I recently came across an interesting articles in nytimes about the e-raters or robo raters that are now used to grade essays in various examinations. A MIT professor by the name Mr Perelman, had did some testing on the e-Rater of the ETS that conducts the gmat and has some pretty interesting insights to share, one of it being to use long sentences and connect them with positive progressive connectors like, 'however' and 'moreover', he also advises students to use complex words instead of simpler ones. The ETS has a pretty witty retort to this claims of one being able to fool the robo raters very easily, E.T.S. officials say that Mr. Perelman's test prep advice is too complex for most students to absorb; if they can, they're using the higher level of thinking the test seeks to reward anyway. In other words, if they're smart enough to master such sophisticated test prep, they deserve a 6. The full nytimes article can be found here.https://www.nytimes.com/2012/04/23/educa ... l?src=recg
I would like to hear my peers and the instructors thought on this. I do remember during my final years in college, my friends who were appearing for their GREs talk of such methods to to score higher in the analytical essays. I wonder of the computer science majors thought of similar ways since its pretty easy to find a work around a machine if you know the basis for an algorithm they would use. I remember, using a different version of this trick to beat basic Chess that comes packed with the windows pc. My argument was, the program was written so expecting we would want to win and do so by making rational decisions where every move is calculated and by the book, but the machine doesnt know, atleast not yet, that humans sometime dont act on logic or use the optimal ways of performing a task.
I would like to hear my peers and the instructors thought on this. I do remember during my final years in college, my friends who were appearing for their GREs talk of such methods to to score higher in the analytical essays. I wonder of the computer science majors thought of similar ways since its pretty easy to find a work around a machine if you know the basis for an algorithm they would use. I remember, using a different version of this trick to beat basic Chess that comes packed with the windows pc. My argument was, the program was written so expecting we would want to win and do so by making rational decisions where every move is calculated and by the book, but the machine doesnt know, atleast not yet, that humans sometime dont act on logic or use the optimal ways of performing a task.
