activities against randombystanders, tourists and even whole villages. In particular,they have attacked journalists who have tried to report onthe situation. As a result, in many cities there is a selfimposed silence of the press due to the terror [10].Citizens, however, have found ways to inform themselvesof the dangers they encounter daily by using social media.Twitter, in particular, has emerged as the tool of choice forspreading information that can help citizens plan their day.Twitter users report, confirm, comment on, and disseminateinformation and alerts about the violence, typically as it

unfolds. Many Mexicans reportedly consult Twitter everymorning the same way that they consult the weatherinformation in order to plan their day.In [21] we describe our initial efforts to collect and analyzedata produced by the community around the #MTYfollowtag in Monterrey, Mexico (MTY is its airport code).Complementary work is presented in [20] with an overviewanalysis of a large data set in several Mexican citiesincluding Monterrey. Both of these papers maintain theanonymity of citizen reporters by introducing pseudonymsfor the major actors in the data to avoid any risk that couldresult by revealing their true identities.Anyone can be a reporter. However, this poses a newproblem: how do we assess the credibility of citizenreporting? [17] When we read news, we usually choose ourinformation sources based on the reputation of the mediaorganization: BBC, New York Times, Der Spiegel, etc. Wetrust these institutions and we expect that their reporting iscredible. Citizen reporting lacks the inherent structures thathelp us evaluate credibility as we do with traditional mediareporting, but sometimes citizen reporters might be the onlysource of information we have. How can we use technologyto help us verify the credibility of such reports?The so-called “Big Data” made possible by ease ofcollecting digital data, including a massive collection ofSocial Web data, holds a lot of promise for studying thepatterns through which Society is operating. However, italso raises concerns about the verification of these observedpatterns [7]. In this paper, we demonstrate how superficialanalysis at the “Big Data” level can miss important storieshiding under the numbers.In this paper, we present some techniques that allow us tomake sense of the data collected, identify important dates ofsignificant events in them, and direct our limited resources todiscover hidden stories that affect the lives and safety ofprominent citizen reporters. In particular, we describe howfocused analysis enabled us to discover an important story inthe life of this community involving the reputation of ananonymous leader in the community. We start by describingthe data collection process and the results of the initialautomatic analysis, along with its limitations. Next we describethe methods we employed to identify and verify the events wediscovered along with interviews of some of the importantcitizen reporters in this community. In the last section, weoutline the characteristics of a semi-automatic system that canenhance the ability of its users to evaluate credibility in realtime sources.Anonymous Citizen ReportersDATA COLLECTIONIn general, it is not trivial to validate claims made byanonymous actors. For reasons that will become apparent,however, we are dropping some of the anonymity in thispaper when we were able to verify, to our satisfaction, boththe validity of several claims and the identity of theimportant actors.This paper is about the emergence of a network ofanonymous users in Monterrey, a large city in NorthernMexico, using (mostly) pseudonymous accounts to informand be informed about events in their city that could exposethem to danger. While they also share reports about trafficaccidents, the most serious danger they encounter comesfrom the fight between ruthless drug cartels that competefor control over land routes inside Mexico. Describedsometimes as “war”, this conflict for control between thecartels and with the Mexican authorities (local, state, andfederal police, the Army, the Navy, etc.) has resulted inthousands of casualties as well as widespread fear anduncertainty in the Mexican population [12], [16].We usually rely on our journalists and news organizationsto keep us informed on the dynamics of events around theglobe, but not every country has a free press or is willing orable to allow the international press to move freely. In somecountries, like Mexico, journalists have been killed byorganized crime or put under pressure by the authorities tostop reporting on certain events [10].The rise of the Social Web has created a new outlet forstaying informed: citizen reporting. The different socialmedia and networking platforms, like YouTube, Flickr,Twitter, and Facebook allow everyone in the world toreport in real-time what is happening in the place they live.The initial data for this study, provided by [21], was basedon a set of keywords related to Mexico events submitted tothe archival service Archivist [1]. The initial datasetanalyzed in this paper was based on the selection of thehashtag #MTYfollow (tag capitalization is not important)and consists of 258,734 tweets written by 29,671 uniqueTwitter accounts, covering 286 days in the time intervalNovember 2010 - August 2011 (see Figure 1).Due to the opportunistic character of the initial datacollection, we knew that the hashtag defining thiscommunity was created earlier than the dates of thecollected data. We supplemented the initial dataset inseveral occasions, based on the discoveries we made duringthe analysis.First, we performed a series of additional data collection inSeptember 2011 to identify the origin of the hashtag and thefollow relationships of the members of the community. Inparticular, we collected all social relations for the users inthe current dataset that were still active in Sept. 2011, aswell as their account information. We collected all tweetsfor accounts created since 2009 with less than 3200 tweets,in order to discover the history of the hashtag #MTYfollowthat defines the community we are studying. We also madeuse of the dataset described in [24] to locate tweets archivedin 2009.

