Uncover HR attributes which best represent a population for a Person to Person query

Returns a data frame that gives a percentage of the group combinations that best represent the population provided. Uses a person to person query. This is used internally within network_p2p().

network_describe(
  data,
  hrvar = c("Organization", "LevelDesignation", "FunctionType")
)

Arguments

data

Data frame containing a vertex table output from network_p2p().

hrvar

Character vector of length 3 containing the HR attributes to be used. Defaults to c("Organization", "LevelDesignation", "FunctionType").

Value

data frame. A summary table giving the percentage of group combinations that best represent the provided data.

See also

Other Network: external_network_plot(), g2g_data, internal_network_plot(), network_g2g(), network_leiden(), network_louvain(), network_p2p(), network_summary(), p2p_data_sim()

Author

Tannaz Sattari Tabrizi Tannaz.Sattari@microsoft.com

Examples

# Simulate a P2P edge list
sim_data <- p2p_data_sim()

# Perform Louvain Community Detection and return vertices
lc_df <-
  sim_data %>%
  network_p2p(
    display = "louvain",
    return = "data"
  )

# Join org data from input edge list
joined_df <-
  lc_df %>%
  dplyr::left_join(
    sim_data %>%
      dplyr::select(TieOrigin_PersonId,
                    TieOrigin_Organization,
                    TieOrigin_LevelDesignation,
                    TieOrigin_City),
    by = c("name" = "TieOrigin_PersonId"))

# Describe cluster 2
joined_df %>%
  # dplyr::filter(cluster == "2") %>%
  network_describe(
    hrvar = c(
      "Organization",
      "LevelDesignation",
      "City"
    )
  ) %>%
  dplyr::glimpse()
#> Rows: 679
#> Columns: 7
#> $ feature_1       <chr> "TieOrigin_Organization", "TieOrigin_Organization", "T…
#> $ feature_1_value <chr> "Org F", "Org A", "Org C", "Org D", "Org B", "Org E", …
#> $ feature_2       <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
#> $ feature_2_value <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
#> $ feature_3       <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
#> $ feature_3_value <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA…
#> $ Percentage      <dbl> 0.22077922, 0.20779221, 0.19480519, 0.11688312, 0.1038…