e-Xploration
Follow
Find
22.1K views | +8 today
e-Xploration
antropologo.net, dataviz, collective intelligence, algorithms, social learning, social change, digital humanities
Curated by luiy
Your new post is loading...
Your new post is loading...
Rescooped by luiy from Global Brain
Scoop.it!

Distributed Intelligence In Design ( PDF ) Download All You Want ...

Distributed Intelligence In Design ( PDF ) Download All You Want ... | e-Xploration | Scoop.it
Tuba Kocatirk, Dr. Benachir Medjdoub, "Distributed Intelligence In Design Filehost Mirrors: Uploaded.net, Rapidgator.net " English | ISBN: 1444333380 | February 8, 2011 | 280 page.

Via Spaceweaver
more...
No comment yet.
Rescooped by luiy from Into the Driver's Seat
Scoop.it!

Vintage Data Visualization: 35 examples from before the Digital Era | #dataviz #history

Vintage Data Visualization: 35 examples from before the Digital Era | #dataviz #history | e-Xploration | Scoop.it

Graphics, charts, diagrams and visual data representations have been published on books, newspapers and magazines since they've existed, not to mention old maps and scientific illustrations...

 

Despite the lack of tools such as the ones we have at our disposal nowadays, they are as inspiring and important as the best contemporary visualizations. Visit the article link for a gallery of vintage visualizations...


Via Lauren Moss, Jim Lerman
more...
Mariana Soffer's comment, July 20, 2013 9:39 AM
my pleasure
Charlley Luz's curator insight, July 20, 2013 10:26 AM

muito legal, os Infográficos antes de existir a internet. 35 exemplos de infográficos no papel :) Achei falta do Marcha para Moscou do Minard http://www.datavis.ca/gallery/re-minard.php ;

Leoncio Lopez-Ocon's curator insight, July 20, 2013 2:57 PM

El brasileño Tiago Veloso, fundador de Visual Loop, nos ofrece 35 interesantísimas representaciones visuales de distintos fenómenos y eventos que permiten hacer un paseo por la historia de la ilustración científica.

Rescooped by luiy from Social and digital network
Scoop.it!

History Of Social Media - Infographic | #cyberculture

History Of Social Media - Infographic | #cyberculture | e-Xploration | Scoop.it

Do you know the history of social media? Think we'll remember Facebook in 20 years? This detailed timeline is a must-see.

 

Social media began decades before the Facebook era. It started, more or less, with CompuServe and Arpanet back in 1969. A couple years later, the first-ever email was sent.

It has evolved over the past few decades into a powerful tool, as seen in this social media history timeline. With so much that’s happened over the past few decades, we can only guess what’s coming next for social media.


Via Lauren Moss, Pascale Mousset
more...
Charles Rein's comment, July 24, 2013 3:38 PM
From the land of "Wired Telephones" USA, we can now look at how explosive Global growth and the potential 5-7 billion people who will always use a Smart or Cell device
Eleonora Guglielman's curator insight, August 1, 2013 9:18 AM

Nice infographic

lbligen's curator insight, August 4, 2013 5:44 AM

Always new background information.

Scooped by luiy
Scoop.it!

Tour the World's Webcams With the Search Engine for the Internet of Things | #surveillance #public

Tour the World's Webcams With the Search Engine for the Internet of Things | #surveillance #public | e-Xploration | Scoop.it
THE INTERNET OF THINGS Shodan A map of the world's publicly available webcams. Image: Shodan
luiy's insight:

When Dan Tentler wants to find something on the internet, he doesn’t use Google or Bing. Tentler, a freelance security consultant, is a road-less-traveled kind of guy. He likes to check out the internet’s alleyways and backroads. And for people like him him, there’s only one search engine. It’s called Shodan.

 

Google has done a masterful job of indexing the human experience — the webpages, books, Word documents, and images and videos that make up our life. But Shodan looks for something simpler. It’s looking for all the stuff that’s connected to the internet, from routers and refrigerators to live webcams that give you a glimpse inside people’s homes to, well, who knows what.

 

These odd little devices, overlooked by Google and Bing, are the things that Tentler finds interesting. Using Shodan, he’s taken a tour of a Scottish country house, explored a stationary GPS receiver in Alaska, and even examined the control panel for a swimming pool. “It’s like looking at a street or a set of the buildings, but not from the front,” he says. “Not from where their marketing department wants you to see it. But from where the shipping and receiving department uses it.”

