Results
Ranks were computed by summing the correlation values across all four levels of comparisons. The sum of the Pearson correlations is used for the official rank of Task 3. However, we also provide a second ranking using the sum of the Spearman correlations, which teams may use for comparison. The column "Difference" shows the difference between these two ranks.
Four systems were submitted after the deadline and are included here in a separate table for comparison, but are unranked.
Team | System | Para-2-Sent | Sent-2-Phr | Phr-2-Word | Word-2-Sense | Official Rank | Spearman Rank | Difference |
---|---|---|---|---|---|---|---|---|
SimCompass | run1 | 0.811 | 0.742 | 0.415 | 0.356 | 1 | 1 | 0 |
ECNU | run1 | 0.834 | 0.771 | 0.315 | 0.269 | 2 | 2 | 0 |
UNAL-NLP | run2 | 0.837 | 0.738 | 0.274 | 0.256 | 3 | 6 | -3 |
SemantiKLUE | run1 | 0.817 | 0.754 | 0.215 | 0.314 | 4 | 4 | 0 |
UNAL-NLP | run1 | 0.817 | 0.739 | 0.252 | 0.249 | 5 | 7 | -2 |
UNIBA | run2 | 0.784 | 0.734 | 0.255 | 0.180 | 6 | 8 | -2 |
UNIBA | run1 | 0.769 | 0.729 | 0.229 | 0.165 | 7 | 10 | -3 |
UNIBA | run3 | 0.769 | 0.729 | 0.229 | 0.165 | 8 | 11 | -3 |
BUAP | run1 | 0.805 | 0.714 | 0.162 | 0.201 | 9 | 13 | -4 |
BUAP | run2 | 0.805 | 0.714 | 0.142 | 0.194 | 10 | 9 | 1 |
Meerkat_Mafia | pairingWords | 0.794 | 0.704 | -0.044 | 0.389 | 11 | 12 | -1 |
HULTECH | run1 | 0.693 | 0.665 | 0.254 | 0.150 | 12 | 16 | -4 |
HULTECH | run3 | 0.669 | 0.671 | 0.232 | 0.137 | 13 | 15 | -2 |
RTM-DCU | run3 | 0.780 | 0.677 | 0.208 | 14 | 17 | -3 | |
HULTECH | run2 | 0.667 | 0.633 | 0.180 | 0.169 | 15 | 14 | 1 |
RTM-DCU | run1 | 0.786 | 0.666 | 0.171 | 16 | 18 | -2 | |
Meerkat_Mafia | SuperSaiyan | 0.834 | 0.777 | 17 | 19 | -2 | ||
Meerkat_Mafia | Hulk2 | 0.826 | 0.705 | 18 | 20 | -2 | ||
RTM-DCU | run2 | 0.747 | 0.588 | 0.164 | 19 | 22 | -3 | |
FBK-TR | run3 | 0.759 | 0.702 | 20 | 23 | -3 | ||
FBK-TR | run1 | 0.751 | 0.685 | 21 | 24 | -3 | ||
FBK-TR | run2 | 0.770 | 0.648 | 22 | 25 | -3 | ||
Duluth | Duluth2 | 0.501 | 0.450 | 0.241 | 0.219 | 23 | 21 | 2 |
AI-KU | run1 | 0.732 | 0.680 | 24 | 26 | -2 | ||
UNAL-NLP | run3 | 0.708 | 0.620 | 25 | 27 | -2 | ||
AI-KU | run2 | 0.698 | 0.617 | 26 | 28 | -2 | ||
TCDSCSS | run2 | 0.607 | 0.552 | 27 | 29 | -2 | ||
JU-Evora | run1 | 0.536 | 0.442 | 0.090 | 0.091 | 28 | 31 | -3 |
TCDSCSS | run1 | 0.575 | 0.541 | 29 | 30 | -1 | ||
Duluth | Duluth1 | 0.458 | 0.440 | 0.075 | 0.076 | 30 | 5 | 25 |
Duluth | Duluth3 | 0.455 | 0.426 | 0.075 | 0.079 | 31 | 3 | 28 |
OPI | run1 | 0.433 | 0.213 | 0.152 | 32 | 36 | -4 | |
SSMT | run1 | 0.789 | 33 | 34 | -1 | |||
DIT | run1 | 0.785 | 34 | 32 | 2 | |||
DIT | run2 | 0.784 | 35 | 33 | 2 | |||
UMCC_DLSI_SemSim | run1 | 0.760 | 36 | 35 | 1 | |||
UMCC_DLSI_SemSim | run2 | 0.698 | 37 | 37 | 0 | |||
UMCC_DLSI_Prob | run1 | 0.023 | 38 | 38 | 0 |
