Analyzes sentiment using Large Language Models (OpenAI or Gemini). Provides nuanced sentiment understanding including sarcasm detection, mixed emotions, and contextual interpretation that lexicon-based methods miss.
Usage
analyze_sentiment_llm(
texts,
doc_names = NULL,
provider = c("openai", "gemini"),
model = NULL,
api_key = NULL,
batch_size = 5,
include_explanation = FALSE,
verbose = TRUE
)Arguments
- texts
Character vector of texts to analyze.
- doc_names
Optional character vector of document names (default: text1, text2, ...).
- provider
AI provider to use: "openai" (default) or "gemini".
- model
Model name. If NULL, uses provider defaults: "gpt-4.1-mini" (OpenAI), "gemini-2.5-flash-lite" (Gemini).
- api_key
API key for OpenAI or Gemini. If NULL, uses environment variable.
- batch_size
Number of texts to process per API call (default: 5). Larger batches are more efficient but may hit token limits.
- include_explanation
Logical, if TRUE includes natural language explanation for each sentiment classification (default: FALSE).
- verbose
Logical, if TRUE prints progress messages (default: TRUE).
Value
A list containing:
- document_sentiment
Data frame with document-level sentiment scores
- summary_stats
Summary statistics of the analysis
- model_used
Model name used for analysis
- provider
AI provider used
Details
LLM-based sentiment analysis offers several advantages over lexicon methods:
Understands context and nuance
Detects sarcasm and irony
Handles mixed emotions
Works across domains without retraining
Examples
if (interactive()) {
abstracts <- TextAnalysisR::SpecialEduTech$abstract[1:5]
sentiment_openai <- analyze_sentiment_llm(abstracts, provider = "openai")
sentiment_gemini <- analyze_sentiment_llm(abstracts, provider = "gemini",
include_explanation = TRUE)
}