While this sounds interesting, the feature selection was done on the same data as the tests were conducted later on. If I read it correctly, that's a biggie, e

While this sounds interesting, the feature selection was done on the same data as the tests were conducted later on. If I read it correctly, that's a biggie, especially given the weak outperformance of chance.