Figure 1. Histogram of the number of tweets collectedautomatically with hashtag #MTYfollow between Nov.2010 and Aug. 2011.For reasons that will be explained in a later section, in June2012 we performed a third sequence of targeted datacollections, seeking accounts that were created in lateMarch, 2011 and had names containing the strings “aguila”and “halcon” (Spanish for “eagle” and “hawk”) as well astweets containing hashtag #aguilasMTY. We collected 654such accounts and to a further qualified subset of 359accounts we collected all their social information and theirtweets.INITIAL DATA ANALYSISIn the initial analysis, we tried to get an overall picture ofthe data [21]. As is often the case with Social Web data, wenoticed that some accounts were tweeting much morefrequently than others. We also noticed that the rate oftweets increased when violence was reported in the city (asindicated by the use of keywords for “shooting”,“explosion”, etc). In some, but not all, instances we wereable to match dates with high rates of tweets to publisheddates of violence in Monterrey, such as the accidental deathof two TEC students who were caught in the crossfirebetween the authorities and the drug cartels on March 22,2011 [27].Next, we collected the followers of the accounts thatdisplayed high activity in our data set and we tried to seehow they were connected through mutual followrelationship. This analysis resulted in the Gephi graph [2]shown in Figure 2, which is remarkable, and not only for its(accidental) visual resemblance to a human brain. Weselected for visualization those accounts that had at least 75mutual followers in the data set, a semi-arbitrary numberindicating higher involvement with the group and interest ineach other’s postings. The cutoff of 75 was selected forcomputational reasons and because it resulted in a singlestrongly connected component of the graph.Figure 2. Visualization (in Gephi) of accounts with atleast 75 mutual followers of accounts using #MTYfollowin our initial data collection. Three parameters areshown: node size (a circle) indicates size of followersgroup; node proximity (Euclidian distance betweencircles) indicates cardinality of common followers; andcolor (for nodes and edges) indicates subgroups withrelatively more followers than others. Of particularinterest in this paper are the green and blue groups.Using Gephi’s functions, we drew the graph as follows:Using the Force Atlas 2 force-directed algorithm [11], wevisualized node proximity, and gave it its overall shape.Using the Louvain community detection module [5], wedetected subgroups that share more connections withinthemselves than with other groups, and gave it its colors.Five large subgroups emerged. Of particular interest in thispaper are the groups colored green and blue, in the lowerpart of the graph. Finally, using Gephi’s logarithmic scalenode size drawing method, we drew larger the nodes withmore followers. This reveals that the nodes with mostfollowers belong mainly to the green group with few onesin the blue and red group.There are a few accounts with many followers in thiscollection; the largest of all with 57,127 followers is@trackmty (which, in [21] is mentioned with thepseudonym @GodFather). Within the community definedby #MTYfollow that we study and visualize above, thisaccount has 9,079 followers, or 36% of all active membersat the time of the data collection.Then, for @trackmty and other prominent accounts in ourdata set (that is, accounts with many followers) we decidedto look at the histograms of their activity. In particular,looking at the volume of tweets that were sent by@trackmty and that mentioned it (see Figure 3) we see asurprising pattern unlike the pattern of any other prominentaccount: The volume of its tweets seems to be falling toabout a third of its average volume around April 17, 2011,and then drops to zero at the end of July, 2011. Moreover,the volume of tweets containing the string “@trackmty”,

