Generate Cluster Labels with AI — generate_cluster

Generates descriptive labels for clusters using either Ollama (local, default) or OpenAI's API. When running locally, Ollama is preferred for privacy and cost-free operation.

Usage

generate_cluster_labels(
  cluster_keywords,
  provider = "auto",
  model = NULL,
  temperature = 0.3,
  max_tokens = 50,
  verbose = TRUE
)

Arguments

cluster_keywords: List of keywords for each cluster.
provider: AI provider to use: "auto" (default), "ollama", or "openai". "auto" will use Ollama if available, otherwise OpenAI.
model: Model name. For Ollama (default: "phi3:mini"). For OpenAI (default: "gpt-3.5-turbo").
temperature: Temperature parameter (default: 0.3).
max_tokens: Maximum tokens for response (default: 50).
verbose: Logical, if TRUE, prints progress messages.

Value

A list of generated labels.

Examples

if (FALSE) { # \dontrun{
keywords <- list("1" = c("machine", "learning", "neural"), "2" = c("data", "analysis"))
labels_ollama <- generate_cluster_labels(keywords, provider = "ollama")
labels_openai <- generate_cluster_labels(keywords, provider = "openai")
} # }