 

Using a network of 24 computers nested in service providers across the world, Shodan reaches out and methodically probes machines across the globe asking them the simplest of questions: What can you tell me about yourself? And you’d be surprised what it has found.....

more...
No comment yet.
Scooped by luiy
Scoop.it!

The Rise of Data Science in the Age of Big Data Analytics | #datascience #analytics

luiy's insight:

The reason why Big Data is important is because we want to use it to make sense of our world. It's tempting to think there's some "magic bullet" for analyzing big data, but simple "data distillation" often isn't enough, and unsupervised machine-learning systems can be dangerous. (Like, bringing-down-the-entire-financial-syste­m dangerous.) Data Science is the key to unlocking insight from Big Data: by combining computer science skills with statistical analysis and a deep understanding of the data and problem we can not only make better predictions, but also fill in gaps in our knowledge, and even find answers to questions we hadn't even thought of yet.

more...
No comment yet.
Scooped by luiy
Scoop.it!

The topography of Tweets | #dataviz

The topography of Tweets | #dataviz | e-Xploration | Scoop.it
Sometimes a map is not enough. What if instead of just showing roads and boundaries, it was a 3D model you could manipulate and explore? That is exactly what our data visualization scientist Nicola......
luiy's insight:

The mountain ranges you see here are not natural geography but the landscape of Tweets — billions of them, visualized across cities. The peaks represent the places most Tweets are sent from, the troughs the fewest. Explore New York closely and you can pick out the Brooklyn and the Queensboro bridge — even the Staten Island ferry.

more...
No comment yet.
Rescooped by luiy from Data is big
Scoop.it!

Scikit-learn, machine learning and #cybercrime attribution | #algorithms #python

Scikit-learn, machine learning and #cybercrime attribution | #algorithms #python | e-Xploration | Scoop.it

Robert Layton http://2013.pycon-au.org/schedule/30019/view_talk The scikit-learn library is a rapidly growing open source toolkit for machine learning in python.


Via ukituki
luiy's insight:

 

The scikit-learn library is a rapidly growing open source toolkit for machine learning in python. It allows for practitioners and researchers to apply machine learning in a variety of applications and is used by companies worldwide. Developed by programmers from around the world, the project has a large (and increasing) number of machine learning algorithms, a very useful set of utility functions and has also spawned a set of detail

more...
No comment yet.
Rescooped by luiy from Urbanisme
Scoop.it!

#Cartographie d'une ville selon les races et ethnies de ses habitants | #dataviz

Maps of racial and ethnic divisions in US cities, inspired by Bill Rankin's map of Chicago, updated for Census 2010. Red is White, Blue is Black, Green is Asian, Orange is Hispanic, Yellow is Other, and each dot is 25 residents.

Via 15marches, Lockall
more...
No comment yet.
Scooped by luiy
Scoop.it!

The #Durkheim Project : Data-mining social networks to spot patterns indicating #suicidal behavior | #dataviz #predict

The #Durkheim Project : Data-mining social networks to spot patterns indicating #suicidal behavior | #dataviz #predict | e-Xploration | Scoop.it
Just as after the Vietnam war there is an epidemic of suicides going on among American veteran of the Middle East wars. According to the VA there is...
luiy's insight:

Unlike the Vietnam days, veterans today search for information (or vent their troubles) about suicide on Google and Facebook. A new project, newly launched by DARPA and Dartmouth University, is trying something new: Data-mining social networks to spot patterns indicating suicidal behavior.

 

Called The Durkheim Project, named for the Victorian-era psychologist, it is asking veterans to offer their Twitter and Facebook authorization keys for an ambitious effort to match social media behavior with indications of suicidal thought. Veterans' online behavior is then fed into a real-time analytics dashboard which predicts suicide risks and psychological episodes.

 

However, there's a caveat: The Durkheim Project, which launched on July 1, is only a study into the effectiveness of predictive analytics for mental health. Veterans who participate will only be monitored, and have to receive any needed mental health assistance through the VA and other sources.

more...
No comment yet.
Rescooped by luiy from Digital Humanities for beginners
Scoop.it!

