To check whether you understood the material we covered on the 20th of October, 
try to find answers to the following questions regarding the methodology employed by
Bos & Markert (2005) "Recognising Textual Entailment with Logical Inference"

1) what is the topic the authors are investigating and how can it
   be formulated as a classification task?

2) what kind of data do the authors use for their experiments?
   what is the size of the corpus? is it annotated? if so,
   with what information?

3) what type of machine learning (supervised or unsupervised) is used?
   which is the particular type of algorithm chosen?

4) how is the data partitioned for training and testing?

5) which features are considered to model each <T,H> pair?

6) which evaluation measures are used to report the results?

7) what is the baseline against which the results are compared?
   is there an upper bound?