Here's a story from The Economist regarding about how the term "social graph", even though it is hot, is not really any thing new. Social graph is being mentioned everywhere on the Net because of Facebook's Mark Zuckerberg and how Facebook is touting and is being given credit to the term "social graph". However, the social graph is something that has been done in Computer Science, according to Eric Schmidt, CEO of Google.
It seems that social graph is the new hot thing in Silicon Valley, and everyone is getting on the bandwagon. It'll be interesting to see if it will still remain hot for a while, or go like the dot com companies.
On Technorati: social graph, Facebook
Friday, November 23, 2007
Sunday, November 11, 2007
Updated my web site, check it out!
I've updated my web site with a new layout which makes it easier to read and easier to navigate. Let me know what you think and if there are any problems.
Thursday, November 08, 2007
World Usability Day 2007
Today is World Usability Day, where the focus is usability in healthcare. In Toronto, the Toronto chapter of SIGCHI called TorCHI is having a meeting of usability experts and presentations at the Bahen Center at the University of Toronto. Ilona Posner is introducing World Usability Day, she is talking about the problems that she experienced in trying to see the webcast of World Usability Day from Boston, explaining about the usability problems of web software and computers. The organizers of this event are TorCHI and Usability Professionals Association Worldwide. Last year World Usability Day 2006 had 40000 participants, 225 events, in 175 cities and 35 countries. In Toronto, there were 2 events and 200+ attendees. Tonight's program is the following:
1) Web 2.0 and Healthcare, 2) Patient Safety and Human Factors, and 3) Reality Checkup: A Conversation with a Physician.
The first presenter is Holly Witteman, PhD Candidate in Mechanical and Industrial Engineering at the University of Toronto. The topic is Usability, eHealth and Web 2.0. Within eHealth, Web 2.0 is characterized by open community and communication in the health area, and the technologies applied to health. Blogs can be used for personal expression. Wikis can be used for collaboration for medical education, for creating repositories of information. Social networks are embedded within the Web 2.0 web sites like patientslikeme. CarePages is also another example of social networking site for health care, also sermo is another example. Mashups can be used to look at disease outbreaks around the world, and sicknesses around locations using mapping tools like Google Maps. Tags are also another technology used in health context like in YouTube. Podcasts are also a popular medium for distributing information in audio. So what does this have to do with usability? User-generated content introduces the notion of credibility, is the information credible and valid. In health care, the information is evaluated by a community of experts to determine the credibility. When it comes to health information online, one size does not fit all. There are individual differences and need to be incorporated in usability assessments.
The second speaker was Anjum Chagpar from the University Health Network who talked about a Systems Approach to Patient Safety. She is the manager of a lab looking at next generation medical devices. She gave an example of Denise Melanson who died because she was infused with 4 days of a drug dose within 4 hours. The cause of this accident was multifold. The label on the drug was difficult to find the dose information, the dose information was in brackets (1.2 mL/hour) instead the nurse read the first number which was 28.8 mL/24 hour (which was given in an hour). Second, there were interface issues with the infusion pump. There was no check for unsafe values to enter so the pump allowed the nurse to enter 28.8. All this shows that there is a need to design systems that minimize errors. Health care is changing from secrecy to disclosure, from a blame culture to a just culture. Why do we have poor design in healthcare systems? Because the devices used in healthcare have different market drivers from consumer technology, the devices are not high tech because there is high risk. Human factors is not incorporated in the design process because due to the complexity of the health care environments and there are no consistent user interfaces. Therefore, health care needs human factors.
I didn't attend the third presentation as I head to head back home.
Happy World Usability Day!
On Technorati: World Usability Day
1) Web 2.0 and Healthcare, 2) Patient Safety and Human Factors, and 3) Reality Checkup: A Conversation with a Physician.
