«Integrating chemists preferences for shape-similarity clustering of series»