An evaluation of Naive Bayesian anti-spam filtering Export

@inproceedings{androutsopoulos00:_bayesian_antispam_eval, abstract = {It has recently been argued that a Naive Bayesian classifier can be used to filter unsolicited bulk e-mail ("spam"). We conduct a thorough evaluation of this proposal on a corpus that we make publicly available, contributing towards standard benchmarks. At the same time we investigate the effect of attribute-set size, training-corpus size, lemmatization, and stop-lists on the filter's performance, issues that had not been previously explored. After introducing appropriate cost-sensitive evaluation measures, we reach the conclusion that additional safety nets are needed for the Naive Bayesian anti-spam filter to be viable in practice.}, archivePrefix = {arXiv}, author = {Androutsopoulos, Ion and Koutsias, John and Chandrinos, Konstantinos V. and Paliouras, George and Spyropoulos, Constantine D.}, booktitle = {Workshop on Machine Learning in the New Information Age}, citeulike-article-id = {682906}, citeulike-linkout-0 = {http://arxiv.org/abs/cs.CL/0006013}, citeulike-linkout-1 = {http://arxiv.org/pdf/cs.CL/0006013}, day = {7}, eprint = {cs.CL/0006013}, keywords = {bayes, machinelearning}, month = {Jun}, pages = {9--17}, posted-at = {2006-06-04 07:22:01}, priority = {2}, title = {An evaluation of Naive Bayesian anti-spam filtering}, url = {http://arxiv.org/abs/cs.CL/0006013}, year = {2000} }

TY - CONF ID - androutsopoulos00:_bayesian_antispam_eval L3 - citeulike-article-id:682906 N2 - It has recently been argued that a Naive Bayesian classifier can be used to filter unsolicited bulk e-mail ("spam"). We conduct a thorough evaluation of this proposal on a corpus that we make publicly available, contributing towards standard benchmarks. At the same time we investigate the effect of attribute-set size, training-corpus size, lemmatization, and stop-lists on the filter's performance, issues that had not been previously explored. After introducing appropriate cost-sensitive evaluation measures, we reach the conclusion that additional safety nets are needed for the Naive Bayesian anti-spam filter to be viable in practice. EP - 17 TI - An evaluation of Naive Bayesian anti-spam filtering T2 - Workshop on Machine Learning in the New Information Age SP - 9 KW - bayes KW - machinelearning AU - Androutsopoulos, Ion AU - Koutsias, John AU - Chandrinos, Konstantinos AU - Paliouras, George AU - Spyropoulos, Constantine PY - 2000/Jun/07/ UR - http://arxiv.org/abs/cs.CL/0006013 ER -