Loading ...
Sorry, an error occurred while loading the content.
 

Link Analysis versus Social Network Analysis

Expand Messages
  • philrjones
    We re having a healthy debate at work about the differences between Link Analysis and SNA. I d be interested in hearing the views of ONA-PRAC members.
    Message 1 of 13 , Nov 10, 2007
      We're having a healthy debate at work about the differences
      between "Link Analysis" and SNA.

      I'd be interested in hearing the views of ONA-PRAC members.

      Regards,

      Phil
    • Cai Kjaer
      Hi Phil, Can you expand just a little? What exactly do you mean by link analysis? I noticed this definition in Wikipedia - is this what you mean? * Link
      Message 2 of 13 , Nov 11, 2007
        Hi Phil,
         
        Can you expand just a little? What exactly do you mean by link analysis? I noticed this definition in Wikipedia - is this what you mean?
         
        "Link analysis is a subset of network analysis, exploring associations between objects. An example may be examining the addresses of suspects and victims, the telephone numbers they have dialed and financial transactions that they have partaken in during a given timeframe, and the familial relationships between these subjects as a part of police investigation. Link analysis here provides the crucial relationships and associations between very many objects of different types that are not apparent from isolated pieces of information."
         
        All my work has been around people more so than 'objects' - with the exception of mapping people's interaction with information sources. I believe Netmap Analytics here in Sydney do this sort of work, but not sure if they are on this list... ( http://www.netmap.com.au/)


        Kind regards

        Cai

        ________________________
        Cai Kjaer, Partner
        Optimice Pty. Ltd.
        Phone: +61 411 569 694
        Email: cai.kjaer@...
        Web: www.optimice.com.au
      • rg244@bath.ac.uk
        Phil and Cai I too am intrigued at the difference between Social Network Analysis and Link Analysis. SNA uses links to represent the relationships between
        Message 3 of 13 , Nov 11, 2007
          Phil and Cai

          I too am intrigued at the difference between Social Network Analysis
          and Link Analysis. SNA uses links to represent the relationships
          between people and people, people and objects and objects and objects
          (and a lot more). I have just looked at the Oz site Phil mentions, and
          based on their demonstration (the Kevin Bacon thing and 6 degrees of
          freedom) I can’t see Netmap Analytics’ unique selling point. They seem
          to be talking about SNA, and the graphical representations we can use
          to show it.

          Re the definition of link analysis from Wiki, this seems to be basic
          SNA to me.

          Am I missing something?

          Roy Greenhalgh




          Quoting Cai Kjaer <cai.kjaer@...>:

          > Hi Phil,
          >
          > Can you expand just a little? What exactly do you mean by link analysis? I
          > noticed this definition in Wikipedia - is this what you mean?
          >
          > *"Link analysis* is a subset of network analysis, exploring associations
          > between objects. An example may be examining the addresses of suspects and
          > victims, the telephone numbers they have dialed and financial transactions
          > that they have partaken in during a given timeframe, and the familial
          > relationships between these subjects as a part of police investigation. Link
          > analysis here provides the crucial relationships and associations between
          > very many objects of different types that are not apparent from isolated
          > pieces of information."
          >
          > All my work has been around people more so than 'objects' - with the
          > exception of mapping people's interaction with information sources. I
          > believe Netmap Analytics here in Sydney do this sort of work, but not sure
          > if they are on this list... (http://www.netmap.com.au/)
          >
          >
          > Kind regards
          >
          > Cai
          >
          > ________________________
          > Cai Kjaer, Partner
          > Optimice Pty. Ltd.
          > Phone: +61 411 569 694
          > Email: cai.kjaer@...
          > Web: www.optimice.com.au
          >
        • Valdis Krebs
          I encounter link analysis when working with law enforcement and counterterrorism clients. Links analysis, from my outsider perspective, is linking together
          Message 4 of 13 , Nov 11, 2007
            I encounter "link analysis" when working with law enforcement and
            counterterrorism clients.

            Links analysis, from my outsider perspective, is linking together
            various objects for evidence mapping. These objects include people,
            cars, houses, businesses, phone numbers, other objects of ownership.
            It is felt that such analysis can uncover cells and gangs and other
            criminal conspiracies -- besides providing nice visuals for court.
            They do not use social network metrics but do a elementary form
            degree and cluster analysis, and have a good understanding of
            structural equivalence. The main tool in play for link analysis is
            i2's Analyst Workbook -- which has a strangle hold on that market.

            http://www.i2inc.com/Products/Analysts_Notebook/

            Valdis


            On Nov 10, 2007, at 6:22 PM, philrjones wrote:

            > We're having a healthy debate at work about the differences
            > between "Link Analysis" and SNA.
          • philrjones
            As Valdis mentions, link analysis is a term used by law enforcement and Intelligence agencies. At its most basic, people use tools such as i2 Analyst s
            Message 5 of 13 , Nov 11, 2007
              As Valdis mentions, link analysis is a term used by law enforcement
              and Intelligence agencies. At its most basic, people use tools such
              as i2 Analyst's notebook to visualise relationships between people,
              things and events - using visualisation as a means for conveying
              their 'mental map' of the network under consideration.

