Published on March 5, 2014
Type of Networks based on Size WWW Networks, Telecom Networks Email Networks Meso Organisational Networks, Chemical Networks, Sports Networks Criminal Networks, Terrorist Networks Economic Networks, BoD Networks, Lobby Networks, Citation, Patents Micro Friendship Networks Team Networks Migration Networks Refugee Networks Macro 2 Social Networks
Network Influencer Identification 3
Brand associations Finding: Based on analysis of Social media chatter around Olympics, we found that Gymnastics is the most popular sport, followed by cycling, basketball, swimming, equestrian, rowing, archery and judo. Evidence: • A sample of 43,048 tweets across 27,683 users were collected between July 29th and July 31st. • Concentration across the US, Europe and Australia. Enterprise Implications: Popularity of games & sports such as Gymnastics, Swimming, Equestrian, Rowing, Archery and Judo is an opportunity for corporates to associate their brands with them. 4 4
Contagion effect in Mutual Funds operating in India Finding: Institutional investors investing in Indian market have large exposures to a few blue-chip stocks. Evidence: • Normally one would associate this behavior at the level of individual investors. • The institutional investors also depict a low risk investment behavior as exhibited in their large exposure to blue-chips stocks. Implications: • This suggest the presence of Bandwagon effect, Peer influence, Conformity and Contagion in the investment behavior of institutional investors. 5 5
Board of Directors: Top 100 most valuable companies in India by market capitalization 46 clusters. The figure shows the clusters and their interconnections through board memberships. The links show board membership ties between the members. This could be applied to, for example, predicting voting patterns 6
State of the Union messages #0: #1: #2: #3: 7 Most influential contexts: american people nation future congress good million agreement year america enemy state iraq terrorist iraqi force #0: #1: #2: #3: Most influential contexts: american people bank country year economy crisis energy america time make large job million create industry
Social Media & The Tale of Two Storms: Hurricane Sandy & Cyclone Nilam 8
Overview • • Social media destinations such as Twitter, Flickr and Facebook were leveraged extensively to spread information during the Victorian bushfires in Australia in 2009. • 9 Social Media sources have played a key role in disaster reporting, relief and rescue efforts. We try to explore the role played by government, civic authorities, law and order, general public, celebrities, activists, journalists and media in the advent of two natural disaster situations: Hurricane Sandy and Cyclone Nilam.
Sandy & Nilam • Hurricane Sandy was the largest Atlantic hurricane on record, as well as the second costliest Atlantic hurricane only surpassed by Hurricane Katrina in 2005. – – – • 10 Hurricane Sandy struck in late October 2012. In the United States, Sandy caused severe damage in New Jersey and New York. It claimed more than 50 lives, left millions without power and caused over US$ 50 billion in damage in the United States. Damages to life and property are spread across Jamaica, Haiti, Dominican Republic, Puerto Rico, Cuba and The Bahamas. Cyclone Storm Nilam, which struck India in late October 2012 caused damages across Indian states such as Tamil Nadu, Andhra Pradesh, Karnataka and Odisha.
The Data • According to Semiocast, the United States has 141.8 million Twitter users, and India has over 15 million Twitter users • We collected tweets containing the hashtag #Sandy from Oct 30th until Nov 6th (500,036 tweets from 306,348 users) – – – – – • 11 We classified #Sandy tweets into four different sets based on the time interval when the tweets occurred Time Interval 1 has 180,489 Nodes and 232,578 Edges Time Interval 2 has 60,351 Nodes and 65,842 Edges Time Interval 3 has 31,033 Nodes and 31,039 Edges Time Interval 4 has 14,550 Nodes and 13,350 Edges We also collected about 1500 tweets with the hashtag #Nilam during 31st Oct 2012.
