Sampling strategies for different data characteristics (i.e. imbalanced, small, exhaustive) have been discussed in the literature for the last two decades. In this study, sampling problem of small and continuous data is focused. Sampling with measured data by employing k-fold and the sampling with synthetic data produced with fuzzy cmeans clustering are applied, and also their performances on genetic programming and adaptive neuro fuzzy inference system (ANFIS) are discussed. When the experimental results are considered fuzzy c-means based synthetic sampling can be proposed.
|Title of host publication||The International Fuzzy Systems Association (IFSA)|
|Number of pages||4|
|Publication status||Published - 2011|
|Event||FUZZYSS'2011: 2nd International Fuzzy Systems Symposium: 17-18 November 2011, Ankara, Turkey - Ankara, Ankara, Turkey|
Duration: Nov 17 2011 → Nov 18 2011
|Conference||FUZZYSS'2011: 2nd International Fuzzy Systems Symposium|
|Period||11/17/11 → 11/18/11|