O'Reilly logo

Natural Language Processing and Cognitive Science by Michael Zock, Mai Yanagimura, Elena Yagunova, Svetlana Volskaya, Jordi Turmo, Susanna Tron, Rocco Tripodi, Radu Topor, Ludovic Tanguy, David Suendermann-Oeft, Elizabeth Rosenfeld, Horacio Rodriguez, Livio Robaldo, Solange O. Rezende, Michaela Regneri, Lidia Pivovarova, Marcello Pelillo, Catherine Pelachaud, Thiago S. Pardo, Ali M. Naderi, Sachiyo Muranishi, Michael Muck, Martin Mory, Francois Morlane-Hondere, Tarek Mehrez, Wieslaw Lubaszewski, Patrick Lange, Shingo Kuroiwa, Manfred Klenner, Diane King, Mohamed Hamed Kholief, Alexei V. Ivanov, Yasuo Horiuchi, Nora Hollenstein, Myriam Hernandez Alvarez, Jiri Havelka, Nabil Hathout, Marcin Hareza, Jose M. Gomez, Emiliano Giovannetti, Nadine Glas, Daniela Gifu, Izabela Gatkowska, Jean-Gabriel Ganascia, Daisuke Furukawa, Richard Frost, Cécile Fabre, Ahmed Magdy Ezzeldin, Yasser El-Sonbaty, Luigi Di Caro, Matthew Crocker, Dan Cristea, Jan Curin, Conrado S. Merley, Pawel Chrzazcz, Jesus Calvillo, Mohamed Amine Boukhaled, Guido Boella, Jared Bernstein, Alessia Bellusci, Andrea Bellandi, Raimo Bakis, Eniafe Festus Ayetiran, Michael Amsler, Nabil Abdullah, César Aguilar, Olga Acosta, Rodolfo Delmonte, Bernadette Sharp

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

Jared Bernstein, Alexei V. Ivanov, and Elizabeth Rosenfeld

Benchmarking Automated Text Correction Services

Abstract: We compared the performance of two automated grammar checkers on a small random sample of student texts to select the better text checker to incorporate in a product to tutor American students, age 12–16, in essay writing. We studied two competing error detection services that located and labeled errors with claimed accuracy well above “90%”. However, performance measurement with reference to a small set of essays (double-annotated for text errors) found that both services operated at similar low accuracies (F1 values in the range of 0.25 to 0.35) when analyzing either sentence- or word-level errors. Several findings emerged. ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required