CODEWITHSUNDEEP
rjsonsparse-matrixcosine-similaritytibble
road
road

Reputation: 479

Converting a Cosine Similarity Matrix to JSON in R

I have a Cosine similarity matrix (csm), which looks like this:

 csm<-matrix( c(1,0,0.4,0.6, 0,0,1, 0.1,0.4,0.7,0.4,0.1,1,0.9,0.5,0.6,0.4,0.9,1,0.6,0,0.7,0.5,0.6,1),nrow=5,ncol=5)
 rownames(csm) <- c("AAB","AAC","AAD","AAE","AAF")
 colnames(csm) <- c("AAB","AAC","AAD","AAE","AAF")

csm
    AAB   AAC   AAD   AAE    AAF
AAB 1     0     0.4   0.6    0
AAC 0     1     0.1   0.4    0.7
AAD 0.4   0.1   1     0.9    0.5
AAE 0.6   0.4   0.9   1      0.6
AAF 0     0.7   0.5   0.6    1

My goal is to build a JSON file that looks like this:

    {"AAB":[{"ID":"AAB","value":1},{"ID":"AAC","value":0},{"ID":"AAD","value":0.4},{"ID":"AAE","value":0.6},{"ID":"AAF","value":0}],
     "AAC":[{"ID":"AAB","value":0},{"ID":"AAC","value":1},{"ID":"AAD","value":0.1},{"ID":"AAE","value":0.4},{"ID":"AAF","value":0.7}],
     "AAD":[{"ID":"AAB","value":0.4},{"ID":"AAC","value":0.3},{"ID":"AAD","value":1},{"ID":"AAE","value":0.9},{"ID":"AAF","value":0.5}],
     "AAE":[{"ID":"AAB","value":0.6},{"ID":"AAC","value":0.4},{"ID":"AAD","value":0.5},{"ID":"AAE","value":1},{"ID":"AAF","value":0.6}],
     "AAF":[{"ID":"AAB","value":0},{"ID":"AAC","value":0},{"ID":"AAD","value":0.7},{"ID":"AAE","value":0},{"ID":"AAF","value":1}]}

What I have done is this:

csm %>% as_tibble() %>% 
    gather( ID, value) %>% 
    complete(ID, value)%>%
    jsonlite::toJSON()

What I am having is :

[{"ID":"AAB","value":0},{"ID":"AAB","value":0},{"ID":"AAB","value":0.1},{"ID":"AAB","value":0.3},{"ID":"AAB","value":0.4},...

Could someone help? Thank you!

Upvotes: 0

Views: 103

Answers (1)

Maurits Evers
Maurits Evers

Reputation: 50678

Here is a base R solution using stack

library(jsonlite)
toJSON(lapply(split(df2 <- data.frame(ID = rownames(df), stack(df)), df2$ind), function(x)
    x[, -3]))
#{"AAB":[{"ID":"AAB","values":1},{"ID":"AAC","values":0},{"ID":"AAD","values":0.4},{"ID":"AAE","values":0.6},{"ID":"AAF","values":0}],"AAC":[{"ID":"AAB","values":0},{"ID":"AAC","values":1},{"ID":"AAD","values":0.1},{"ID":"AAE","values":0.4},{"ID":"AAF","values":0.7}],"AAD":[{"ID":"AAB","values":0.4},{"ID":"AAC","values":0.1},{"ID":"AAD","values":1},{"ID":"AAE","values":0.9},{"ID":"AAF","values":0.5}],"AAE":[{"ID":"AAB","values":0.6},{"ID":"AAC","values":0.4},{"ID":"AAD","values":0.9},{"ID":"AAE","values":1},{"ID":"AAF","values":0.6}],"AAF":[{"ID":"AAB","values":0},{"ID":"AAC","values":0.7},{"ID":"AAD","values":0.5},{"ID":"AAE","values":0.6},{"ID":"AAF","values":1}]}

Or the tidyverse way

df %>%
    rownames_to_column("row") %>%
    gather(ID, value, -row) %>%
    split(.$row) %>%
    map(~.x[, -1]) %>%
    toJSON()
#{"AAB":[{"ID":"AAB","value":1},{"ID":"AAC","value":0},{"ID":"AAD","value":0.4},{"ID":"AAE","value":0.6},{"ID":"AAF","value":0}],"AAC":[{"ID":"AAB","value":0},{"ID":"AAC","value":1},{"ID":"AAD","value":0.1},{"ID":"AAE","value":0.4},{"ID":"AAF","value":0.7}],"AAD":[{"ID":"AAB","value":0.4},{"ID":"AAC","value":0.1},{"ID":"AAD","value":1},{"ID":"AAE","value":0.9},{"ID":"AAF","value":0.5}],"AAE":[{"ID":"AAB","value":0.6},{"ID":"AAC","value":0.4},{"ID":"AAD","value":0.9},{"ID":"AAE","value":1},{"ID":"AAF","value":0.6}],"AAF":[{"ID":"AAB","value":0},{"ID":"AAC","value":0.7},{"ID":"AAD","value":0.5},{"ID":"AAE","value":0.6},{"ID":"AAF","value":1}]}

Sample data

df <- read.table(text =
    "    AAB   AAC   AAD   AAE    AAF
AAB 1     0     0.4   0.6    0
AAC 0     1     0.1   0.4    0.7
AAD 0.4   0.1   1     0.9    0.5
AAE 0.6   0.4   0.9   1      0.6
AAF 0     0.7   0.5   0.6    1")

Upvotes: 1

Related Questions