====================================================
TAC KBP 2016 CHINESE SLOT FILLING EVALUATION RESULTS
====================================================


Team ID:  LDC
Organization:  Linguistic Data Consortium


*************************************************************

Run ID:  LDC_SF_CMN_1
Did the run access the live Web during the evaluation window:  Yes
Did the run extract relations from the Cold Start source corpus: Yes
Did the run generate meaningful confidence values: Yes

Slot Filling Evaluation:

Metric                   RunID        Hop Prec   Recall F1    
SF-ALL-Micro             LDC_SF_CMN_1 0   0.7757 0.2210 0.3440
SF-ALL-Micro             LDC_SF_CMN_1 1   0.7500 0.2609 0.3871
SF-ALL-Micro             LDC_SF_CMN_1 ALL 0.7687 0.2304 0.3545
SF-ALL-Macro             LDC_SF_CMN_1 0   0.7805 0.6105 0.6348
SF-ALL-Macro             LDC_SF_CMN_1 1   0.3974 0.3710 0.3743
SF-ALL-Macro             LDC_SF_CMN_1 ALL 0.6030 0.4996 0.5141
LDC-MAX-ALL-Micro        LDC_SF_CMN_1 0   0.7546 0.2299 0.3524
LDC-MAX-ALL-Micro        LDC_SF_CMN_1 1   0.7792 0.3261 0.4598
LDC-MAX-ALL-Micro        LDC_SF_CMN_1 ALL 0.7625 0.2545 0.3816
LDC-MAX-ALL-Macro        LDC_SF_CMN_1 0   0.7444 0.6047 0.6244
LDC-MAX-ALL-Macro        LDC_SF_CMN_1 1   0.5455 0.5093 0.5138
LDC-MAX-ALL-Macro        LDC_SF_CMN_1 ALL 0.6547 0.5617 0.5745
LDC-MEAN-ALL-Macro       LDC_SF_CMN_1 0   0.7444 0.6047 0.6244
LDC-MEAN-ALL-Macro       LDC_SF_CMN_1 1   0.4318 0.4009 0.4039
LDC-MEAN-ALL-Macro       LDC_SF_CMN_1 ALL 0.6035 0.5128 0.5250

*ALL-Macro Prec, Recall and F1 refer to mean-precision, mean-recall and mean-F1.

NIL-DETECTION P/R/F1:				0.2650 0.7886 0.3967