[Adapt] Labeling of image conceptualization

[赵凯祺]

Hi adapters,


We need your help in the evaluation of image conceptualization. 
In the demo site: http://202.120.38.145:4081, you can select 50 different queries in drop-down list, and the system will return the image clusters(different entities) corresponds to the query. (See the snapshot in the bottom) On top of each cluster, you will see 10 different terms which describes the entity for the cluster. Could you please evaluate the correctness of the terms? If it is related to the entity, label it as correct, otherwise incorrect.
So, for each cluster of each query, please record the number of correct terms among the 10 terms.


Please write down the result in the following format in a txt file, and send it to me by tomorrow morning.


query 1
# of correct terms in cluster 1 of query 1
# of correct terms in cluster 2 of query 1
...
# of correct terms in cluster 3 of query 1


query 2
# of correct terms in cluster 1 of query 2
...


For example:
acrobat
10
9
8
7
...


adam
10
10
...








Thanks!
Kaiqi

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20130515/8f0d14fc/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ??3.png
Type: image/png
Size: 186726 bytes
Desc: not available
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20130515/8f0d14fc/attachment-0001.png>