Introduction 3
Results for CLEF 2006 Ad-Hoc Robust Tracks List of Submitted Experiments 7 colesir daedalus daedalus daedalus daedalus hummingbird hummingbird colesir daedalus daedalus daedalus daedalus dcu dcu hildesheim hildesheim hummingbird hummingbird reina reina colesir daedalus daedalus daedalus daedalus dcu dcu hummingbird hummingbird reina reina colesir daedalus daedalus daedalus daedalus dcu dcu dcu hildesheim hildesheim hildesheim hummingbird hummingbird reina reina unine unine unine colesir daedalus daedalus daedalus daedalus dcu dcu
AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC AUTOMATIC Track Overview Results and Graphs 11 0.5 −1 0% 0% 10% 20% 30% dex011deRFW3FS3 [MAP 38.63%; Not Pooled] s t n e irmdex021deRFW3FS3 [MAP 38.31%; Not Pooled] e p x E
deFSdeR3S [MAP 38.03%; Not Pooled] CoLesIRdeTst [MAP 37.21%; Not Pooled] deFSdeR2S [MAP 34.06%; Not Pooled] 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Mean Average Precision
Ad−Hoc Robust Monolingual German track − Tukey T test with "top group" highlighted humDE06Rtde
deFSdeR3S CoLesIRdeTst −1 1 0.5 0.5 −1 hummingbird [humDE06Rtde; R−Prec 46.34%; Not Pooled] colesir [CoLesIRdeTst; R−Prec 36.49%; Not Pooled] daedalus [deFSdeR2S; R−Prec 34.80%; Not Pooled]
Ad−Hoc Robust Monolingual German track − Box Plot of the Topics humDE06Rtde [R−Prec 46.34%; Not Pooled] humDE06Rtd [R−Prec 41.68%; Not Pooled] dex011deRFW3FS3 [R−Prec 38.76%; Not Pooled] s t n e irmdex021deRFW3FS3 [R−Prec 38.15%; Not Pooled] e p x E
deFSdeR3S [R−Prec 37.76%; Not Pooled] CoLesIRdeTst [R−Prec 36.49%; Not Pooled] deFSdeR2S [R−Prec 34.80%; Not Pooled] 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
R−Precision
Ad−Hoc Robust Monolingual German track − Tukey T test with "top group" highlighted
deFSdeR3S CoLesIRdeTst 100,000 5,008 3,266
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 1 0.5 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) esFSesRLit2S Docs Cutoff Levels Precision at DCL (%) 5 docs 34.20 10 docs 30.30 15 docs 28.07 20 docs 25.60 30 docs 23.13 100 docs 14.87 200 docs 10.22 500 docs 5.52 1000 docs 3.27 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 28.75 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 10
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 0% 0% 10% 20% 30%
Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) esFSesRLit3S Docs Cutoff Levels Precision at DCL (%) 5 docs 39.20 10 docs 35.00 15 docs 32.13 20 docs 29.55 30 docs 26.23 100 docs 16.09 200 docs 10.89 500 docs 6.03 1000 docs 3.50 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 31.00 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 10 15 20 Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment Precision averages (%) for individual queries
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision
Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment Precision averages (%) for individual queries 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) −1 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085
Topic Identifier
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 086 to 130) −1 086 087 088 089 090 091 092 093 094 095 096 097 098 099 110 111 112 113 114 115 116 117 118 119 130
Topic Identifier
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 131 to 165) −1 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 160 161 162 163 164 165
Topic Identifier
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 166 to 200) esx021esRLitFW3FS3 esx021esRLitFW3FS3 Docs Cutoff Levels Precision at DCL (%) 5 docs 38.00 10 docs 34.70 15 docs 32.27 20 docs 29.50 30 docs 26.20 100 docs 16.09 200 docs 10.89 500 docs 6.03 1000 docs 3.50 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 31.39 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 10 15 20 Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment Precision averages (%) for individual queries
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision dcu
Query Construction Source Language Topic Fields Pooled 1 AUTOMATIC Italian title, description, narrative false spanish bilingual standard 0% 0% 10% 20% 30%
Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Mean Average Precision Precision averages (%) for individual queries dcuitqynar sp1005 dcuitqynar sp1005 dcuitqynar sp1005 dcu Docs Cutoff Levels Precision at DCL (%) 5 docs 42.00 10 docs 40.10 15 docs 36.87 20 docs 33.45 30 docs 29.13 100 docs 17.63 200 docs 11.90 500 docs 6.60 1000 docs 3.86 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 34.17 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 10 15 20 Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment Precision averages (%) for individual queries
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision dcuitqynar sp1005 dcuitqynar sp1005 dcuitqynar sp1005 dcuitqynar sp1005
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Exact R−Precision by Topic (Topics 041 to 085) dcu
Query Construction Source Language Topic Fields Pooled 2 AUTOMATIC Italian title, description false spanish bilingual with desc+title 0% 0% 10% 20% 30% 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Mean Average Precision Precision averages (%) for individual queries
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) dcuitqydescsp12075 dcuitqydescsp12075 dcu Docs Cutoff Levels Precision at DCL (%) 5 docs 41.60 10 docs 37.40 15 docs 35.13 20 docs 31.85 30 docs 27.90 100 docs 17.09 200 docs 11.57 500 docs 6.40 1000 docs 3.76 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 33.55 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 10 15 20 Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers Precision averages (%) for individual queries
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision reina MT + local expansion with term windows, title+desc fields Precision averages (%) for individual queries 20 t n e m irpe15 x E e h ftso10 cp i o T f o re 5 b m u N 0% 0% 10% 20% 30% 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) reina Docs Cutoff Levels Precision at DCL (%) 5 docs 44.40 10 docs 40.80 15 docs 37.93 20 docs 34.70 30 docs 30.80 100 docs 19.29 200 docs 13.00 500 docs 6.81 1000 docs 3.86 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 36.73 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 10 15 20 Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers Precision averages (%) for individual queries 25
Ad−Hoc Robust Bilingual Spanish track − Retrieved documents vs Precision reinaIT2EStdtest
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision reina 100,000 5,008 3,529 MT + local expansion with term windows, only title field Precision averages (%) for individual queries 40 ten35 m irpe30 x E ftsohe2205 cp i foT15 o re10 b m u 5 N 0% 0% 10% 20% 30% 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) reina Docs Cutoff Levels Precision at DCL (%) 5 docs 38.00 10 docs 33.60 15 docs 29.87 20 docs 28.90 30 docs 25.90 100 docs 16.94 200 docs 11.37 500 docs 6.18 1000 docs 3.53 R-Precision (precision after R document retrieved, where R = Relevant retrieved) 31.64 100% 90% 80% 70% 60% 40% 30% 20% 10% 0%5 30 100 200 Retrieved Documents (logarithmic scale) Ad−Hoc Robust Bilingual Spanish track − Box plot of the Topics of the Experiment Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers Precision averages (%) for individual queries 30
Ad−Hoc Robust Bilingual Spanish track − Retrieved documents vs Precision reinaIT2ESttest 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision
Ad−Hoc Robust Bilingual Spanish track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision reinaIT2ESt est reinaIT2ESt est reinaIT2ESt est reinaIT2ESt est
Ad−Hoc Robust Bilingual Spanish track − Comparison to Median Mean Exact R−Precision by Topic (Topics 041 to 085) Overall statistics for 100 queries : Total number of documents over all queries Retrieved Relevant Relevant retrieved 100,000 2,717 2,084 Precision averages (%) for individual queries 0% 0% 10% 20% 30% 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment 1 0.5 ce reen 0 if D−0.5 0.5 ce reen 0 if D−0.5 1 1 0.5 ce reen 0 if D−0.5 1 0.5 ce reen 0 if D−0.5 25 t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Dutch track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) nlFSnlRLfr2S nlFSnlRLfr2S 60% 40% 30% 20% 10% 0%5
