# A Comparison of State-of-the-Art Classification Techniques with Application to Cytogenetics

Boaz Lerner, Ben Gurion University
Neil D. Lawrence, University of Sheffield

Neural Computing and Applications 10, pp 39-47

#### Abstract

Several state-of-the-art techniques: a neural network, Bayesian neural network, support vector machine and naive Bayesian classifier are experimentally evaluated in discriminating fluorescence in-situ hybridization (FISH) signals. Highly accurate classification of signals from real data and artefacts of two cytogenetic probes (colours) is required for detecting abnormalities in the data. More than 3,100 FISH signals are classified by the techniques into colour and as real or artefact with accuracies of around 98% and 88%, respectively. The results of the comparison also show a trade-off between simplicity represented by the naive Bayesian classifier and high classification performance represented by the other techniques.

  @Article{lerner-comparison01, title = {A Comparison of State-of-the-Art Classification Techniques with Application to Cytogenetics}, journal = {Neural Computing and Applications}, author = {Boaz Lerner and Neil D. Lawrence}, pages = {39}, year = {2001}, volume = {10}, number = {1}, month = {00}, edit = {https://github.com/lawrennd//publications/edit/gh-pages/_posts/2001-04-01-lerner-comparison01.md}, url = {http://inverseprobability.com/publications/lerner-comparison01.html}, abstract = {Several state-of-the-art techniques: a neural network, Bayesian neural network, support vector machine and naive Bayesian classifier are experimentally evaluated in discriminating fluorescence in-situ hybridization (FISH) signals. Highly accurate classification of signals from real data and artefacts of two cytogenetic probes (colours) is required for detecting abnormalities in the data. More than 3,100 FISH signals are classified by the techniques into colour and as real or artefact with accuracies of around 98% and 88%, respectively. The results of the comparison also show a trade-off between simplicity represented by the naive Bayesian classifier and high classification performance represented by the other techniques.}, key = {Lerner-comparison01}, linkpsgz = {http://www.thelawrences.net/neil/comparison.ps.gz}, OPTgroup = {} }
 %T A Comparison of State-of-the-Art Classification Techniques with Application to Cytogenetics %A Boaz Lerner and Neil D. Lawrence %B %C Neural Computing and Applications %D %F lerner-comparison01 %J Neural Computing and Applications %P 39--47 %R %U http://inverseprobability.com/publications/lerner-comparison01.html %V 10 %N 1 %X Several state-of-the-art techniques: a neural network, Bayesian neural network, support vector machine and naive Bayesian classifier are experimentally evaluated in discriminating fluorescence in-situ hybridization (FISH) signals. Highly accurate classification of signals from real data and artefacts of two cytogenetic probes (colours) is required for detecting abnormalities in the data. More than 3,100 FISH signals are classified by the techniques into colour and as real or artefact with accuracies of around 98% and 88%, respectively. The results of the comparison also show a trade-off between simplicity represented by the naive Bayesian classifier and high classification performance represented by the other techniques. 
 TY - CPAPER TI - A Comparison of State-of-the-Art Classification Techniques with Application to Cytogenetics AU - Boaz Lerner AU - Neil D. Lawrence PY - 2001/04/01 DA - 2001/04/01 ID - lerner-comparison01 SP - 39 EP - 47 UR - http://inverseprobability.com/publications/lerner-comparison01.html AB - Several state-of-the-art techniques: a neural network, Bayesian neural network, support vector machine and naive Bayesian classifier are experimentally evaluated in discriminating fluorescence in-situ hybridization (FISH) signals. Highly accurate classification of signals from real data and artefacts of two cytogenetic probes (colours) is required for detecting abnormalities in the data. More than 3,100 FISH signals are classified by the techniques into colour and as real or artefact with accuracies of around 98% and 88%, respectively. The results of the comparison also show a trade-off between simplicity represented by the naive Bayesian classifier and high classification performance represented by the other techniques. ER - 
 Lerner, B. & Lawrence, N.D.. (2001). A Comparison of State-of-the-Art Classification Techniques with Application to Cytogenetics. Neural Computing and Applications 10(1):39-47