[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki

[ Back to the navigation ]

You are here: start » courses » rg » 2013 » composite-activities

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
courses:rg:2013:composite-activities [2013/04/23 18:06]
machys 		courses:rg:2013:composite-activities [2013/09/29 21:35] (current)
machys
Line 4: Line 4:
    - Thinking about the scripts:    - Thinking about the scripts:
       * What is the main reason (the biggest advantage) of using scripts? What kind of information does it bring? (Hint: page 2, page 8)       * What is the main reason (the biggest advantage) of using scripts? What kind of information does it bring? (Hint: page 2, page 8)
-      * The authors don't get the "knowledge" of scripts straightforwardly. How is that "knowledge" represented in the model and which (two) ways are used to get it?+      * The authors don't get the "knowledge" of scripts straightforwardly. How is that "knowledge" represented in the model and which (four) ways are used to get it?
    - In the last paragraph of Section 3, a method is described that enhances the robustness of the model (binarization of all association weights <latex>w^z_i</latex>). Answer one of the following questions (choose one):    - In the last paragraph of Section 3, a method is described that enhances the robustness of the model (binarization of all association weights <latex>w^z_i</latex>). Answer one of the following questions (choose one):
       * Why does it work? (=> Why should it work best?)       * Why does it work? (=> Why should it work best?)
       * Do you have any idea how to do it differently?       * Do you have any idea how to do it differently?
    - Experiments: Which "tricks"/"parts of processing" enhanced the //Attribute recognition// and //Composite activity classification// tasks "the most"? Try to answer why.    - Experiments: Which "tricks"/"parts of processing" enhanced the //Attribute recognition// and //Composite activity classification// tasks "the most"? Try to answer why.
 +
 +====== Answers ======
 +
 +   - First set
 +      * list components [[https://docs.google.com/drawings/d/12wIoIDVDV3b6EbJAVVIzZn9h0ymIgHRHi_sTauGM08A/edit|(Google doc graph)]]
 +      * dependance of components (the same graph)
 +   - Scripts
 +      * reason?: Cheap source of training data, Many combinations, unseen variants, “decsriptions” of the same thing
 +      * four ways: 2x2: 1) direct use of words from data or 2) mapping word classes from WordNet X 3) simple word frequency or 4) TF*IDF
 +   - There was a discussion about 3rd set of question. We are not sure why authors do that. There was strongly supported opinion that autohors do a lot unnecessary work, which is lost by binarization.
 +   - 4th: Majority people in aswers nominated the use of TF*IDF in case of no training data as the best idea.
  

[ Back to the navigation ] [ Back to the content ]