Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Dear Antoine,
Please consider this pull request and implement a teleport parameter explicitly in the Sun object.
Some discussion for context:
Antoine: I think your randomisations may induce a different 'interventional' null distribution than one we can obtain via a family of teleport parameters in [0,1]. It is worth studying.For each teleport parameter, in the personalised pagerank (with perhaps something like the normalised or observed correlation with/without the current randomisation in rooster object), we should get aggressiveness and sensitivity measures, I suppose.
I think it may be worthwhile implementing this pagerank in the sun object with a user-specified teleport parameter.
We could then see the effect of teleport parameter with and without the initial randomisation. I am not sure if the current randomisation to obtain 'normalised correlations' and then using these correlations as edge-weights in a directed graph of concept-ids (events) to further run personalised page rank to get measures of aggresiveness and sensitivity is obviously equivalent to obtaining observed correlations and then running personalised pagerank for each teleport parameter in a set say {0/10, 1/10, 2/10, ..., 10/10}.But, I believe that the teleport parameter as it approaches 1 does capture a natural homotopy from pure structure in the directed correlation-weighted graph over events to pure randomness in the complete graph over the same set of events. Moreover, it is a definite generalisation of the current implementation and is worth investigation by Albin, Liam and myself, if you have the time to implement it in the latest version of pathogen directly in: https://github.com/aamend/pathogen after accepting our PR just made.