Actually, these are 7 types of language patterns that are difficult to analyse with automated algorithms:

1. "A land of milk and honey" becomes "A land of Milken Honey" (algorithm trained on Wall Street Journal from the 1980's where Michael Milken was mentioned much more than milk)
2. "She threw up her dinner" vs. "She threw up her hands"
3. "I ate a tomato with salt" vs. "I ate a tomato with my mother" or "I ate a tomato with a fork"
4. Words ending with -ing, e.g. "They were entertaining people"
5. "He washed and dried the dishes", vs. "He drank and smoked cigars" (in the latter case he did not drunk cigars)
6. "The lamb was ready to eat" vs. "Was the lamb hungry and wanting some grass?"
7. Words with multiple meaning (e.g. a bay can be a color, type of window or body of water)