              More advanced link analysis involves systematic node and link
              extraction from source material and potentially look at a range of
              inter-personal relationship types. At this point the boundary between
              link analysis and SNA seems to get blurred.

              My working hypothesis for what makes SNA distinctive from link
              analysis is:

              * SNA makes use of social theory and the social context of the
              network under consideration to underpin the range of link definitions
              and subsequent analysis (though my impression is that SNA is often
              focused on production of graph theoretic metrics - which one could
              say is simply Network Analysis, not SNA).

              * There is some fusion of evidence (e.g. multiple reports of a
              relationship between person A and B)to provide an assessment of the
              link, prior to undertaking SNA.

              * It makes use of science (social and mathematical) principles to:
              improve the objectivity of analysis / visualisation; and potentially
              to undertake link and node inference where missing data is an issue.

              Do others agree ... what have I missed.

              Regards,

              Phil
            • philrjones
              For that matter what makes SNA and / or ONA distinctive from Network Analysis? Is it purely to do with the goal of the analysis - insight into social systems
              Message 6 of 13 , Nov 11, 2007
                For that matter what makes SNA and / or ONA distinctive from Network
                Analysis? Is it purely to do with the goal of the analysis - insight
                into social systems and organisations, or are there more fundamental
                attributes?

                Phil
              • philrjones
                Not sure where my reply went ... I ll have another go, sorry if it subsequently appears! As Valdis mentions, link analysis is used by law enforcement and
                Message 7 of 13 , Nov 11, 2007
                  Not sure where my reply went ... I'll have another go, sorry if it
                  subsequently appears!

                  As Valdis mentions, link analysis is used by law enforcement and
                  intelligence agencies to look at the relationships between people,
                  things and events.

                  At its most basic, people use tools like i2 Analyst's Notebook to
                  visualise their mental map of a network. More advanced link analysis
                  will systematically identify nodes and links from source material and
                  may use a range of link types to represent different inter-personal
                  link types. At this point the boundary between link analysis and SNA
                  gets blurred.

                  What I think makes SNA distinctive from link analysis is that:

                  * SNA makes use of social constructs, theory and context to define
                  the link types that will be used.

                  * Some fusion of information takes place, prior to undertaking
                  quantitative analysis (e.g. to resolve multiple pieces of evidence
                  about links between person A and person B).

                  * SNA makes use of social and mathematical (mainly graph theory)
                  science to support both analysis and visualisation of data.


                  Using these definitions, I wonder how much SNA is, in reality little
                  more than network analysis ?

                  Phil
                • Valdis Krebs
                  As I think I posted here last year, or maybe Patti did, ONA was created by Gerry Falkowski, when his IBM clients [both internal and external, squirmed at the
                  Message 8 of 13 , Nov 11, 2007
                    As I think I posted here last year, or maybe Patti did, ONA was
                    created by Gerry Falkowski, when his IBM clients [both internal and
                    external, squirmed at the word "social" -- "we're business people,
                    what does that have to do with social?" was one of the infamous
                    quotes. So he changed the wording -- and nothing else -- and started
                    lining up ONA clients left and right. This was back in the mid
                    1990s. I remember the conversation when he told me... pretty funny.

                    When I took grad courses @ Mich State from the Comm Dept in network
                    analysis -- it was called just that -- "network analysis", once in a
                    while you heard "communication network analysis". But plain network
                    analysis soon got associated with computer networks, so people
                    started adding a descriptor up front.

                    Valdis


                    On Nov 11, 2007, at 3:44 PM, philrjones wrote:

                    > For that matter what makes SNA and / or ONA distinctive from Network
                    > Analysis? Is it purely to do with the goal of the analysis - insight
                    > into social systems and organisations, or are there more fundamental
                    > attributes?
                    >
                    > Phil
                    >
                    >
                    >
                    >
                    >
                    >
                    >
                    >
                    > Yahoo! Groups Links
                    >
                    >
                    >
                  • Patti Anklam
                    I don t think I knew that Gerry story, but I like it. We named this group ONA-PRAC so that we could distinguish ourselves from the greater world of social
                    Message 9 of 13 , Nov 11, 2007

                      I don’t think I knew that Gerry story, but I like it.  We named this group ONA-PRAC so that we could distinguish ourselves from the greater world of social network analysis – which does deal with epidemiology and other sociological applications. When I make this distinction in presentations, I emphasize that ONA is a subset of the larger practice of social network analysis – we want to look at people in organizations (mostly business organizations). The tools and the methods are essentially the same, but as Valdis says, the word “social” carries a lot of connotations so ONA sometimes works better.