The Rise of the Machines | #DH

The Rise of the Machines | #DH | e-Xploration | Scoop.it
Stephen Mitchell suffered from allergies. “When the trees come out, I can’t see. People stand around saying, ‘Isn’t it lovely,’ but I weep,” he told the New

Via Pierre Levy
luiy's insight:

During the previous fifteen years, computers had been making inroads into the humanities, emerging as more than a passing curiosity by the mid 1960s. The digital humanities—or “humanities computing,” as it was then known—used machines the size of small cars, punch cards, and data recorded on magnetic tape. To many scholars, its methods, which depended on breaking down texts into data elements, seemed alien, as did the antiseptic atmosphere of the computer lab. But for those who didn’t mind working away from the comforting smell of musty old books, a new field was opening up, a hybrid discipline that would receive significant assistance from the National Endowment for the Humanities, which made its debut in 1965.

 

A PRIEST AND A SCIENTIST WALK INTO A COMPUTER LAB . . .


In the years following World War II, the sciences—physical, biological, and social—embraced computers to work on complex calculations, but it took the humanities a little longer to see the value of computing. One of the biggest challenges for humanists was the question of how to turn language, the core operating system of the humanities, into numbers in order to be compiled and calculated. At this point in the history of computing, all data had to be in numerical form. It’s not sur-prising, then, that some of the first humanities projects were indexes and concordances, since the location of a word could be given a numerical value.....

more...
No comment yet.
Rescooped by luiy from Social Network Analysis
Scoop.it!

Use These Secret #NSA Google Search Tips to Become Your Own Spy Agency | Wired | #privacy #surveillance

Use These Secret #NSA Google Search Tips to Become Your Own Spy Agency | Wired | #privacy #surveillance | e-Xploration | Scoop.it
Want to know how to "'hack" Google like the pros? The NSA has released a book it produced for its workers on how to find intelligence on the web.

Via Pierre Levy, Jay Ratcliff
luiy's insight:

The 643-page tome, called Untangling the Web: A Guide to Internet Research (.pdf), was just released by the NSA following a FOIA request filed in April by MuckRock, a site that charges fees to process public records for activists and others.

 

The book was published by the Center for Digital Content of the National Security Agency, and is filled with advice for using search engines, the Internet Archive and other online tools. But the most interesting is the chapter titled “Google Hacking.”

 

Say you’re a cyberspy for the NSA and you want sensitive inside information on companies in South Africa. What do you do?

Search for confidential Excel spreadsheets the company inadvertently posted online by typing “filetype:xls site:za confidential” into Google, the book notes.

 

Want to find spreadsheets full of passwords in Russia? Type “filetype:xls site:ru login.” Even on websites written in non-English languages the terms “login,” “userid,” and “password” are generally written in English, the authors helpfully point out.

 

Misconfigured web servers “that list the contents of directories not intended to be on the web often offer a rich load of information to Google hackers,” the authors write, then offer a command to exploit these vulnerabilities — intitle: “index of” site:kr password.

 

“Nothing I am going to describe to you is illegal, nor does it in any way involve accessing unauthorized data,” the authors assert in their book. Instead it “involves using publicly available search engines to access publicly available information that almost certainly was not intended for public distribution.” You know, sort of like the “hacking” for which Andrew “weev” Aurenheimer was recently sentenced to 3.5 years in prison for obtaining publicly accessible information from AT&T’s website.

 

Stealing intelligence on the internet that others don’t want you to have might not be illegal, but it does come with other risks, the authors note: “It is critical that you handle all Microsoft file types on the internet with extreme care. Never open a Microsoft file type on the internet. Instead, use one of the techniques described here,” they write in a footnote. The word “here” is hyperlinked, but since the document is a PDF the link is inaccessible. No word about the dangers that Adobe PDFs pose. But the version of the manual the NSA released was last updated in 2007, so let’s hope later versions cover it.

more...
No comment yet.
Scooped by luiy
Scoop.it!

Interactive Atlas Highlights Water Use Issues in California | Wired | #dataviz #hackathon #WaterRights

Interactive Atlas Highlights Water Use Issues in California | Wired | #dataviz #hackathon #WaterRights | e-Xploration | Scoop.it
Water is always a contentious issue in California. With 38 million thirsty people, a $43 billion agriculture industry, often erratic rainfall, and a bewildering tangle of policies, it's no wonder.
luiy's insight:

The creators of anew California water atlas are hoping to make data on the state’s water resources more accessible and comprehensible through a series of interactive maps.

