LearnSNATeam solutions

T.R.O.L – Temporally Recurrent Optimal Learning – Case Telenor SNA

7
votes

 

Article – SNA-Telenor – Team T.R.O.L. 1.Business Understanding, 2.Data Understanding, 3. Data Preparation

TROL-0002 4. Modeling

T.R.O.L-0003,4.Modeling, 5. Evaluation

T.R.O.L-0004, 4.Modeling, 5. Evaluation

T.R.O.L – 6. Deployment

 

 

 

 

#Case_Telenor  #SNA

Microsoft Azure

2 vauchers

W62UL1PZZRFHVCBH9R

WWZO8QDZDKJK8VG7QB

 

Share this

4 thoughts on “T.R.O.L – Temporally Recurrent Optimal Learning – Case Telenor SNA

  1. 1
    votes

    Very well written. I like even humor parts :), but I would point to one segment before pushing final article:
    “This means that most of the people has called as many people as they were called by. After a short research it appears that some experts consider such behaviour in communicational SNA-s for “normal”. (Example for such research https://www.aaai.org/ocs/index.php/ICWSM/ICWSM11/paper/viewFile/2812/3224
    Visual proof can be seen in the diagram above ”
    Visual proof, e.g. diagram is missing from article.

  2. 0
    votes

    Questions during your presentation:
    1. @va1io
    For the Closeness between an outside node and Group 5 – why you don’t use the connection strength also?
    2. @drceenish
    Do you think its a starting of new kind of social network?

  3. 1
    votes

    I see you have tried to form Cliques… But did you decide final version of all Cliques and the total number of formed Social Groups and all members per Social Group (Clique). leaders should be defined per Group (Clique) not for the network in general…. It is very important for Viral communication or risk score. Then calculate all metrics as density or degrees Per group, not for the network in general. I am sure this will give you complacently different results. 😉

Leave a Reply