Search

Many Error-Correcting Output Codes (ECOC) algorithms had been proposed based on the hard coding (HC) schemes: binary coding {1, 0} or ternary coding {+ 1, -1, 0}. This paper introduces two novel strategies to recode the original code matrices with the mean values and the intervals of learners' outputs, which are named Mean Value Recoding (MVR) and Interval Recoding (IR) strategies. Both strategies are designed to reduce the distance between the outputs of base learners and the target codewords, aiming to produce more accurate results compared with the HC schemes. It is the first time that two concepts, soft recoding, and learner dependent, are injected into the ECOC framework to the best of our knowledge. To verify the effectiveness of our strategies, four data-independent ECOC algorithms and two data-dependent ECOC algorithms are deployed in the experiments based on UCI data sets. The experiments are carried out using the original HC strategies and our soft recoding strategies, and results verify that our strategies outperform the HC-based algorithms in most cases by producing balanced results among classes. In short, our strategies can improve the performance of different ECOC algorithms. Our python code and the corresponding data sets are available for non-commercial or research use at: https://github.com/MLDMXM2017/softcoding-ECOC. [ABSTRACT FROM AUTHOR]

The Open-Set recognition is an important topic in the pattern recognition research field. Different from the close-set recognition task, in the open-set recognition problem, the test data contains unknown classes that do not appear in the training phase. Consequently, the recognition of the open-set data is much more difficult than that of the close-set problem. This study applies the Error-Correcting Output Codes (ECOC) framework to handle the open-set problem by dynamically adding new functions to deal with the unknown classes, named ECOC-OS. Our algorithm includes two steps: (1) the unknown data discovery step based on a rejection strategy; (2) the code matrix expanding step for the separation of the unknown classes from the known classes. Due to the wide and chaotic distribution of the unknown class samples, this paper refines the unknown class into multiple sub-classes, and each sub-class has its own feature distribution. After preliminary row and column expansion and class splitting for the unknown class, the clustering algorithm is used to continuously refine the characteristics of the unknown class, dividing it into several sub-classes. Then the algorithm adds multiple coding rows and multiple "one-to-all" basic classifiers, so as to distinguish each unknown sub-class from multiple known classes. Finally, without re-training the existing learners, the zero symbols in the code matrix are selectively re-encoded according to the basic learners' preference. The experiments deploy 16 data sets for the test, and the results confirm that ECOC-OS algorithm effectively improves the performance compared with other open-set recognition methods. [ABSTRACT FROM AUTHOR]