The first presenter is Holly Witteman, PhD Candidate in Mechanical and Industrial Engineering at the University of Toronto. The topic is Usability, eHealth and Web 2.0. Within eHealth, Web 2.0 is characterized by open community and communication in the health area, and the technologies applied to health. Blogs can be used for personal expression. Wikis can be used for collaboration for medical education, for creating repositories of information. Social networks are embedded within the Web 2.0 web sites like patientslikeme. CarePages is also another example of social networking site for health care, also sermo is another example. Mashups can be used to look at disease outbreaks around the world, and sicknesses around locations using mapping tools like Google Maps. Tags are also another technology used in health context like in YouTube. Podcasts are also a popular medium for distributing information in audio. So what does this have to do with usability? User-generated content introduces the notion of credibility, is the information credible and valid. In health care, the information is evaluated by a community of experts to determine the credibility. When it comes to health information online, one size does not fit all. There are individual differences and need to be incorporated in usability assessments.
The second speaker was Anjum Chagpar from the University Health Network who talked about a Systems Approach to Patient Safety. She is the manager of a lab looking at next generation medical devices. She gave an example of Denise Melanson who died because she was infused with 4 days of a drug dose within 4 hours. The cause of this accident was multifold. The label on the drug was difficult to find the dose information, the dose information was in brackets (1.2 mL/hour) instead the nurse read the first number which was 28.8 mL/24 hour (which was given in an hour). Second, there were interface issues with the infusion pump. There was no check for unsafe values to enter so the pump allowed the nurse to enter 28.8. All this shows that there is a need to design systems that minimize errors. Health care is changing from secrecy to disclosure, from a blame culture to a just culture. Why do we have poor design in healthcare systems? Because the devices used in healthcare have different market drivers from consumer technology, the devices are not high tech because there is high risk. Human factors is not incorporated in the design process because due to the complexity of the health care environments and there are no consistent user interfaces. Therefore, health care needs human factors.
I didn't attend the third presentation as I head to head back home.
Happy World Usability Day!
On Technorati: World Usability Day
Google's OpenSocial API
It looks like my thoughts as to why Google hasn't explored the social networking space is now answered. Besides having Google creating a large social network graph (according to Eric Schmidt), Google is also creating APIs to allow applications to easily use social networking information. Google has something called the OpenSocial API, more information from one of Google's employees on their blog. It seems Google has quite a number of partners on board using their API. This is something to definitely look at, as you can't reckon with a force like Google.
On Technorati: OpenSocial, Google, social network
On Technorati: OpenSocial, Google, social network
Sunday, November 04, 2007
Social Networking in the Learning Sciences - Social Networking Conference @ U of T
“A Wiki-Based Exchange Community for the Learning Sciences” Jim Slotta, Associate Professor OISE/ UT
In this talk, some of the information that Jim talked about overlapped during his talk at the CASCON Second Working Conference on Social Computing and Business. Social networking is providing new opportunities for knowledge communities. The whole idea is to connect students in the classroom with social computing tools that students are using. WISE is a research platform that allows students to collaborate and is available on SourceForge. Jim says that to make a community is not SourceForge. To create a community around SourceForge is through wikis to build online community.
On Technorati: Social Networking Symposium
In this talk, some of the information that Jim talked about overlapped during his talk at the CASCON Second Working Conference on Social Computing and Business. Social networking is providing new opportunities for knowledge communities. The whole idea is to connect students in the classroom with social computing tools that students are using. WISE is a research platform that allows students to collaborate and is available on SourceForge. Jim says that to make a community is not SourceForge. To create a community around SourceForge is through wikis to build online community.
On Technorati: Social Networking Symposium
Friday, November 02, 2007
Making Personal Network Analysis More Accessible - Social Networking Conference @ U of T
Making Personal Network Analysis More Accessible
Bernie Hogan, Research Director, NetLab, UofT
In this talk, Bernie is talking about tools to make use of personal network analysis and make it accessible to the average user. In yesterday’s presentation, Bernie talked about the Connected Lives project which studies individuals from East York. It is difficult to analyze data that comes about from name generators. So the idea is to create a software to help to analyze the data that come out from name generators. Bernie and his colleagues at NetLab created visualizations of network data using participant-aided sociogram.
