Session (web analytics)

In web analytics , a session , or visit is a unit of measurement of a user’s actions taken within a period of time or with regard to completion of a task. Sessions are also used in operational analytics and provision of user-specific recommendations . There are two primary methods used to define a session time-oriented approaches based is continuity in user activity and navigation-based approaches based continuity is in a chain of requested pages.

Definition

The definition of “session” varies, when applied to search engines . [1] Generally, a session is understood to consist of “a sequence of requests made by a single end-user during a visit to a particular site”. [2] In the context of search engines , “sessions” and “query sessions” have at least two definitions. [1] A session or query session may be made by a user in a particular time period [3] or it may also be a series of queries or navigations with a consistent underlying user need. [4] [5]

Uses

Per user sessions can be used as a measure of website use. [6] [7] Other metrics used in research and applied to web analytics include session length, [8] and user actions per session. [9] Session length is seen as a more accurate alternative to measuring page views . [10]

Reconstructed sessions have been used to measure total user input, including to measure the number of hours taken to construct Wikipedia . [11] Sessions are also used for operational analytics, data anonymization , identifying networking anomalies , and synthetic workload generation for testing servers with artificial traffic. [12] [13]

Session reconstruction

Essential to the use of sessions in web analytics is being able to identify them. This is known as “session reconstruction”. Approaches to session reconstruction can be divided into two main categories: time-oriented, and navigation-oriented. [14]

Time-oriented approaches

Time-oriented approaches to session rebuilding a user’s period of inactivity commonly called an “inactivity threshold.” Once this period of inactivity is reached, the user is assumed to have left the site completely closed. Further requests from the same user are considered at the second session. A common value for the inactivity threshold is 30 minutes and sometimes described as the standard industry. [15] [16] Some 30-minute artifacts are naturally occurring over long periods of time and have experienced with other thresholds. [17] [18] Others simply state: “no time threshold is effective at identifying [sessions]”. [19]

One alternative that has been proposed using user-specific thresholds rather than a single, global threshold for the entire dataset. [20] [21] This is the problem of assuming that the thresholds follow a bimodal distribution , and is not suitable for datasets that cover a long period of time. [17]

Navigation-oriented approaches

Navigation-oriented approaches exploit the structure of websites – specifically, the presence of hyperlinks and the tendency of users to navigate through the pages of the site. [14] One way of identifying sessions by: http://www.youtube.com/watch of the previously-accessed pages. This will take backtracking, where a user will retrace their steps before opening a new page. [22] A simpler approach, which does not take backtracking into account, is to simply require that the HTTP referrerof each request be a page that is already in the session. If it is not, a new session is created. [23] This class of heuristics “exhibits very poor performance” on websites that contain framesets . [24]

References

  1. ^ Jump up to:b Gayo-Avello 2009 , p. 1824.
  2. Jump up^ Arlitt 2000, p. 2.
  3. Jump up^ Donato 2010, p. 324.
  4. Jump up^ Gayo-Avello 2009, p. 1825.
  5. Jump up^ Lam 2007, p. 147.
  6. Jump up^ Weischdel 2006, p. 464.
  7. Jump up^ Catledge 1995, p. 5.
  8. Jump up^ Jansen 2006, p. 10.
  9. Jump up^ Jansen 2000, p. 12.
  10. Jump up^ Khoo 2008, p. 377.
  11. Jump up^ Geiger 2014, p. 1.
  12. Jump up^ Meiss 2009, p. 177.
  13. Jump up^ Arlitt 2000, p. 8.
  14. ^ Jump up to:b Spiliopoulou 2003 , p. 176.
  15. Jump up^ Ortega 2010, p. 332.
  16. Jump up^ Eickhoff 2014, p. 3.
  17. ^ Jump up to:b Mehrzadi 2012 , p. 3.
  18. Jump up^ He 2002, p. 733.
  19. Jump up^ Jones 2008, p. 2.
  20. Jump up^ Murray 2006, p. 3.
  21. Jump up^ Mehrzadi 2012, p. 1.
  22. Jump up^ Cooley 1999, p. 19.
  23. Jump up^ Cooley 1999, p. 23.
  24. Jump up^ Berendt 2003, p. 179.