MPSUM: Entity Summarization with Predicate-based Matching
This work addresses entity summarization for Semantic Web applications, but it appears incremental as it builds upon existing topic models with specific enhancements.
The authors tackled the problem of generating concise and representative summaries for entities in the Semantic Web by proposing MPSUM, a method that integrates predicate-uniqueness and object-importance into a probabilistic topic model, and they reported improvements in summarization quality on DBpedia and LinkedMDB datasets.
With the development of Semantic Web, entity summarization has become an emerging task to generate concrete summaries for real world entities. To solve this problem, we propose an approach named MPSUM that extends a probabilistic topic model by integrating the idea of predicate-uniqueness and object-importance for ranking triples. The approach aims at generating brief but representative summaries for entities. We compare our approach with the state-of-the-art methods using DBpedia and LinkedMDB datasets.The experimental results show that our work improves the quality of entity summarization.