that is, tweets that either mention the account or retweetedits contents verbatim, similarly drops by about half aroundthe same time period, though these tweets persist after@trackmty has stopped tweeting. And the two time periodsare separated with the highest spike in tweet activity onApril 17. This is rather remarkable, given the fact that thehistogram of all tweets in the community does not show anysuch pattern in the tweet rate over time (see Figure 1).thousands of tweets serially or at random, since we knewwhere to look. We decided to engage Spanish-speakingstudents (see Acknowledgement section) to help ustranslate a few dozen random tweets posted at around thetimes of the spikes. We also used the automatic Googletranslate feature of Chrome [13] to translate a few blogpostings.At this point we realized that we needed to examine theevents prior to April 17 more closely to make sense of theactivity changes for @trackmty.Before providing more information on what we discoveredthrough our research, and to make it easier for the reader tofollow the description, we will use this subsection tointroduce the different actors of the #MTYfollowcommunity.The actorsThe News Organizations and AuthoritiesReporters in Mexico have often been violently attacked forreporting on the war among drug cartels for control oftrafficking passages and cartel conflict with the Army,Marines, and local police. While there are several newsorganizations reporting on the situation, their reporting isnot without gaps. Investigative reporting has diminisheddue to brutal executions of journalists.Figure 3. Timeline of tweets related to @trackmty. Weobserve that this user is tweeting with higher averagevolume before April 17, 2011 (red rectangles). Thecommunity’s tweets mentioning it or tweeting itsmessages (blue rectangles) follows a similar pattern.Notice the spike of activity in mid-April, 2011.THE LIMITS OF “BIG DATA” ANALYSISUp until this point, we had examined the collected datausing basic statistical and visualization methods foranalyzing large volumes of data without having read any ofthe contents of the tweets. To employ more sophisticateddata mining methods did not seem a promising path foranalysis. The answer to the puzzle of activity patternchange we were seeking was in the content of the tweets.While our ability to understand the developments withoutreading specific tweets was diminishing, the questionspersisted: What happened shortly before April 17 thatresulted in the spike of mentions and reduced activity of@trackmty? What caused the other spikes in January,February, and March 2011? Were the highest peaks themost important in the community’s development? And whydid @trackmty stop tweeting in late July?These unexpected patterns in activity provide guidance onwhere to look for explanations, but for non-Spanishspeakers, they cannot help further. However, the initialanalysis was already very useful in guiding us to selectwhat to examine. We did not need to read hundred ofDistrust of the police further complicates the situation onthe ground in Monterrey. It is common belief there thatlocal, state and federal police have been infiltrated by thecartels, especially the powerful Golf cartel and the ruthlessZetas. Most people trust the Army and the Marines, buteven those institutions are immune to mistrust.The citizen reporters of #mtyfollowThe majority of the Twitter accounts represented in Figure2 are citizen reporters. These reporters mainly usepseudonyms out of fear of exposure. In a few occasions,bodies have been found with notes stating that they wereexecuted because they were “talking on the internet” [8].The term “curators” coined by [20] indicates the moreprominent citizen reporters on Twitter. [20] has moreinformation about them, together with some interviews.Curators reportedly spend long hours in altruistic efforts toinform the public of potential dangers in the city, yet theydo not often collaborate, as they see other curators ascompetitors who may “steal their tweets”.The citizen reporter @trackmtyAn important actor in the events we are describing is thecitizen reporter with account name @trackmty, who hasbeen tweeting since early 2010 with the pseudonym“Melissa Lotzer”. She claims to be the originator of the#mtyfollow community, something disputed by somemembers of the community. She is definitely the mostactive member of this community, and her 3 twitteraccounts (@MelissaLotzer, @monitoreoMNR being theother two) have often come under attack by trolls. She isalso the creator of a series of social media accounts namedNew Revolution of Mexico (“Mexico Nueva Revolucion”,or MNR [22]), present on social media platforms thatinclude Facebook, Wordpress, and Twitter. In early 2010,

she organized the MNR twitter group, and in late March2011 the eagles (“aguilas”) group.The BlogsThe vacuum created by the lack of reporting by officialnews organizations has been filled by several high-

"The Rise and the Fall of a Citizen Reporter." To be presented at WebSci13, May 2-4, 2013, Paris, France. The Rise and the Fall of a Citizen Reporter Panagiotis Metaxas Wellesley College Computer Science Department pmetaxas@wellesley.edu Eni Mustafaraj Wellesley College Computer Science Department emustafa@wellesley.edu

Related Documents:

Silat is a combative art of self-defense and survival rooted from Matay archipelago. It was traced at thé early of Langkasuka Kingdom (2nd century CE) till thé reign of Melaka (Malaysia) Sultanate era (13th century). Silat has now evolved to become part of social culture and tradition with thé appearance of a fine physical and spiritual .

May 02, 2018 · D. Program Evaluation ͟The organization has provided a description of the framework for how each program will be evaluated. The framework should include all the elements below: ͟The evaluation methods are cost-effective for the organization ͟Quantitative and qualitative data is being collected (at Basics tier, data collection must have begun)

̶The leading indicator of employee engagement is based on the quality of the relationship between employee and supervisor Empower your managers! ̶Help them understand the impact on the organization ̶Share important changes, plan options, tasks, and deadlines ̶Provide key messages and talking points ̶Prepare them to answer employee questions

Dr. Sunita Bharatwal** Dr. Pawan Garga*** Abstract Customer satisfaction is derived from thè functionalities and values, a product or Service can provide. The current study aims to segregate thè dimensions of ordine Service quality and gather insights on its impact on web shopping. The trends of purchases have

On an exceptional basis, Member States may request UNESCO to provide thé candidates with access to thé platform so they can complète thé form by themselves. Thèse requests must be addressed to esd rize unesco. or by 15 A ril 2021 UNESCO will provide thé nomineewith accessto thé platform via their émail address.

Chính Văn.- Còn đức Thế tôn thì tuệ giác cực kỳ trong sạch 8: hiện hành bất nhị 9, đạt đến vô tướng 10, đứng vào chỗ đứng của các đức Thế tôn 11, thể hiện tính bình đẳng của các Ngài, đến chỗ không còn chướng ngại 12, giáo pháp không thể khuynh đảo, tâm thức không bị cản trở, cái được

Food outlets which focused on food quality, Service quality, environment and price factors, are thè valuable factors for food outlets to increase thè satisfaction level of customers and it will create a positive impact through word ofmouth. Keyword : Customer satisfaction, food quality, Service quality, physical environment off ood outlets .

More than words-extreme You send me flying -amy winehouse Weather with you -crowded house Moving on and getting over- john mayer Something got me started . Uptown funk-bruno mars Here comes thé sun-the beatles The long And winding road .