He is talking about how there is a problem with existing applications. They are designed for a single network (UCINET, NetDraw, Pajek), they have no GUI and steep learning curve (R, JUNG). So what they have done is modify existing applications, for example, GUESS (from Eytan Adar) + GraphModifier. Another problem is that the applications have virtually no interactive analysis. Batch processing of data has high fixed cost (have to know loops in R). So, the applications currently push in data, and then answers come out. What we want is data that goes in, answers come out and become source of new data. To address these issues, they created Egotistics software which is available on Sourceforge. In Egotistics, users can program, and batch process cohesive subgroups like k-plexes (I could have used that for my analysis!). One of the things to improve and encourage others to use Egotistics is to provide a web API to enable people to do analysis (not yet but should do).
I believe this talk really addresses how we need tools to discover communities and our social networks, so I'm going to look into these tools in a little more detail.
On Technorati: Social Networking Symposium, Egotistics
Bernie Hogan, Research Director, NetLab, UofT
In this talk, Bernie is talking about tools to make use of personal network analysis and make it accessible to the average user. In yesterday’s presentation, Bernie talked about the Connected Lives project which studies individuals from East York. It is difficult to analyze data that comes about from name generators. So the idea is to create a software to help to analyze the data that come out from name generators. Bernie and his colleagues at NetLab created visualizations of network data using participant-aided sociogram.
He is talking about how there is a problem with existing applications. They are designed for a single network (UCINET, NetDraw, Pajek), they have no GUI and steep learning curve (R, JUNG). So what they have done is modify existing applications, for example, GUESS (from Eytan Adar) + GraphModifier. Another problem is that the applications have virtually no interactive analysis. Batch processing of data has high fixed cost (have to know loops in R). So, the applications currently push in data, and then answers come out. What we want is data that goes in, answers come out and become source of new data. To address these issues, they created Egotistics software which is available on Sourceforge. In Egotistics, users can program, and batch process cohesive subgroups like k-plexes (I could have used that for my analysis!). One of the things to improve and encourage others to use Egotistics is to provide a web API to enable people to do analysis (not yet but should do).
I believe this talk really addresses how we need tools to discover communities and our social networks, so I'm going to look into these tools in a little more detail.
On Technorati: Social Networking Symposium, Egotistics
Networks, Job Search and Labour Markets: Information Sharing as a Structured Process - Social Networking Conference @ U of T
"Networks, Job Search and Labour Markets: Information Sharing as a Structured Process"
Alexandra Marin, Assistant Professor, Department of Sociology, UofT
In this talk, Alexandra talks about how to use information sharing for job search. She is applying social capital to the process of job search and how social networks of contacts can be used for finding jobs. I asked the question about studying job search using social network sites like LinkedIn and getting a job through the LinkedIn network. Alexandra mentioned how people do not really use LinkedIn and use their physical contacts rather than from a web site.
On Technorati: Social Networking Symposium"
Alexandra Marin, Assistant Professor, Department of Sociology, UofT
In this talk, Alexandra talks about how to use information sharing for job search. She is applying social capital to the process of job search and how social networks of contacts can be used for finding jobs. I asked the question about studying job search using social network sites like LinkedIn and getting a job through the LinkedIn network. Alexandra mentioned how people do not really use LinkedIn and use their physical contacts rather than from a web site.
On Technorati: Social Networking Symposium"
A New Research Agenda: The Emergence of Online Social Networking Systems - Social Networking Conference @ U of T
A New Research Agenda: The Emergence of Online Social Networking Systems
Stefan Sariou and Nick Koudas
Department of Computer Science
University of Toronto
In this talk, Stefan discussed about research work that their groups are doing with studying and improving online social networking systems. Before, you didn't see much work in Computer Science on this area, but now, this is a hot topic with fertile areas for research. Specifically, Stefan is looking at social networks for access control to content, search, and content delivery and aggregation. Stefan is researching on social networking-based access for personal content. He says that the push model is an inefficient way to share content. For example, e-mail is a push model and e-mail was never designed to push content. Another way to share content is to use social networking sites for sharing content. However, sharing content online is a mess because you can start creating so many social identities and be part of so many social networks as a result. In real life, users have just one social network, but online, they have multiple social networks from different social networking sites. For example, you may have an account on Flickr, LinkedIn, YouTube, MySpace and Facebook, and you have social networks in these sites. But the people that are in your actual social network, is just one network. The online networks are just instances of your own social network. Therefore, there is a need to separate social information from content serving. I wholeheartedly agree with this.
