Information monitoring, propagation analysis and community detection in Social Network Sites

[Italian version]
We are happy to announce that the SIGSNA team has just been awarded of some new research funds. The funds coming from the FIRB – future in research – funding action of the Italian Minister of Research and University are awarded only to young researcher under the age of 32 yo.
Our research project: “Information monitoring, propagation analysis and community detection in Social Network Sites ” brings the work we’ve done so far to a next level aiming at working within a multiple SNSs scenario.
The main objectives of this project are:
1) The development of monitoring tools and models, to filter relevant messages among millions of textual interactions and to identify distributed conversations made
of these messages.
2) The study of how information propagates in SNSs, following complementary approaches: the evaluation of mathematical propagation models on real data, the
identification of propagation patterns and the development of sociological models to explain these patterns.
3) The definition of new metrics to provide meaningful descriptions of large SNSs and their application to identify communities and other relevant structures hidden
inside the networks.
The project – as well as the SIGSNA prject – is based on a collaboration between the Department of Communication Studies of the University of Urbino and the Department of Computer Science of the University of Bologna.
The length of the FIRB research funds (3 years) and the large amount of funds (approx. 760K USD) will allow us to move deeper into the direction of high performance computing techniques applied to social network data. This is a research direction that we barely started and that seems to be quite promising.
[eng version]
Siamo estremamente felici di comunicare che il gruppo di ricerca SIGSNA ha vinto un importante finanziamento FIRB (futuro in ricerca) riservato ai progetti di ricerca presentati da ricercatori non strutturati con meno di 32 anni. Il progetto che abbiamo presentato si configura come un evoluzione di molte delle linee di ricerca che abbiamo portato avanti fino a questo momento. Al tempo stesso la lunghezza e l’importo del finanziamento (3 anni e ca 560mila € in totale) ci permettono di progettare un nuovo e più avanzato livello.
La sfida è infatti quella di analizzare un contesto estremamente complesso senza ridurlo ad un unico ambiente digitale ma affrontando l’intero ecosistema digitale. Ovviamente in quest’ottica alcuni dei lavori che abbiamo sviluppato fin ad ora (come il ML-model) insieme all’utilizzo di tecnologie HPC ci saranno di grande aiuto.

PRIN: Online social relations and identity: Italian experience in Social Network Sites.

We are happy to announce that the SIGSNA team is now officially part of a larger research group, counting 5 Italian universities (University of Urbino, University of Bologna, Catholic University of Milano, University of Bergamo, Calabria University), that has recently been awarded of PRIN research funds for the project: Online social relations and identity: Italian experience in Social Network Sites.
PRIN (short for the Italian version of: Research project of national interest) funds are awarded to large research group working of high level projects. The Online Social Relations and Identity project aims at investigating how social relations are reshaped and constructed Online. The research will focus on Italian people and will use Social Network Sites (SNS) as selected online environment for observation. Research methodology will be based on an advanced interdisciplinary approach based both on quantitative computer-driven analysis and on qualitative sociological analysis. This is obviously where all the expertise we’ve collected within the SIGNSA group will be most valuable.
The main research questions of the project are:
– What kinds of communication practices support social relationships in SNSs?
– What is the role of media production, fruition and sharing in this context?
– What kinds of practices support the public self-narration of users’ identity in SNSS?
– How is the distinction between private and public space being reshaped in SNSs? How this distinction in strategically used by users?
– How emotional relations (such as friendships or family) are redefined in SNSs?
In addition to that, since the research will use a large amount of automatically extracted data form SNS, a methodological question will be risen:
– What is an effective way to combine computer-based analysis of large quantity of SNS data with in-depth sociological interviews?
While ad ad hoc web site of the project will be established soon, we’re keeping publishing any relevant update also here.

Multi-Layer Networks: data available

We have made the data of our latest work on multi-layer networks available in the Data section. The zip file contains two networks in cvs format, one between twitter users and the other between friendfeed users, and a file with account correspondences indicating pairs of accounts in twitter and friendfeed belonging to the same person. We will add new networks in a while so stay tuned!
And since we’re having a great time in Kaohsiung we are very happy to publish a picture of us with the great staff of the conference.

