100,000 templated questions and answers covering 75,000 movie-related entities derived from the Open Movie Database (OMDb). The dataset provides a structured knowledge base for question answering tasks within the film domain.
Use Cases
- Train a question-answering model to identify movie entities based on templated natural language queries
- Evaluate knowledge graph completion techniques by mapping entities to their corresponding OMDb attributes
- Develop natural language interfaces for movie databases using the question-answer pairs as training data
Strengths
- Contains approximately 100,000 question-answer pairs
- Features 75,000 unique entities sourced from the Open Movie Database
- Utilizes templated natural language questions to ensure consistency across the knowledge graph