Therefore, Stefan says that people should manage their social networks and maintain one social network. Everyone has a personal address book which they are familiar with and use. Let sites serve content and offer access control based on your social network in your address book. He says that there should not be a person or company that should manage your social network or even aggregate social networks, something of which Google is trying to do to create one huge social network (aggregations of multiple social networks combined together).
So from this, Stefan's research group is looking and developing new internet applications: Social Flickr will be released November 2007, Social BitTorrent in December 2007, and Social Google calendar in January 2008. Those are pretty aggressive time schedules for releasing the software.
Nick's work deals with social media aggregation to build a system to share information with others. His research group has created a system called BlogScope that mines the blogs in the blogosphere and it is currently tracking over 14.28 million blogs with 127.61 million posts. BlogScope can assist the user in discovering interesting information from these millions of blogs via a set of numerous unique features including popularity curves, identification of information bursts, related terms, and geographical search. From social media, based on content, we can extract communities for recommendation (which I believe they could use my work).
On Technorati: social networking symposium, blogscope
Stefan Sariou and Nick Koudas
Department of Computer Science
University of Toronto
In this talk, Stefan discussed about research work that their groups are doing with studying and improving online social networking systems. Before, you didn't see much work in Computer Science on this area, but now, this is a hot topic with fertile areas for research. Specifically, Stefan is looking at social networks for access control to content, search, and content delivery and aggregation. Stefan is researching on social networking-based access for personal content. He says that the push model is an inefficient way to share content. For example, e-mail is a push model and e-mail was never designed to push content. Another way to share content is to use social networking sites for sharing content. However, sharing content online is a mess because you can start creating so many social identities and be part of so many social networks as a result. In real life, users have just one social network, but online, they have multiple social networks from different social networking sites. For example, you may have an account on Flickr, LinkedIn, YouTube, MySpace and Facebook, and you have social networks in these sites. But the people that are in your actual social network, is just one network. The online networks are just instances of your own social network. Therefore, there is a need to separate social information from content serving. I wholeheartedly agree with this.
Therefore, Stefan says that people should manage their social networks and maintain one social network. Everyone has a personal address book which they are familiar with and use. Let sites serve content and offer access control based on your social network in your address book. He says that there should not be a person or company that should manage your social network or even aggregate social networks, something of which Google is trying to do to create one huge social network (aggregations of multiple social networks combined together).
So from this, Stefan's research group is looking and developing new internet applications: Social Flickr will be released November 2007, Social BitTorrent in December 2007, and Social Google calendar in January 2008. Those are pretty aggressive time schedules for releasing the software.
Nick's work deals with social media aggregation to build a system to share information with others. His research group has created a system called BlogScope that mines the blogs in the blogosphere and it is currently tracking over 14.28 million blogs with 127.61 million posts. BlogScope can assist the user in discovering interesting information from these millions of blogs via a set of numerous unique features including popularity curves, identification of information bursts, related terms, and geographical search. From social media, based on content, we can extract communities for recommendation (which I believe they could use my work).
On Technorati: social networking symposium, blogscope
Conversations in Social Hypertext: Telecommunity and Post-Industrial Work - Social Networking Conference @ U of T
Conversations in Social Hypertext: Telecommunity and Post-Industrial Work - Social Networking Conference @ U of T
Mark Chignell
Department of Mechanical and Industrial Engineering
University of Toronto
In this talk, Mark talked about social computing tools for telework using a software that the Interactive Media Lab created called Vocal Village which is a great tool for spatializing audio (better than Skype!). The software was tested in a Japanese company. As well, Mark introduced work about looking at community in online environments, specifically the vaccination groups which is part of my PhD research work.
On Technorati: social networking conference, Vocal Village, Interactive Media Lab
Mark Chignell
Department of Mechanical and Industrial Engineering
University of Toronto
In this talk, Mark talked about social computing tools for telework using a software that the Interactive Media Lab created called Vocal Village which is a great tool for spatializing audio (better than Skype!). The software was tested in a Japanese company. As well, Mark introduced work about looking at community in online environments, specifically the vaccination groups which is part of my PhD research work.
