作者: Alexandre Klementiev , Dan Roth , Kevin Small
DOI: 10.1007/978-3-540-74958-5_60
关键词:
摘要: Many applications in information retrieval, natural language processing, data mining, and related fields require a ranking of instances with respect to specified criteria as opposed classification. Furthermore, for many such problems, multiple established models have been well studied it is desirable combine their results into joint ranking, formalism denoted rank aggregation. This work presents novel unsupervisedlearning algorithm aggregation (ULARA) which returns linear combination the individual functions based on principle rewarding ordering agreement between rankers. In addition presenting ULARA, we demonstrate its effectiveness fusion task across ad hoc retrieval systems.