 

“We’re hoping we can inspire dialogue around some of these policy issues,” says Laci Videmsky, a designer and one of the directors of the project, which is funded by Patagonia and the Resource Renewal Institute, a non-profit organization based in Mill Valley, California. Videmsky and his co-director Chacha Sikes, a programmer, hope the new digital atlas will help journalists, activists, and ordinary citizens learn more about how water is allocated and used in the state.

 

The duo met last year at Code for Oakland, a hackathon for civic-minded software developers, and started brainstorming projects that could have state-wide impact.

 

Chacha Sikes (left) and Laci Videmsky. Photo: Alex Washburn/Wired

A major inspiration for the new project was a1979 water atlas commissioned by California governor Jerry Brown (yes, he was governor back then too). The book became a cult classic among cartography geeks for its beautiful maps and infographics, Videmsky says, but it didn’t reach many people and didn’t have much impact on water policy.

more...
No comment yet.
Scooped by luiy
Scoop.it!

The Startup Universe : A Visual Guide to #Startups, #Founders & #Venture Capitalists | #dataviz

The Startup Universe : A Visual Guide to #Startups, #Founders & #Venture Capitalists | #dataviz | e-Xploration | Scoop.it
Uncover new stories and make sense of the complex relationships between startups, founders and Venture Capitalists with Visually's Startup Universe
luiy's insight:
The Startup UniverseA Visual Guide to Startups, Founders & Venture Capitalists

The Startup Universe displays and explores the relationships between startup companies and their founders and investors (Venture Capitalists) since 1990.

 

Startups are grouped into 19 categories, based on the type of products or services they deliver. Each category is represented by an unique color. In addition to that, startups are visually sized according to the amount of financing they have raised, with each individual round displayed as well.

Startups are positioned within the main interface on a horizontal interactive timeline, based on the year they were founded. Details about each company are provided below the timeline.

 

By clicking on the left or right columns - Venture Capitalist or Founders' names - tangential stories can be explored: how many startups has a VC company funded and are there any time or startup category trends or patterns? How many startups have been founded by the same person?

Startups, VCs and founders can also be directly browsed by entering a name in the search field.

 

All data comes from the CrunchBase API.

 

The Startup Universe aims at being a comprehensive exploratory view of the past 22 years (and counting) of startup life. As you explore our tool, we hope you will discover new stories, relationships and trends.

more...
No comment yet.
Rescooped by luiy from Social Network Analysis #sna
Scoop.it!

Building a follower model from scratch | #dataviz #SNA

Building a follower model from scratch | #dataviz #SNA | e-Xploration | Scoop.it
Our follower graph has millions of nodes and billions of edges, making it an interesting challenge to maintain and scale data as we build out the interest graph. The model is similar to those of...

Via M. Edward (Ed) Borasky, ukituki
luiy's insight:

Our follower graph has millions of nodes and billions of edges, making it an interesting challenge to maintain and scale data as we build out the interest graph. The model is similar to those of Twitter or Facebook, but with some key differences based around interests that we account for in the product development and design phases.

 

The final version of the Pinterest follower service was developed, migrated and deployed in about 8 weeks with one full time engineer and 2-3 part time engineers.

 

Here I’ll explain how the service-oriented architecture has helped us develop and maintain the service as a unit of its own.

more...
No comment yet.
Rescooped by luiy from Politique des algorithmes
Scoop.it!

Big Data bullshit. by Christian Fauré | #controverses #critics #bigdata

Big Data bullshit. by Christian Fauré | #controverses #critics #bigdata | e-Xploration | Scoop.it
Je suis particulièrement étonné par le discours actuel sur les big data ; discours selon lequel nous serions passé de la causalité à la corrélation.

Via Dominique Cardon
luiy's insight:

Je suis particulièrement étonné par le discours actuel sur les big data ; discours selon lequel nous serions passé de la causalité à la corrélation. Je pense surtout à la thèse de Viktor Mayer-Schönberger et Kenneth Cukier, dans leur livre Big Data : une révolution qui va transformer notre façon de vivre, de travailler et penser. (voir l’excellent article de recension de Hubert Guillaud : Big Data : nouvelle étape de l’informatisation du monde.)

more...
No comment yet.
Rescooped by luiy from artificial intelligence for students
Scoop.it!

Internet of Things | Wired | #trends #tendances

Internet of Things | Wired | #trends #tendances | e-Xploration | Scoop.it
THE INTERNET OF THINGS FEATURED 07.08.13 Tour the World's Webcams With the Search Engine for the Internet

Via Scott Turner
more...
No comment yet.
Scooped by luiy
Scoop.it!