The ML-model for multi-layer social networks


In few weeks we’re going to Kaohsiung, Taiwan to present at the International Conference on Advances in Social Network Analysis and Mining (ASONAM 2011). We’re really excited about this conference (last year was great!) and we want to share the paper we’re presenting. It’s been a while since we’ve started working on the ML model (multi-layer model) as a better way to describe and measure our contemporary SNSs experience. With the variety and different strengths of different social networking online services we are using today, it is becoming increasingly common to have different online networks supporting different functions. For instance, facebook can be used to interact with friends and family, Twitter can be used to broadcast to the world and LinkedIn can be used with colleagues and business partners. Given this scenario the traditional SNA approach catches only one small part of the complex system of interconnected networks we’re part of. The ML model proposes a way to keep several networks together while performing SNA analysis.
The model is just a first step but we hope, as one of the reviewers said, “to have scratched the tip of a potentially big iceberg”.
Read the paper [PDF]

Conversation retrieval from Social Media

Next week Matteo is going to Dublin for the annual European Conference on Information Retrieval (ECIR). We are presenting a demo of our Conversation retrieval system for Social Media and Social Network Sites. While the addition of Social aspect to traditional online searches has been around for some times we are following a different approach. So far social search used what we can define as an ego-centric approach that means that informational objects around the web get somehow recommended by your online contacts.
We are doing something different. We are moving from online search aimed at retrieving information toward what we call a conversational search. This means that the object of our search is no longer a single information but a set of messages and users that can be described (and ranked) according to many social aspects.
Therefore some of the ranking criteria that can be used are:
Text relevance, User centrality (e.g., degree, page rank, audience), Message popularity (e.g., retweets, likes, sharing), Timeliness (i.e., distance from a given timestamp), Length (i.e., number of messages), Density (i.e., emotions and interest).
We’ve done some blind comparison between our system (tuned with different ranking parameters) and Google on some Friendfeed conversations searches. Users were asked to judge (according to their personal interest) a set of Friendfeed conversations about a specific topic. Here you can see the results (Google is green, The other two are our system with the ranking based on popularity [purple] and density [blue]), higher values mean higher a better judgement on results showed by the search system.

SIGSNA TwitterGet: a Twitter downloading tool

When we started our researches on Twitter propagation we, as many others, used the well known Twapperkeeper service in order to save and download Tweets. Later we moved toward an ad hoc solution which was highly customised for our specific needs and infrastructure.
Recently due to a change in Twitter policy Twapperkeeper removed the export and download capability, leaving many researchers without an important research tool.
We then decided to share a simplified version of our system that should be able to run properly on many systems. You can download it from the tools page. The attached manual should explain everything you need to know to install and use the system.
Please note that this is an alpha version so please give us feedback or request for new functions.
We really hope this will help many researchers looking for a (relatively) simple to installa alternative to Twapperkeeper.

#rifarelitalia dynamic ReTweet Chain


[Italian]
“Per Rifare l’Italia” was a conference hold in the Italian Lower Chamber (Campera dei Deputati) on Feb. 2 2011, organised by Telecom Italia – Working Capital about the strategies to bring a higher level of innovation in Italy (the hashtag sound like “rebuilding Italy). Keynote speaker was Edmund S. Phelps (winner of the 2006 Bank of Sweden Prize in Economic Sciences in Memory of Alfred Nobel), You can read the full transcript of his speech here.
Following the official hashtag we have done this short animation that clearly show the dynamic evolution of the ReTweets related to the event. It is interesting to point out that while during the conference there was a good level of activity (most due to a numbers of blogger live tweetting the event) very few interactions happens during the afternoon.
From a technical poing of view the video (made with Gephi) shows only the ReTweet chains of the messages with the #rifarelitalia official hashtag. Both size and colour of the nodes indicate the Degree centrality of the node.
[English Version]
“Per Rifare l’Italia”, evento organizzato il 2 Febbraio 2011 da Telecom Italia presso la Camera dei Deputati, ha visto la partecipazione del premio Nobel per l’economia Edmund S. Phelps. La visualizzazione che qui proponiamo – basata sui tweet contenenti l’hashtag ufficiale #rifarelitalia – mostra le dinamiche di propagazione dei messaggi durante l’evento (decisamente elevate) e – al tempo stesso – come al termine dell’evento gran parte della propagazione in rete tenda ad esaurirsi.
Da un punto di vista tecnico il video – realizzato con Gephi – mostra la rete dei ReTweet (l’aggiustamento è dei nodi è ottenuto tramite l’applicazione di un Force Atlas Layout) su intervalli temporali di un’ora. Sia le dimensioni che l’intensità di colore dei nodi sono calcolare sulla base della degree centrality del nodo stesso.