embed_list_visits: Generate Embeddings of Visits from a List

In adamgdobrakowski/memr: Multisource Embeddings for Medical Records

Description

Generate embeddings of given visits, contained in two named vectors. One vector contains the descriptions of interviews and the second – the descriptions of examinations. This function interatively calls embed_visit function.

Usage

embed_list_visits(inter_descriptions, exam_descriptions,
  inter_term_vectors, exam_term_vectors)

Arguments

inter_descriptions	A vector of descriptions of interviews, named by visits' IDs; terms in the descriptions are separated by ", "
exam_descriptions	A vector of descriptions of examinations, named by visits' IDs; the visits' IDs do not have to be identical to inter_descriptions
inter_term_vectors	A matrix of embeddings of the interview terms
exam_term_vectors	A matrix of embeddings of the examination terms

Value

A matrix of embeddings of visits – in each row there is an embedding of one visit.

See Also

embed_visit, embed_description

Examples

inter_term_vectors <- embed_terms(interviews, embedding_size = 10L,
  term_count_min = 1L)
exam_term_vectors <- embed_terms(examinations, embedding_size = 10L,
  term_count_min = 1L)

visits_vectors <- embed_list_visits(interviews,
                                    examinations,
inter_term_vectors, exam_term_vectors)
nrow(visits_vectors)
visits_vectors
visits_vectors[is.na(visits_vectors)] <- 0
D <- as.matrix(dist(visits_vectors))

# Two closest visits
examinations[rownames(which(D == min(D[D > 0]),
                  arr.ind = TRUE))]

visits_vectors <- embed_list_visits(interviews, "",
  inter_term_vectors, exam_term_vectors)
nrow(visits_vectors)
visits_vectors

adamgdobrakowski/memr documentation built on Sept. 4, 2021, 3:45 a.m.