Hurricane Sandy: Time Interval 1 12
Hurricane Sandy: Time Interval 2 13
Hurricane Sandy: Time Interval 3 & 4 14
Hurricane Sandy 15
Centrality Measures Betweenness Centrality Degree Time 1 16 Time 2 Time 3 Time 4 Time 1 azipaybarah Occuweather NYGovCuomo buell003 Newyorkist nycarecs DanielSquadron NYDNHammond NewYorkPost SallyGold adamlisberg MTAInsider fema nowthisnews AntDeRosa sahnetaeter nydailynews YourAnonNews OccupyWallSt Ows_Casper chipsandpickles PSteely Shabbyginger EFF ShaneFloyd NYCFUNRIDE ginjula briandonohue FreeUnivNYC sanjayguptaCNN NYeHealth GOOD 923NowFM Agent350 sloane csanati PhunkyLondon SenatorMenendez NYHealthScape TimGuinee diannank mzelma MAMIVERSE rootlessme FlourMagic proudamericans julienegrin SOYLAMAR sunshineejc dianalimongi MYERTECH CocktailBodega LarisaLive Romniac GoodwillRescue Laura_Byrnes 32BJ_SEIU ElisaBatista peterfhart crampell janedoenut stevenacook DyckmanBar AsBestRecipes urbanjibaro MarianaByDesign JeffreyKyle JohnGaltReport MoScarlet DorDor29 tonytorero krys730 tinackp sookietex Sharpie_Says PalanteLatino metnyc lafamiliacool GinnyMackles unitedcommon FrancisBoulle MINDBLOWlNG EmWatson NYGovCuomo realDonaldTrump MindbIowingFact NewYorkPost nowthisnews kurtdietrich katespencer YourAnonNews ArianaGrande bananarams CrazyHoneyBooBo RealMindBlowing NathanTheWanted JimGaffigan iamdiddy AnthonyShaw_ TheRealNickMara Time 2 NYGovCuomo FrancisBoulle whateverson sanjayguptaCNN AC360 MTAInsider MINDBLOWlNG RealMindBlowing morrowchris ReaderMom71 ERogTweets2Much 9GAG NathanTheWanted realDonaldTrump MindbIowingFact EmWatson BarackObama alexanderludwig EndHateRadio NewYorker Time 3 Time 4 NYGovCuomo Alyssa_Milano CarrieFairygirl essamz chrisrockozfan ConEdison MINDBLOWlNG anamariecox FrancisBoulle SteveNiles HuffingtonPost BW EmWatson AKRPR BarackObama RedCross MikeBloomberg fema GovChristie Gizmodo TheRealXtina NYGovCuomo YourAnonNews RBPundit fema AKRPR halfadams RedCross AntDeRosa EmWatson sophieraworth realjeffreyross Gothamist GovChristie andersoncooper BarackObama WNYC Janedoenut alexa_chung Chrisrockozfan
Conclusions • Sandy was 'Instagrammed': Number of #Sandy tweets originating from Instagram ranked fourth, preceded only by Tweets originating from iPhone, Web and Android and in the case of Nilam, Twitpic scored high. – – In disaster scenarios, visual means of message propagation assume prominence. This is in contrast to our observations with regard to message propagation during the Olympic Games, where Tweets originating from Instagram scored very low • • News Media leveraging Social Media: Journalists and News Media extensively leveraged Social Media for news propagation and for news amplification. News Media sources and Journalists were the heaviest users of Social media in the events of both Sandy and Nilam. • 17 Government and officials leveraging Social media: @NYGovCuomo, which is the Official Twitter account for the Governor of New York State, Mr. Andrew Cuomo consistently emerged as one of the most important entities. No such trends in the case of Cyclone Sandy where we did not see even a single tweet from a government official or utility. Individual Journalists leveraged their Twitter follower base to spread messages faster and they invariably figured higher up in the network centrality measures as against the twitter accounts of Media houses which employ them.
Conclusions • Cyclone Nilam saw a low overall coverage in Social Media in India and surrounding affected regions. • Celebrities in the United States were active in social media during Hurricane Sandy as against celebrities in India who chose to ignore Cyclone Nilam. • Measures of centrality, can show who creates the most connections in the network. – • Lists of words frequently used in the network of those who mention #sandy & #nilam can be clustered by their closest affiliations and most commonly used words – • 18 In our case, the Twitter handles of NYC Guv and several leading journalists play this role This can help trace communities within the network formed by the recurring use of particular words Using the URLs used in Tweets from the entire network as well as from each of the smaller groups within the network, we can trace a pattern of interests, values, and engagement that drew the most attention across the network.
Thank you jai3.ganesh3@Cognizant.com 19
Die soziale Netzwerkanalyse ist eine Methode der empirischen Sozialforschung zur Erfassung und Analyse sozialer Beziehungen und sozialer Netzwerke.
Network Analysis for Wikipedia F. Bellomi and R. Bonato Dipartimento di Informatica, Universita` di Verona C`a Vignal, 2 Strada Le Grazie, 15 I-37134 ...
INSNA INSNA is the professional association for researchers interested in social network analysis. INSNA Past President Honored. The University of South ...
Social network analysis [SNA] is the mapping and measuring of relationships and flows between people, groups, organizations, computers, URLs, and other ...
Network analysis (or social network analysis) is a set of mathematical methods used in social psychology, sociology, ethology, and anthropology.
10 Free Network Analysis Tools. Network management can be a tall task without the right statistics and data about traffic flows, device configurations and ...
Colasoft provides free network analyzer, network sniffer and retrospective network analysis solution for network monitoring, network troubleshooting ...
Generally speaking, network analysis is any structured technique used to mathematically analyze a circuit (a “network” of interconnected components).
Capsa Free network analyzer (packet sniffer) software for Windows platform, sniffer packets and analyze protocols, best FREE Network Analysis tool for ...
What is Social Network Analysis? Social network analysis is focused on uncovering the patterning of people's interaction. It is about the kind of ...