Team | System | Para-2-Sent | Sent-2-Phr | Phr-2-Word | Word-2-Sense | Official Rank | Spearman Rank | Difference |
---|---|---|---|---|---|---|---|---|
Baseline (LCS) | 0.527 | 0.562 | 0.165 | 0.109 | ||||
RTM-DCU | run1-late | 0.845 | 0.750 | 0.305 | ||||
RTM-DCU | run2-late | 0.785 | 0.698 | 0.221 | ||||
RTM-DCU | run3-late | 0.786 | 0.663 | 0.171 | ||||
Meerkat_Mafia | pairingWords (corrected) | 0.794 | 0.704 | 0.457 | 0.389 |
Spearman Correlation Results
Team | System | Para-2-Sent | Sent-2-Phr | Phr-2-Word | Word-2-Sense | sum |
---|---|---|---|---|---|---|
SimCompass | run1 | 0.801 | 0.728 | 0.424 | 0.344 | 2.297 |
ECNU | run1 | 0.821 | 0.757 | 0.306 | 0.263 | 2.147 |
Duluth | Duluth3 | 0.725 | 0.660 | 0.399 | 0.322 | 2.106 |
SemantiKLUE | run1 | 0.802 | 0.739 | 0.218 | 0.327 | 2.086 |
Duluth | Duluth1 | 0.726 | 0.658 | 0.385 | 0.311 | 2.080 |
UNAL-NLP | run2 | 0.820 | 0.710 | 0.249 | 0.236 | 2.009 |
UNAL-NLP | run1 | 0.803 | 0.717 | 0.258 | 0.231 | 2.015 |
UNIBA | run2 | 0.775 | 0.718 | 0.247 | 0.174 | 1.914 |
BUAP | run2 | 0.804 | 0.709 | 0.169 | 0.196 | 1.878 |
UNIBA | run1 | 0.762 | 0.713 | 0.221 | 0.162 | 1.858 |
UNIBA | run3 | 0.762 | 0.713 | 0.221 | 0.162 | 1.858 |
Meerkat_Mafia | pairingWords | 0.776 | 0.709 | 0.001 | 0.380 | 1.866 |
BUAP | run1 | 0.804 | 0.709 | 0.175 | 0.164 | 1.852 |
HULTECH | run2 | 0.688 | 0.633 | 0.260 | 0.124 | 1.704 |
HULTECH | run3 | 0.688 | 0.633 | 0.259 | 0.124 | 1.704 |
HULTECH | run1 | 0.666 | 0.633 | 0.260 | 0.126 | 1.685 |
RTM-DCU | run3 | 0.769 | 0.683 | 0.201 | 1.653 | |
RTM-DCU | run1 | 0.778 | 0.669 | 0.166 | 1.613 | |
Meerkat_Mafia | SuperSaiyan | 0.817 | 0.760 | 1.577 | ||
Meerkat_Mafia | Hulk2 | 0.799 | 0.726 | 1.525 | ||
Duluth | Duluth2 | 0.553 | 0.473 | 0.235 | 0.225 | 1.486 |
RTM-DCU | run2 | 0.734 | 0.581 | 0.152 | 1.467 | |
FBK-TR | run3 | 0.770 | 0.695 | 1.465 | ||
FBK-TR | run1 | 0.759 | 0.681 | 1.440 | ||
FBK-TR | run2 | 0.775 | 0.642 | 1.417 | ||
AI-KU | run1 | 0.727 | 0.646 | 1.373 | ||
UNAL-NLP | run3 | 0.708 | 0.607 | 1.315 | ||
AI-KU | run2 | 0.700 | 0.612 | 1.312 | ||
TCDSCSS | run2 | 0.642 | 0.599 | 1.241 | ||
TCDSCSS | run1 | 0.623 | 0.563 | 1.186 | ||
JU-Evora | run1 | 0.533 | 0.440 | 0.096 | 0.075 | 1.144 |
DIT | run1 | 0.777 | 0.777 | |||
DIT | run2 | 0.777 | 0.777 | |||
SSMT | run1 | 0.777 | 0.777 | |||
UMCC_DLSI_SemSim | run1 | 0.747 | 0.747 | |||
OPI | run1 | 0.424 | 0.188 | 0.131 | 0.743 | |
UMCC_DLSI_SemSim | run2 | 0.674 | 0.674 | |||
UMCC_DLSI_Prob | run1 | 0.054 | 0.054 |
Team | System | Para-2-Sent | Sent-2-Phr | Phr-2-Word | Word-2-Sense | sum |
---|---|---|---|---|---|---|
Baseline | 0.613 | 0.626 | 0.162 | 0.130 | 1.528 | |
RTM-DCU | run1-late | 0.829 | 0.734 | 0.295 | ||
RTM-DCU | run2-late | 0.778 | 0.687 | 0.219 | ||
RTM-DCU | run3-late | 0.778 | 0.667 | 0.166 | ||
Meerkat_Mafia | pairingWords (corrected) | 0.776 | 0.709 | 0.448 | 0.380 |