Weight Pair-wise Cumulative Temporal Projection Scheme
Source:R/graph.R
weight_scheme_pairwise_cum_temporal.Rd
This weight scheme sums the deleted node adjacent edges when re-wired
in the projection graph that *occur before* the temporal edge.
Differently than weight_scheme_cum_temporal
, however,
the weight value used for every edge is of both developers to the deleted node,
which causes some duplication.
Arguments
- temporally_ordered_projected_graph
A temporally row-wise ordered table semi-processed bipartite projection network resulting from
temporal_graph_projection
when specifying weight_scheme_function = NA and lag = all_lag.
Details
Refer to the unit tests to see a concrete example.
Note this function assumes the rows of `temporally_ordered_projected_graph`
are temporally ordered. This property is only guaranteed by the
temporal_graph_projection
function if `lag = "all_lag"`. For
`lag=one_lag`, use weight_scheme_sum_edges
or
weight_scheme_count_deleted_nodes
.
Refer to this weight scheme unit tests for examples.
References
M. Joblin, W. Mauerer, S. Apel, J. Siegmund and D. Riehle, "From Developer Networks to Verified Communities: A Fine-Grained Approach," 2015 IEEE/ACM 37th IEEE International Conference on Software Engineering, Florence, 2015, pp. 563-573, doi: 10.1109/ICSE.2015.73.
See also
Other weight_scheme:
weight_scheme_count_deleted_nodes()
,
weight_scheme_cum_temporal()