Sequential testing in classifier evaluation yields biased estimates of effectiveness.

@inproceedings{webber-2013-sequential, author = {William Webber and Mossaab Bagdouri and David D. Lewis and Douglas W. Oard}, editor = {Gareth J. F. Jones and Paraic Sheridan and Diane Kelly and Maarten de Rijke and Tetsuya Sakai}, title = {{Sequential testing in classifier evaluation yields biased estimates of effectiveness}}, booktitle = {{The 36th International ACM SIGIR conference on research and development in Information Retrieval, SIGIR '13, Dublin, Ireland - July 28 - August 01, 2013}}, pages = {933--936}, publisher = {ACM}, year = {2013}, url = {https://doi.org/10.1145/2484028.2484159}, doi = {https://doi.org/10.1145/2484028.2484159} }