On Technorati: social networking conference, Vocal Village, Interactive Media Lab
Content-based Social Network Analysis of Online Communities - Social Networking Conference @ U of T
Content-based Social Network Analysis of Online Communities
Anatoliy Gruzd and Caroline Haythornthwaite
School of Library and Information Science, University of Illinois
In this talk, they analyze online communities like bulletin boards to gain more information and insight about nodes, relations and ties. Very few systems look at relational information so they focus on nodes and tie discovery. Their goal is to identify who are the actors in the network. Their approach is to use natural language processing to enhance the current techniques of building social networks. So how to obtain the social networks from online communities? There are two methods. First, you can do a chain network which is based on the chain of posting of posts and comments (like what I do for my PhD research). One of the problems with the chain network (which I also encountered as well) is what is the relation of the 3rd commenter, do they comment on the posting or the previous comment? A solution around this is to look at tie strength to the previous commenter or the poster to determine if the person is posting to the previous commenter or the poster. The second method is to do a name network by pulling the names from within the body of the text. Here is where the NLP comes into play.
The idea in the name network is to make use of node and information in text of posting. How to disambiguate names/nicknames from text, those that mean the same person. How to know the name is in the subject, is it being discussed? To determine this, they did hand coding of the items to see the categories of names. They then compared the name network with the chain network and performed ego network analysis for posts and comments. Another problem is that many times when you reply, the previous message is embedded in the post so you don't want to include this in the name generator to duplicate this. So, they removed the previous message embedded in the reply to the post.
Anatoliy Gruzd and Caroline Haythornthwaite
School of Library and Information Science, University of Illinois
In this talk, they analyze online communities like bulletin boards to gain more information and insight about nodes, relations and ties. Very few systems look at relational information so they focus on nodes and tie discovery. Their goal is to identify who are the actors in the network. Their approach is to use natural language processing to enhance the current techniques of building social networks. So how to obtain the social networks from online communities? There are two methods. First, you can do a chain network which is based on the chain of posting of posts and comments (like what I do for my PhD research). One of the problems with the chain network (which I also encountered as well) is what is the relation of the 3rd commenter, do they comment on the posting or the previous comment? A solution around this is to look at tie strength to the previous commenter or the poster to determine if the person is posting to the previous commenter or the poster. The second method is to do a name network by pulling the names from within the body of the text. Here is where the NLP comes into play.
The idea in the name network is to make use of node and information in text of posting. How to disambiguate names/nicknames from text, those that mean the same person. How to know the name is in the subject, is it being discussed? To determine this, they did hand coding of the items to see the categories of names. They then compared the name network with the chain network and performed ego network analysis for posts and comments. Another problem is that many times when you reply, the previous message is embedded in the post so you don't want to include this in the name generator to duplicate this. So, they removed the previous message embedded in the reply to the post.
Social Networking conference at U of T
I just finished presenting my talk on "Structural Analysis of Social Hypertext for Finding Sense of Community" at the Social Networking conference at U of T this morning. The gremlins of presentation attacked me today. During the last couple of slides of my talk, I accidentally kicked the AC plug (which seemed to happen to the previous speaker), and then the digital projector turned off. So, I had to finish my talk without slides, but I was fortunate that I could still read the slides off the laptop and my notes, though I wasn't quite happy with that and it kind of throwed me off. Second of all my slides didn't show up properly on the laptop in the room, normally I use the laptop in the room instead of mine to avoid switching and having my laptop reboot in the process (it's actually happened couple of times, the last time at the CASCON conference). Third, I recorded the talk on my iPod but for some strange reason it actually didn't save on the iPod (it actually didn't even start recording). Aghh!
But I do have the slides from my talk so the slides that I wanted to show, are available on my web site.
This is the first time where I could not rely on the laptop in the room, than use my own laptop.
Anyways, if you have any comments on my talk, feel free to contact me at my e-mail (achin AT cs DOT toronto DOT edu).
But I do have the slides from my talk so the slides that I wanted to show, are available on my web site.
This is the first time where I could not rely on the laptop in the room, than use my own laptop.
Anyways, if you have any comments on my talk, feel free to contact me at my e-mail (achin AT cs DOT toronto DOT edu).
