get_cluster_recommendations: Get Recommendations for Clusters

In adamgdobrakowski/memr: Multisource Embeddings for Medical Records

Description

Get recommendations prescribed by doctors to patients from given clusters. The categories of recommendation terms can be specified.

Usage

get_cluster_recommendations(recom_descriptions, clusters,
  category = "all", recom_table)

Arguments

recom_descriptions	A vector of descriptions of recommendations, named by visits' IDs; terms in the descriptions are separated by ", "
clusters	An output of the function cluster_visits
category	A vector of categories of the terms to be returned or "all" (default: "all")
recom_table	(necessary if category != "all") A data frame with columns: term a term of recommendation category a category of the term

Value

A list of data frames. For each cluster there is a data frame with columns:

recommendation	A term of recommendation
count	A number of occurrences of the recommentadion in the cluster
frequency	A percentage of visits containing the reccomentation

Rows are sorted by the frequency.

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)
clusters <- cluster_visits(visits_vectors, visits,
                           spec = "internist",
                           cluster_number = 2L)

get_cluster_recommendations(recommendations, clusters, recom_table = terms_categories)
get_cluster_recommendations(recommendations, clusters, recom_table = terms_categories,
category = "anatomic")
get_cluster_recommendations(recommendations, clusters, recom_table = terms_categories,
category = "all")

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