                       


                      From: ona-prac@yahoogroups.com [mailto: ona-prac@yahoogroups.com ] On Behalf Of Valdis Krebs
                      Sent: Sunday, November 11, 2007 5:13 PM
                      To: ona-prac@yahoogroups.com
                      Subject: Re: [ona-prac] Re: Link Analysis versus Social Network Analysis

                       

                      As I think I posted here last year, or maybe Patti did, ONA was
                      created by Gerry Falkowski, when his IBM clients [both internal and
                      external, squirmed at the word "social" -- "we're business people,
                      what does that have to do with social?" was one of the infamous
                      quotes. So he changed the wording -- and nothing else -- and started
                      lining up ONA clients left and right. This was back in the mid
                      1990s. I remember the conversation when he told me... pretty funny.

                      When I took grad courses @ Mich State from the Comm Dept in network
                      analysis -- it was called just that -- "network analysis", once in a
                      while you heard "communication network analysis". But plain network
                      analysis soon got associated with computer networks, so people
                      started adding a descriptor up front.

                      Valdis

                      On Nov 11, 2007, at 3:44 PM, philrjones wrote:

                      > For that matter what makes SNA and / or ONA distinctive from Network
                      > Analysis? Is it purely to do with the goal of the analysis - insight
                      > into social systems and organisations, or are there more fundamental
                      > attributes?
                      >
                      > Phil
                      >
                      >
                      >
                      >
                      >
                      >
                      >
                      >
                      > Yahoo! Groups Links
                      >
                      >
                      >

                    • Bordeaux, John
                      Perhaps I m just being cynical, but some usage of the link analysis term stems apparently from the unfortunate privacy associations that are made when
                      Message 10 of 13 , Nov 11, 2007
                        Perhaps I'm just being cynical, but some usage of the "link analysis" term stems apparently from the unfortunate privacy associations that are made when someone in government says "data mining." If I'm right, it's only a matter of time before "link analysis" becomes tainted as well.

                        -----Original Message-----
                        From: ona-prac@yahoogroups.com on behalf of Patti Anklam
                        Sent: Sun 11/11/2007 6:04 PM
                        To: ona-prac@yahoogroups.com
                        Subject: RE: [ona-prac] Re: Link Analysis versus Social Network Analysis

                        I don't think I knew that Gerry story, but I like it. We named this group ONA-PRAC so that we could distinguish ourselves from the greater world of social network analysis - which does deal with epidemiology and other sociological applications. When I make this distinction in presentations, I emphasize that ONA is a subset of the larger practice of social network analysis - we want to look at people in organizations (mostly business organizations). The tools and the methods are essentially the same, but as Valdis says, the word "social" carries a lot of connotations so ONA sometimes works better.



                        ________________________________

                        From: ona-prac@yahoogroups.com [mailto:ona-prac@yahoogroups.com] On Behalf Of Valdis Krebs
                        Sent: Sunday, November 11, 2007 5:13 PM
                        To: ona-prac@yahoogroups.com
                        Subject: Re: [ona-prac] Re: Link Analysis versus Social Network Analysis



                        As I think I posted here last year, or maybe Patti did, ONA was
                        created by Gerry Falkowski, when his IBM clients [both internal and
                        external, squirmed at the word "social" -- "we're business people,
                        what does that have to do with social?" was one of the infamous
                        quotes. So he changed the wording -- and nothing else -- and started
                        lining up ONA clients left and right. This was back in the mid
                        1990s. I remember the conversation when he told me... pretty funny.

                        When I took grad courses @ Mich State from the Comm Dept in network
                        analysis -- it was called just that -- "network analysis", once in a
                        while you heard "communication network analysis". But plain network
                        analysis soon got associated with computer networks, so people
                        started adding a descriptor up front.

                        Valdis

                        On Nov 11, 2007, at 3:44 PM, philrjones wrote:

                        > For that matter what makes SNA and / or ONA distinctive from Network
                        > Analysis? Is it purely to do with the goal of the analysis - insight
                        > into social systems and organisations, or are there more fundamental
                        > attributes?
                        >
                        > Phil
                        >
                        >
                        >
                        >
                        >
                        >
                        >
                        >
                        > Yahoo! Groups Links
                        >
                        >
                        >
                      • Valdis Krebs
                        A little Monday humor... in light of our recent discussion on the terms which include network analysis & link analysis you folks may find this funny. My
                        Message 11 of 13 , Nov 12, 2007
                          A little Monday humor... in light of our recent discussion on the terms
                          which include "network analysis" & "link analysis" you folks may find
                          this funny. My blog post about it is here...