Tuesday, October 30, 2007
Jon Kleinberg CS lecture at U of T
Right now is the lecture by Jon Kleinberg, Department of Computer Science, Cornell University which is on Challenges in Mining Social Network Data: Processes, Privacy and Paradoxes. He has generated seminal results in social networks and document retrieval. I've read his research work on the HITS algorithm which uses hubs and authorities in order to classify search, and from which Google's PageRank is somewhat related to. I've never heard Jon speak so I'm very glad to hear him speak.
He will also speak tonight to kick off the Social Networking Week at U of T. What can computer science contribute to social networks? Today there is a convergence of social and technological networks, computing and information systems with intrinsic social structure. Social network data is a very active area in sociology, social psychology and anthropology. So what can the different fields learn from each other (sociology, social psychology, anthropology from computer science)? This is the research area which I am also part of as well, and it's an exciting research area in my opinion with the emergence of social networking sites like Facebook and MySpace. Mining social networks has a long history in social sciences eg. with Wayne Zachary's PhD work on the university karate club, observing social ties and rivalries. Split in the network could be explained by the minimum cut in the social network.
Social network data spans many orders of magnitude. For example there were 240 million nodes of all IM communication over one month on Microsoft Instant Messenger (Leskovec-Horvitz '07), 4.4 million nodes of declared friendships on blogging community LiveJournal (Liben-Nowell et al., 2005). How can we find the point where the lines of research in large scale and small scale networks converge? In social networks, we can find behaviours of diffusion in social networks that cascade from node to node like an epidemic, which is identified by radial structures in the graph. There have been empirical studies of diffusion in the social sciences like the spread of new agricultural and medical practices (Coleman et al., 1966). The diffusion curves are based on the probability of adopting new behaviour which depends on number of friends who have adopted (Bass 1969, Granovetter 1978 and Schelling 1978). All of the diffusion curves seem to have diminishing returns property, for example in editing a Wikipedia article (Cosley et al., 2007) and joining a LiveJournal community (Backstrom et al., 2006).
These results can then be used for general prediction. Given a network and v's position in it at t1, estimate the probability v will join a given group by t2. Kleinberg has formulated this as a probability estimation problem (Backstrom-Huttenlocher-Kleinberg-Lan 2006). Do disconnected friends or connected friends make joining more likely? Disconnected friends provide an informational advantage but connected friends provide safety/trust advantages. For example in LiveJournal, joining probability increases significantly with more connections among friends in the group (in otherwise friends that are within a clique than not).
If connectedness among friends promotes joining, do highly "clustered" groups grow more quickly? Kleinberg defines clustering to be # of triangles / # of open triads and you can determine community by examining the growth from t1 to t2 as a function of clustering. Leskovec, McGlohon, Faloutsos, Glance and Hurst (2007) have looked into the diffusion of topics in networks of news media and bloggers which shows cascading behaviour. Leskovec, Adamic and Huberman (2006) describe how incentives can be used to propagate interesting recommendations along social network links. How to push questions to people within the social network? (Kleinberg, Raghavan, 2005)
One of the most important questions in mining social network data is how to protect privacy in the dataset. There has been some research where anonymizing data actually caused problems from using on-line pseudonyms and using search engine query logs. If you are part of a small network and based on connectivity, you may be able to find yourself, so anonymization doesn't help. An attacker can attack an anonymized network by being part of the system. Kleinberg has done some work on this by creating a template (Backstrom, Dwork, Kleinberg, 2007). The idea is an attacker creates a small network of nodes through creating accounts called subgraph H and attach them to targeted nodes in the original network. From Ramsey theory, in a random n-node graph, H is unique.
Take home message: how do we build deeper models of the processes at work inside large-scale social networks? How do we make data available without compromising privacy?
It was great to finally meet and talk with Jon Kleinberg!
