R-从单个列中选择满足多个条件的ID

Question

I have a list of IDs, each having multiple events. The data looks like an event log, ie one event per ID per row. For example:

n.ID=4
n.events=5
set.seed(1234)
df <- setNames(melt(replicate(n.ID, sort(sample(letters[c(1:10)], n.events))))[c(2:3)], c("ID", "Event"))
df

    > df
   ID Event
1   1     b
2   1     e
3   1     f
4   1     h
5   1     i
6   2     a
7   2     b
8   2     d
9   2     e
10  2     g
11  3     b
12  3     c
13  3     e
14  3     g
15  3     j
16  4     b
17  4     c
18  4     g
19  4     i
20  4     j

I want to select those IDs, that meet a list of criteria, that either use AND or OR.

For example:

1. those IDs that have events "b" AND "c" AND "g" --> results in ID 3 & 4
2. those IDs that have events "a" OR "h" --> results in ID 1 & 2

The criteria vectors can be any length.

EDIT:

I am aware of %in% and "|", however,

keep.if <- c("b", "c", "g") # This list can be of any length
subset(df, Event %in% keep.if)
ID Event
1   1     b
7   2     b
10  2     g
11  3     b
12  3     c
14  3     g
16  4     b
17  4     c
18  4     g

I only want those that have 3 rows in the results, so i can do a table on this results, and select those IDs where the Freq == length(keep.if)... but I guess there should be an easier, less messy method...

I guess for the OR version I can just take:

unique(subset(df, Event %in% keep.if)$ID)

Answer 1

I would create a table then use tidyr::spread to create a contigency table type object. Then I would use data.table for easier sub setting and logical operations:

library(tidyr)

df.table<-as.data.frame(table(df)) %>% spread(Event, Freq)
df.table

ID a b c d e f g h i j
1 0 1 0 0 1 1 0 1 1 0
2 1 1 0 1 1 0 1 0 0 0
3 0 1 1 0 1 0 1 0 0 1
4 0 1 1 0 0 0 1 0 1 1

library(data.table)
##easier to subset with

df.table<-data.table(df.table)
df.table[b & c & g]

ID a b c d e f g h i j
3 0 1 1 0 1 0 1 0 0 1
4 0 1 1 0 0 0 1 0 1 1

df.table[a | h]

ID a b c d e f g h i j
1 0 1 0 0 1 1 0 1 1 0
2 1 1 0 1 1 0 1 0 0 0

Those are the 2 examples you gave in the question. You should be able to do just about any operation you want. Also, if you only want to know which IDs satisfy your logic (and not their entire contingency table) then:

df.table[b & c & g]$ID
[1] 3 4