Title :
Evaluation of combination methods
Author :
Impedovo, S. ; Salzo, A.
Author_Institution :
Dipt. di Inf., Bari Univ., Italy
Abstract :
This paper presents a new approach to evaluate the performances of combination methods, which takes into account both the recognition rates of the experts combined and the correlation among them. At the purpose, a suitable estimator of correlation is defined. Two combination methods have been considered: majority vote and Dempster Shafer. A statistical test, based on the analysis of variance, has also been used to infer some interesting considerations on the behaviour of combination methods. The paper shows how the proposed approach allows the selection of the best combination method for each set of experts
Keywords :
expert systems; inference mechanisms; optical character recognition; statistical analysis; Dempster Shafer method; combination methods; correlation estimator; experts; majority vote method; performance evaluation; recognition rates; statistical test; variance analysis; Analysis of variance; Character recognition; Degradation; Handwriting recognition; Optical character recognition software; Performance analysis; Postal services; Statistical analysis; Testing; Voting;
Conference_Titel :
Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth International Conference on
Conference_Location :
Bangalore
Print_ISBN :
0-7695-0318-7
DOI :
10.1109/ICDAR.1999.791807