On Technorati: jon kleinberg, social network
He will also speak tonight to kick off the Social Networking Week at U of T. What can computer science contribute to social networks? Today there is a convergence of social and technological networks, computing and information systems with intrinsic social structure. Social network data is a very active area in sociology, social psychology and anthropology. So what can the different fields learn from each other (sociology, social psychology, anthropology from computer science)? This is the research area which I am also part of as well, and it's an exciting research area in my opinion with the emergence of social networking sites like Facebook and MySpace. Mining social networks has a long history in social sciences eg. with Wayne Zachary's PhD work on the university karate club, observing social ties and rivalries. Split in the network could be explained by the minimum cut in the social network.
Social network data spans many orders of magnitude. For example there were 240 million nodes of all IM communication over one month on Microsoft Instant Messenger (Leskovec-Horvitz '07), 4.4 million nodes of declared friendships on blogging community LiveJournal (Liben-Nowell et al., 2005). How can we find the point where the lines of research in large scale and small scale networks converge? In social networks, we can find behaviours of diffusion in social networks that cascade from node to node like an epidemic, which is identified by radial structures in the graph. There have been empirical studies of diffusion in the social sciences like the spread of new agricultural and medical practices (Coleman et al., 1966). The diffusion curves are based on the probability of adopting new behaviour which depends on number of friends who have adopted (Bass 1969, Granovetter 1978 and Schelling 1978). All of the diffusion curves seem to have diminishing returns property, for example in editing a Wikipedia article (Cosley et al., 2007) and joining a LiveJournal community (Backstrom et al., 2006).
These results can then be used for general prediction. Given a network and v's position in it at t1, estimate the probability v will join a given group by t2. Kleinberg has formulated this as a probability estimation problem (Backstrom-Huttenlocher-Kleinberg-Lan 2006). Do disconnected friends or connected friends make joining more likely? Disconnected friends provide an informational advantage but connected friends provide safety/trust advantages. For example in LiveJournal, joining probability increases significantly with more connections among friends in the group (in otherwise friends that are within a clique than not).
If connectedness among friends promotes joining, do highly "clustered" groups grow more quickly? Kleinberg defines clustering to be # of triangles / # of open triads and you can determine community by examining the growth from t1 to t2 as a function of clustering. Leskovec, McGlohon, Faloutsos, Glance and Hurst (2007) have looked into the diffusion of topics in networks of news media and bloggers which shows cascading behaviour. Leskovec, Adamic and Huberman (2006) describe how incentives can be used to propagate interesting recommendations along social network links. How to push questions to people within the social network? (Kleinberg, Raghavan, 2005)
One of the most important questions in mining social network data is how to protect privacy in the dataset. There has been some research where anonymizing data actually caused problems from using on-line pseudonyms and using search engine query logs. If you are part of a small network and based on connectivity, you may be able to find yourself, so anonymization doesn't help. An attacker can attack an anonymized network by being part of the system. Kleinberg has done some work on this by creating a template (Backstrom, Dwork, Kleinberg, 2007). The idea is an attacker creates a small network of nodes through creating accounts called subgraph H and attach them to targeted nodes in the original network. From Ramsey theory, in a random n-node graph, H is unique.
Take home message: how do we build deeper models of the processes at work inside large-scale social networks? How do we make data available without compromising privacy?
It was great to finally meet and talk with Jon Kleinberg!
On Technorati: jon kleinberg, social network
Monday, October 29, 2007
Busy busy week this week
I've got a busy week ahead of me for this week. I'm giving an Ignite presentation about finding subgroups in TorCamp at DemoCampToronto15 tonight at Hart House, then have to finish marking assignments, then finish writing a paper, and then giving a talk at the Social Networking Week at U of T on Friday.
But I enjoy doing this kind of stuff, so I don't mind it. So don't expect too many blog entries this week!
But I enjoy doing this kind of stuff, so I don't mind it. So don't expect too many blog entries this week!
Wednesday, October 24, 2007
Pervasive 2007 Conference trip report in IEEE Pervasive Computing magazine
Just found out that the Pervasive 2007 conference trip report which I helped co-author is now out in the IEEE Pervasive Computing magazine Vol. 6 No. 4 (October - December 2007). You can read the article here.
