@inproceedings{pranckeviciene2006class,
  title={Class separability in spaces reduced by feature selection},
  author={Pranckeviciene, Erinija and Ho, TinKam and Somorjai, Ray},
  booktitle={Proceedings of 18th International Conference on Pattern Recognition, 2006. ICPR 2006. August 20-24, Hong Kong},
  volume={3},
  pages={254--257},
  year={2006},
  publisher={IEEE},
  url_Link={http://ieeexplore.ieee.org/document/1699514/}, 
  abstract={We investigated the geometrical complexity of several high-dimensional, small sample classification problems and its changes due to two popular feature selection procedures, forward feature selection (FFS) and linear programming support vector machine (LPSVM). We found that both procedures are able to transform the problems to spaces of very low dimensionality where class separability is improved over that in the original space. The study shows that geometrical complexities have good potentials for comparing different feature selection methods in aspects relevant to classification accuracy, yet independent of particular classifier choices.},
keywords={support vector machines, support vector machine classification, biomedical measurements, linear programming, labeling, geometry}, 
url_Paper={http://erinijapranckeviciene.net/data_science_projects/ICPR2006_Class_Separability.pdf}

}

{"_id":"s4vAay4EmN7QQ9KBb","bibbaseid":"pranckeviciene-ho-somorjai-classseparabilityinspacesreducedbyfeatureselection-2006","downloads":0,"creationDate":"2017-02-13T03:40:38.580Z","title":"Class separability in spaces reduced by feature selection","author_short":["Pranckeviciene, E.","Ho, T.","Somorjai, R."],"year":2006,"bibtype":"inproceedings","biburl":"http://erinijapranckeviciene.net/erinija_publications.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"Class separability in spaces reduced by feature selection","author":[{"propositions":[],"lastnames":["Pranckeviciene"],"firstnames":["Erinija"],"suffixes":[]},{"propositions":[],"lastnames":["Ho"],"firstnames":["TinKam"],"suffixes":[]},{"propositions":[],"lastnames":["Somorjai"],"firstnames":["Ray"],"suffixes":[]}],"booktitle":"Proceedings of 18th International Conference on Pattern Recognition, 2006. ICPR 2006. August 20-24, Hong Kong","volume":"3","pages":"254–257","year":"2006","publisher":"IEEE","url_link":"http://ieeexplore.ieee.org/document/1699514/","abstract":"We investigated the geometrical complexity of several high-dimensional, small sample classification problems and its changes due to two popular feature selection procedures, forward feature selection (FFS) and linear programming support vector machine (LPSVM). We found that both procedures are able to transform the problems to spaces of very low dimensionality where class separability is improved over that in the original space. The study shows that geometrical complexities have good potentials for comparing different feature selection methods in aspects relevant to classification accuracy, yet independent of particular classifier choices.","keywords":"support vector machines, support vector machine classification, biomedical measurements, linear programming, labeling, geometry","url_paper":"http://erinijapranckeviciene.net/data_science_projects/ICPR2006_Class_Separability.pdf","bibtex":"@inproceedings{pranckeviciene2006class,\r\n title={Class separability in spaces reduced by feature selection},\r\n author={Pranckeviciene, Erinija and Ho, TinKam and Somorjai, Ray},\r\n booktitle={Proceedings of 18th International Conference on Pattern Recognition, 2006. ICPR 2006. August 20-24, Hong Kong},\r\n volume={3},\r\n pages={254--257},\r\n year={2006},\r\n publisher={IEEE},\r\n url_Link={http://ieeexplore.ieee.org/document/1699514/}, \r\n abstract={We investigated the geometrical complexity of several high-dimensional, small sample classification problems and its changes due to two popular feature selection procedures, forward feature selection (FFS) and linear programming support vector machine (LPSVM). We found that both procedures are able to transform the problems to spaces of very low dimensionality where class separability is improved over that in the original space. The study shows that geometrical complexities have good potentials for comparing different feature selection methods in aspects relevant to classification accuracy, yet independent of particular classifier choices.},\r\nkeywords={support vector machines, support vector machine classification, biomedical measurements, linear programming, labeling, geometry}, \r\nurl_Paper={http://erinijapranckeviciene.net/data_science_projects/ICPR2006_Class_Separability.pdf}\r\n\r\n}\r\n\r\n\r\n","author_short":["Pranckeviciene, E.","Ho, T.","Somorjai, R."],"key":"pranckeviciene2006class","id":"pranckeviciene2006class","bibbaseid":"pranckeviciene-ho-somorjai-classseparabilityinspacesreducedbyfeatureselection-2006","role":"author","urls":{" link":"http://ieeexplore.ieee.org/document/1699514/"," paper":"http://erinijapranckeviciene.net/data_science_projects/ICPR2006_Class_Separability.pdf"},"keyword":["support vector machines","support vector machine classification","biomedical measurements","linear programming","labeling","geometry"],"downloads":0,"html":""},"search_terms":["class","separability","spaces","reduced","feature","selection","pranckeviciene","ho","somorjai"],"keywords":["support vector machines","support vector machine classification","biomedical measurements","linear programming","labeling","geometry"],"authorIDs":[],"dataSources":["wyhwdvMCTbALq9k63"]}