High Performance #Predictive #Analytics in R and Hadoop

Presentation given by US Chief Scientist, Mario Inchiosa, at the June 2013 Hadoop Summit in San Jose, CA. ABSTRACT: Hadoop is rapidly being adopted as a major p
luiy's insight:

I'm a bit late catching up on this, but Mario Inchosa (Revolution Analytics US Chief Scientist) gave a standing-room-only talk on high-performance predictive analytics in R and Hadoop at last month'sHadoop Summit. In the talk, he described some of the progress we've made integrating the ScaleR parallel external-memory algorithms into the Hadoop platform. He described some of the design considerations that makes the ScaleR algorithms so fast with big data, and how the architecture is being integrated into the Hadoop platform. He also described the interface from the perspective of an R programmer working at a desktop, but using the power of a remote Hadoop cluster for the analytics. I've embedded his slides below:

more...
No comment yet.
Rescooped by luiy from Cooperation Theory & Practice
Scoop.it!

Trends in #Cognitive Sciences - Human #cooperation

"Why should you help a competitor? Why should you contribute to the public good if free riders reap the benefits of your generosity? Cooperation in a competitive world is a conundrum. Natural selection opposes the evolution of cooperation unless specific mechanisms are at work. Five such mechanisms have been proposed: direct reciprocity, indirect reciprocity, spatial selection, multilevel selection, and kin selection. Here we discuss empirical evidence from laboratory experiments and field studies of human interactions for each mechanism. We also consider cooperation in one-shot, anonymous interactions for which no mechanisms are apparent. We argue that this behavior reflects the overgeneralization of cooperative strategies learned in the context of direct and indirect reciprocity: we show that automatic, intuitive responses favor cooperative strategies that reciprocate."


Via Howard Rheingold
luiy's insight:

Highlights:

Theoretical work has revealed five mechanisms for the evolution of cooperation.These are direct and indirect reciprocity, and spatial, multilevel, and kin selection.We present experimental evidence for each mechanism operating in human behavior.We show that reciprocation is an automatic response, which implies that reciprocity has a key role.

 

 

more...
Howard Rheingold's curator insight, July 15, 2013 3:34 PM

Reciprocation is deeply embedded. Nowak et al's work on the evolution of cooperation yields clues for present-day human behavior.

Peter C. Newton-Evans's curator insight, July 20, 2013 5:13 PM

From an assumed, unproven instinct for competition, we have gone to a demonstrable instinct for cooperation.

Rescooped by luiy from OSINT News
Scoop.it!

Global #Surveillance, Remapped | #dataviz #privacy

Global #Surveillance, Remapped | #dataviz #privacy | e-Xploration | Scoop.it
 A 'heat map' published by The Guardian last month exposed the massive collection of the world's electronic data by the US National Security Agency over a 30-day period in March 2013.

Via Paulo Félix
luiy's insight:

A ‘heat map’ published by The Guardian last month exposed the massive collection of the world’s electronic data by the US National Security Agency over a 30-day period in March 2013. The chart sparked immediate controversy and diplomatic outcry, as nations allied to the United States discovered they were subject to a massive programme of unrestricted electronic surveillance.

 

The original map, however, was misleading. Its colour-code reflected the absolute number of data elements (DNI) gathered, rather than the intensity of snooping. However most people care about their own privacy: that is, given the country in which you live, how much of your personal communication is being scooped by the NSA?

 

To answer this puzzle, Factblink reanalysed the Guardian heat-map data. The original map uses a continuous colour-coding scheme, so by extracting the red-green-blue information we are able to obtain a continuous “colour intensity” variable for all countries. From the slide we already have actual DNI figures for six countries, as well as knowledge of the total number of items, and together with a few extra parameters (e.g. no negative values, non-linear scaling) we can impute the remaining values. Combined with internet usage data by country, this allows us to work out the intensity of collection, on a per internet user basis (below).

more...
No comment yet.
Rescooped by luiy from War, Cyberwar, Geopolitics
Scoop.it!

65% of Internet Users Are #Cybercrime Victims [INFOGRAPHIC]

65% of Internet Users Are #Cybercrime Victims [INFOGRAPHIC] | e-Xploration | Scoop.it
What are your chances of being hacked? Find out some easy ways to prevent a breach of your system.