                          http://www.networkweaving.com/blog

                          Valdis
                        • Tim Keller
                          As a term of art LA is at least a decade old, probably more like 15 years. I ve seen a paper out of the Intelligence Community from the mid-90s that uses it.
                          Message 12 of 13 , Nov 19, 2007
                            As a term of art LA is at least a decade old, probably more like 15
                            years. I've seen a paper out of the Intelligence Community from the
                            mid-90s that uses it. I think it just came about to distinguish
                            itself from other types of data mining, no conspiracies needed.

                            --- In ona-prac@yahoogroups.com, "Bordeaux, John" <John_Bordeaux@...>
                            wrote:
                            >
                            > Perhaps I'm just being cynical, but some usage of the "link
                            analysis" term stems apparently from the unfortunate privacy
                            associations that are made when someone in government says "data
                            mining." If I'm right, it's only a matter of time before "link
                            analysis" becomes tainted as well.
                            >
                            > -----Original Message-----
                            > From: ona-prac@yahoogroups.com on behalf of Patti Anklam
                            > Sent: Sun 11/11/2007 6:04 PM
                            > To: ona-prac@yahoogroups.com
                            > Subject: RE: [ona-prac] Re: Link Analysis versus Social Network Analysis
                            >
                            > I don't think I knew that Gerry story, but I like it. We named this
                            group ONA-PRAC so that we could distinguish ourselves from the greater
                            world of social network analysis - which does deal with epidemiology
                            and other sociological applications. When I make this distinction in
                            presentations, I emphasize that ONA is a subset of the larger practice
                            of social network analysis - we want to look at people in
                            organizations (mostly business organizations). The tools and the
                            methods are essentially the same, but as Valdis says, the word
                            "social" carries a lot of connotations so ONA sometimes works better.
                            >
                            >
                            >
                            > ________________________________
                            >
                            > From: ona-prac@yahoogroups.com [mailto:ona-prac@yahoogroups.com] On
                            Behalf Of Valdis Krebs
                            > Sent: Sunday, November 11, 2007 5:13 PM
                            > To: ona-prac@yahoogroups.com
                            > Subject: Re: [ona-prac] Re: Link Analysis versus Social Network Analysis
                            >
                            >
                            >
                            > As I think I posted here last year, or maybe Patti did, ONA was
                            > created by Gerry Falkowski, when his IBM clients [both internal and
                            > external, squirmed at the word "social" -- "we're business people,
                            > what does that have to do with social?" was one of the infamous
                            > quotes. So he changed the wording -- and nothing else -- and started
                            > lining up ONA clients left and right. This was back in the mid
                            > 1990s. I remember the conversation when he told me... pretty funny.
                            >
                            > When I took grad courses @ Mich State from the Comm Dept in network
                            > analysis -- it was called just that -- "network analysis", once in a
                            > while you heard "communication network analysis". But plain network
                            > analysis soon got associated with computer networks, so people
                            > started adding a descriptor up front.
                            >
                            > Valdis
                            >
                            > On Nov 11, 2007, at 3:44 PM, philrjones wrote:
                            >
                            > > For that matter what makes SNA and / or ONA distinctive from Network
                            > > Analysis? Is it purely to do with the goal of the analysis - insight
                            > > into social systems and organisations, or are there more fundamental
                            > > attributes?
                            > >
                            > > Phil
                            > >
                            > >
                            > >
                            > >
                            > >
                            > >
                            > >
                            > >
                            > > Yahoo! Groups Links
                            > >
                            > >
                            > >
                            >
                          • Sérgio Storch
                            Dear friends We´d like to have one of you who lives in Europe to tell us about best practices in utilization of social networks in government and public
                            Message 13 of 13 , Nov 21, 2007
                              Dear friends
                              We´d like to have one of you who lives in Europe to tell us about best practices in utilization of social networks in government and public services, in Dec 12 or 13, when the 6th Brazilian Conference on KM for Government will take place.
                              Would those who are interested: please rush and send a short profile on the work you´ve been doing, to me and to paulo.fresneda@.... Because of the funding available, a mandatory requirement is that you live in Europe.

                              Regards
                              --
                              Sérgio Storch
                              sergiostorch@...
                              fixo: 11-3666.9005
                              cel : 11-8644.0721 (novo!)
                              skype: sstorch

                              "Eu não sei quem descobriu a água, mas certamente não foi um peixe" (Marshall MacLuhan)

                            Your message has been successfully submitted and would be delivered to recipients shortly.