Skip to contents

Generate Content from Topic Terms

Source: R/ai_integration.R
generate_topic_content.Rd

Uses Large Language Models (LLMs) to generate various types of content based on topic model terms. Supports multiple content types with optimized default prompts, or fully custom prompts.

Usage

generate_topic_content(
  topic_terms_df,
  content_type = c("survey_item", "research_question", "theme_description",
    "policy_recommendation", "interview_question", "custom"),
  topic_var = "topic",
  term_var = "term",
  weight_var = "beta",
  provider = c("openai", "ollama"),
  model = "gpt-3.5-turbo",
  temperature = 0,
  system_prompt = NULL,
  user_prompt_template = NULL,
  max_tokens = 150,
  api_key = NULL,
  output_var = NULL,
  verbose = TRUE
)

Arguments

topic_terms_df

A data frame with topic terms, containing columns for topic identifier, term, and optionally term weight (beta).

content_type

Type of content to generate. One of:

"survey_item"

Likert-scale survey items for scale development

"research_question"

Research questions for literature review

"theme_description"

Theme descriptions for qualitative analysis

"policy_recommendation"

Policy recommendations for policy analysis

"interview_question"

Interview questions for qualitative research

"custom"

Custom content using user-provided prompts

topic_var

Name of the column containing topic identifiers (default: "topic").

term_var

Name of the column containing terms (default: "term").

weight_var

Name of the column containing term weights (default: "beta").

provider

LLM provider: "openai" or "ollama" (default: "openai").

model

Model name. For OpenAI: "gpt-3.5-turbo", "gpt-4", etc. For Ollama: "llama3", "mistral", etc.

temperature

Sampling temperature (0-2). Lower = more deterministic (default: 0).

system_prompt

Custom system prompt. If NULL, uses default for content_type.

user_prompt_template

Custom user prompt template with {terms} placeholder. If NULL, uses default for content_type.

max_tokens

Maximum tokens for response (default: 150).

api_key

OpenAI API key. If NULL, reads from OPENAI_API_KEY environment variable.

output_var

Name of the output column (default: based on content_type).

verbose

Logical, if TRUE, prints progress messages.

Value

A data frame with generated content joined to original topic terms.

Details

The function generates one piece of content per unique topic. Each content type has optimized default prompts, but these can be overridden with custom prompts.

For OpenAI, requires an API key set via the api_key parameter or OPENAI_API_KEY environment variable (can be loaded from .env file).

For Ollama, requires a local Ollama installation with the specified model.

See also

Other ai: analyze_contrastive_similarity(), call_ollama(), check_ollama(), create_label_selection_data(), format_label_candidates(), generate_survey_items(), generate_topic_labels_langgraph(), get_content_type_prompt(), get_content_type_user_template(), get_recommended_ollama_model(), list_ollama_models(), run_rag_search(), validate_topic_labels_langgraph()

Examples

if (FALSE) { # \dontrun{
# Generate survey items
survey_items <- generate_topic_content(
  topic_terms_df = top_terms,
  content_type = "survey_item",
  provider = "openai",
  model = "gpt-3.5-turbo"
)

# Generate research questions
research_qs <- generate_topic_content(
  topic_terms_df = top_terms,
  content_type = "research_question",
  provider = "ollama",
  model = "llama3"
)

# Generate with custom prompt
custom_content <- generate_topic_content(
  topic_terms_df = top_terms,
  content_type = "custom",
  system_prompt = "You are an expert in educational policy...",
  user_prompt_template = "Based on {terms}, generate a learning objective:"
)
} # }