Via Pierre Levy
luiy's insight:

Cybersecurity is a rising concern globally — for individuals, businesses and nations. Unfortunately, many either don't take the threat seriously or aren't doing enough to protect themselves from cybercrime.

A Ponemon Institute survey of 583 U.S. companies found 90% had been hacked in the past 12 months. Of those companies, most admitted their networks had been breached more than once, and more than half expressed little to no confidence that they could ward off such attacks in the future.

 

SEE ALSO: 10 Spooky Cyberattacks in 2012 [INFOGRAPHIC]

America is not the most-hacked nation, but we still have a long way to go in terms of preventative measures. The most common password, after all, is still simply "password," the worst choice you could make to secure your online identity.

Check out the infographic below to learn more about the state of cybersecurity and how to protect your online privacy.

more...
No comment yet.
Scooped by luiy
Scoop.it!

Nations Buying as Hackers Sell Flaws in Computer Code | #hacking #NSA

Nations Buying as Hackers Sell Flaws in Computer Code | #hacking #NSA | e-Xploration | Scoop.it
Governments pay hundreds of thousands of dollars to learn about and exploit weaknesses in the computer systems of foreign adversaries.
luiy's insight:

The hackers, Luigi Auriemma, 32, and Donato Ferrante, 28, sell technical details of such vulnerabilities to countries that want to break into the computer systems of foreign adversaries. The two will not reveal the clients of their company, ReVuln, but big buyers of services like theirs include the National Security Agency— which seeks the flaws for America’s growing arsenal of cyberweapons — and American adversaries like the Revolutionary Guards of Iran.

 

All over the world, from South Africa to South Korea, business is booming in what hackers call “zero days,” the coding flaws in software like Microsoft Windows that can give a buyer unfettered access to a computer and any business, agency or individual dependent on one.

 

Just a few years ago, hackers like Mr. Auriemma and Mr. Ferrante would have sold the knowledge of coding flaws to companies like Microsoft and Apple, which would fix them. Last month, Microsoft sharply increased the amount it was willing to pay for such flaws, raising its top offer to $150,000.

more...
No comment yet.
Scooped by luiy
Scoop.it!

Data Mining for Human Cognition | #datascience

Data Mining for Human Cognition | #datascience | e-Xploration | Scoop.it
Lumosity hopes to advance the study of human cognition with a new project that will leverage big data collected from its popular website.
luiy's insight:

Since launching in 2007, Lumosity has attracted plenty of attention on the Web with its quirky ads that tout the capability of its online games to make your brain stronger. There may be some debate whether the games actually make you smarter. But with a study recently published in the journal, Frontiers in Neuroscience, there's little doubt that the San Francisco, California company is serious about advancing the understanding of human cognition.

 

"New technologies and research platforms have the potential to transform the speed, scale, efficiency, and range of topics in which neuroscience research is conducted," said P. Murali Doraiswamy, Professor of Psychiatry at DukeUniversityMedicalCenter and co-author of the study. "This study is interesting because it brings to light the possibilities of what we can uncover by taking a big data approach to cognitive performance research."

more...
No comment yet.
Scooped by luiy
Scoop.it!

On the Interplay between Social and Topical Structure | #dataviz #predictions

On the Interplay between Social and Topical Structure | #dataviz #predictions | e-Xploration | Scoop.it
Your friends and your topics of interests are intuitively related – people form friendships through mutual interests and at the same time people discover new interests through friends.
luiy's insight:

Your friends and your topics of interests are intuitively related – people form friendships through mutual interests and at the same time people discover new interests through friends. We are interested in exploring the ways in which social and topical structures can predict each other. We ask two basic questions:

How well can a person’s topical interests predict who her friends are?How well can the social connections among the people interested in a topic predict the future popularity of that topic?

In order to answer these questions we study 5 million Twitter users. We study their hashtag usage to identify topical interests and their follower/@-messages to identify two different kinds of social relationships.

To predict whether two users have a social relationship based on their hashtags, we use logistic regression models trained on a wide range of distance measures, measuring topical similarity. Interestingly, one of the most predictive measures is also one of the simplest ones to compute: the size of the smallest hashtag shared by the two users.

Our full model has an accuracy of 77% when predicting follower relationships and 86% when predicting @-message relationships. We also find that predicting strong ties is much easier that predicting weak ties. Our model achieves an accuracy of up to 98% when predicting the strongest pairs, which exchanged more than 20 @-messages.

more...
No comment yet.