Ad−Hoc Robust Bilingual Dutch track − Retrieved documents vs Precision 30 100 200 Retrieved Documents (logarithmic scale) Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers Precision averages (%) for individual queries 25 t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision
Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision Overall statistics for 100 queries : Total number of documents over all queries Retrieved Relevant Relevant retrieved 1 AUTOMATIC French title, description, narrative false Precision averages (%) for individual queries 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085 ce reen 0 if D−0.5 ce reen 0 if D−0.5 25 t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 0% 0% 10% 20% 30% 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Dutch track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) nlFSnlRLfr6S nlFSnlRLfr6S 60% 40% 30% 20% 10% 0%5
Ad−Hoc Robust Bilingual Dutch track − Retrieved documents vs Precision 30 100 200 Retrieved Documents (logarithmic scale) Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers Precision averages (%) for individual queries 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085 ce reen 0 if D−0.5 ce reen 0 if D−0.5 1 0.5 ce reen 0 if D−0.5 30 t n iem25 r e xp E20 e h tfso15 cp i o fT10 o r e b m 5 u N 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision
Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision Overall statistics for 100 queries : Total number of documents over all queries Retrieved Relevant Relevant retrieved 3 AUTOMATIC French title, description, narrative false 0% 0% 10% 20% 30% 40% 50% 60%
Interpolated Recall 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment Precision averages (%) for individual queries 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085 ce reen 0 if D−0.5 ce reen 0 if D−0.5 1 0.5 ce reen 0 if D−0.5 25 t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Dutch track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) nlx021nlRLfrFW4FS6 nlx021nlRLfrFW4FS6 60% 40% 30% 20% 10% 0%5
Ad−Hoc Robust Bilingual Dutch track − Retrieved documents vs Precision 30 100 200 Retrieved Documents (logarithmic scale) Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers Precision averages (%) for individual queries 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085 ce reen 0 if D−0.5 0.5 ce reen 0 if D−0.5 1 1 0.5 ce reen 0 if D−0.5 1 0.5 ce reen 0 if D−0.5 25 t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Exact R−Precision
Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100% Exact R−Precision Overall statistics for 100 queries : Total number of documents over all queries Retrieved Relevant Relevant retrieved 2 AUTOMATIC French title, description, narrative false 20% 10% 0% 0% 10% 20% 30% 40% 50% 60%
Interpolated Recall 70% 80% 90% 100%
Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision Ad−Hoc Robust Bilingual Dutch track − Distribution of the Topics of the Experiment Precision averages (%) for individual queries 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085 ce reen 0 if D−0.5 0.5 ce reen 0 if D−0.5 1 1 0.5 ce reen 0 if D−0.5 1 0.5 ce reen 0 if D−0.5 25 t n e irm20 e xp E the15 f so c iop10 T f o r e b 5 m u N 00% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% 70% 75% 80% 85% 90% 95% 100%
Mean Average Precision
Ad−Hoc Robust Bilingual Dutch track − Comparison to Median Mean Average Precision by Topic (Topics 041 to 085) nlx011nlRLfrFW4FS6 nlx011nlRLfrFW4FS6 100% 90% 80% 70% 60% 40% Ad−Hoc Robust Bilingual Dutch track − Box plot of the Topics of the Experiment Exact R-Precision Maximum Minimum First Quartile Second Quartile Third Quartile Interquartile range Mean Standard Deviation Lower Outlier Threshold Upper Outlier Threshold Mean With No Outliers Std With No Outliers 041 042 043 044 045 046 047 048 049 060 061 062 063 064 065 066 067 068 069 080 081 082 083 084 085 1
Ad−Hoc Robust Bilingual Dutch track − Comparison to Median Mean Exact R−Precision by Topic (Topics 041 to 085)