Surdeanu_ACL11_customizing an information extraction system to a new domain

The paper implements a pipeline system for relation extraction. First NE, the RE. The RE system features are selected based on the ACE corpus. The test domain is the sports news.

They showed that gazetteer doesn't help that much.
 Identify the head of arguments is important. They proposed several heuristics for this:

1. append "It was" to the entity, then parse the entity by the Stanford parser.
2. remove dashes as it is not common in the Penn treebank.
3. Guide the Stanford parser such that the final tree contains a constituent with the same span. (Need the check how to do this.)


They show domain specific deterministic rules also helps. such as "teamFinalScore(G, S) :- teamInGame(T,G), teamScoringAll(T,S)"

all kinds of steam buns

Making steam buns today.

The dough:
1 Table Spoon (T) active yeast.
1Cup (C) warm water.
4 C all purpose flour.
2 T sugar
1 t baking powder
1 t salt.

put ingredients into the bread machine. Found that my yeast is expired -_- Then I rush to Sobeys bought a new one. Came back, hope every thing will be OK.

stuffing:
three types.
1. ground pork: with ginger, garlic, cooking wine, white pepper. 15minutes steam.
2. sesame: with sugar. 1:1. 5 minutes steam.
3. red bean: mashed with red sugar. 10 minutes steam.

result:
OK, not enough water.... the ground pork tastes good, but sugar is not enough for sesame. The skin is not soft, too chewy... I guess it's because we left for 30 minutes for the yeast....

Try again next month. Too tired.