On Technorati: Pervasive2007
On Technorati: Pervasive2007
Finished my CASCON short paper talk
I just finished my CASCON short paper talk on "Identifying Active Subgroups in Online Communities" about an hour ago. It went well, I had some great feedback. I'll post the talk which I recorded and slides soon. Now, I can concentrate on the rest of my PhD work. No rest for a PhD student! But I enjoy giving talks and meeting with people and discussing about my research, it's exciting and engaging. If you have any comments or feedback from my paper or talk, write some comments on this blog back to me!
Tuesday, October 23, 2007
CASCON 2007 Conference, Day 1
I just finished co-chairing a session on Tagging as a Social Contract along with Mark Chignell and Sara Darvish, where we had 4 talks about issues surrounding tagging in a business environment. This was the second session as part of the Second Working Conference on Social Computing and Business at the CASCON 2007 conference. It was a great workshop and great session, and great discussion. I talked about how community can be inferred from tagging using the YouTube vaccination videos as an example. Podcasts and slides from the workshop should be available, so check the CASCON blog.
I also showed a demo of a community-based web portal that our lab created to support vaccination groups in our exhibit at the Technology Showcase called "Video Web 2.0: Collaborative Tagging in Web Video". If you're at CASCON, check it out!
Photos from CASCON are on my Flickr account.
Friday, October 19, 2007
Google hints at social network plan
I've been wondering when Google would start to think about social networking and how it could be used. So far, Yahoo has been the leader in social networking, with Flickr and Upcoming and Yahoo My Web Beta. Google's Orkut still does not compare in the same calibre with Facebook or MySpace, huge social networking web sites. However, Google is now beginning to hint how they will use social networking data in their own web search and to share the social data with others. According to Eric Schmidt, Google's CEO, from this article from NY Times, Google has something up its sleeve.
Will Google be able to top Yahoo and other social networking sites like Facebook? Only time will tell.
Will Google be able to top Yahoo and other social networking sites like Facebook? Only time will tell.
Wednesday, October 17, 2007
Passed the thesis proposal!
I just did my thesis proposal today and passed! Just need to make changes and do some more analysis and I should be hopefully done before April of next year.
Saturday, October 13, 2007
Thesis proposal and busy rest of October!
I just finished writing the thesis proposal which I will send to my committee, because I will have a meeting with them on Wednesday. Hopefully, everything goes well and if everything goes according to plan, I can finish the dissertation and defence by end of this year!
It's going to be a busy rest of the October. I'm co-chairing a workshop at CASCON called Tagging as a Social Contract on Monday, October 22. If you're going to be at CASCON, sign up for this workshop, it promises to be an interesting one. For more information, check the CASCON blog. After that, I'm going to be presenting my paper at the CASCON conference on Wednesday, October 24 called "Identifying Active Subgroups in Online Communities". And then, I will be giving a talk at the Social Networking Symposium at U of T on Friday, November 2nd from 9:50 to 10:15 am called "Structural Analysis of Social Hypertext for Finding Sense of Community" right after my supervisor talks.
It's going to be a busy rest of the October. I'm co-chairing a workshop at CASCON called Tagging as a Social Contract on Monday, October 22. If you're going to be at CASCON, sign up for this workshop, it promises to be an interesting one. For more information, check the CASCON blog. After that, I'm going to be presenting my paper at the CASCON conference on Wednesday, October 24 called "Identifying Active Subgroups in Online Communities". And then, I will be giving a talk at the Social Networking Symposium at U of T on Friday, November 2nd from 9:50 to 10:15 am called "Structural Analysis of Social Hypertext for Finding Sense of Community" right after my supervisor talks.
Monday, October 08, 2007
Married!
Yes, I just got married about a week ago, the wedding was great and the weather was just perfect. Couldn't have asked for a better day. Thanks to everyone who helped out in the wedding and for those that attended. My wife and I were so happy to see you there, and even though it was a tiring day, we thoroughly enjoyed it and will treasure this for the rest of our lives.
I'm very thankful this Thanksgiving for such a beautiful, amazing and considerate wife. And marriage life feels so great, I wouldn't trade it for anything else.
For those that are interested, I'll post wedding photos online soon.
I'm very thankful this Thanksgiving for such a beautiful, amazing and considerate wife. And marriage life feels so great, I wouldn't trade it for anything else.
For those that are interested, I'll post wedding photos online soon.
Subscribe to:
Posts (Atom)