Print

Print


*****  To join INSNA, visit http://www.insna.org  *****

Gloria-

   In all likelihood, the canonical resource for longitudinal network
analysis is that of Tom Snijders at Oxford (
https://www.stats.ox.ac.uk/~snijders/siena/) as well as his students such
as Christian Stadtfeld now at ETHZ (
https://www.uni-konstanz.de/universitaet/aktuelles-und-medien/oeffentliche-veranstaltungen/veranstaltungen/veranstaltungsdetails/2017/1/31/event/23366-Christoph-Stadtfeld-Zric/tx_cal_phpicalendar/
).

  I hesitate in calling this work 'seminal,' although it is in the context
of network literature, because it's strongly related to earlier work done
in psychometrics on multimode principal components such as PARAFAC (e.g.,
this tutorial,
https://www.cs.cmu.edu/~pmuthuku/mlsp_page/lectures/Parafac.pdf), Pieter
Kroonenberg's book *Three-mode Principal Component Analysis* and his 'Three
Mode Company' (http://three-mode.leidenuniv.nl/), Ledyard Tucker's original
decompositions of the data cube (
https://en.wikipedia.org/wiki/Tucker_decomposition) as well as the work of
many scholars in longitudinal multidimensional scaling and the methods they
developed to decompose matrices enabling dynamic analysis (e.g.,
https://en.wikipedia.org/wiki/Multidimensional_scaling) including J.D.
Carroll, William Kruskal, Wayne DeSarbo or Forrest Tucker's ALSCAL -- the
original alternating-least squares algorithm which continues to see wide
use today by machine learning enthusiasts.

Thomas Ball

On Fri, Jul 21, 2017 at 8:24 AM, Gloria Alvarez <[log in to unmask]> wrote:

> ***** To join INSNA, visit http://www.insna.org *****
>
> Hi,
>
> I am researching about longitudinal social networks and some questions
> came up
>
> 1) In terms of epistemology, which is the difference between "dynamic
> social network" and "longitudinal social network". Does it come from the
> approach? ie. Longitudinal looks associated with time-based approach and
> dynamic with "actor-based" models? From different terminology across
> different disciplines? others?
>
> 2) Is there any common framework that synthesizes all different
> longitudinal/dynamic models/approaches? I found some in Wasserman & Faust,
> McCulloh & Carley, Holme & Saramäki,...  but I have problems to located
> some models in those classifications and I also find difficult the
> differences between disciplines approach (social science, physics, computer
> science...).
>
> Thanks very much in advance for your support and have a nice weekend,
>
> Gloria
> _____________________________________________________________________
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send an email
> message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET
> in the body of the message.

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.