Top Queries on Uncertain Data On Score Distribution and Typical Answers Tingjian Ge Computer Science Department Brown University Providence RI USA tigecs
127K - views

Top Queries on Uncertain Data On Score Distribution and Typical Answers Tingjian Ge Computer Science Department Brown University Providence RI USA tigecs

brownedu Stan Zdonik Computer Science Department Brown University Providence RI USA sbzcsbrownedu Samuel Madden CSAIL MIT Cambridge MA USA maddencsailmitedu 03 125 11 19 1050 T7 05 58 9 25 1050 T6 10 56 12 7 1049 T5 03 80 10 19 1050 T4 04 110 9 25 10

Download Pdf

Top Queries on Uncertain Data On Score Distribution and Typical Answers Tingjian Ge Computer Science Department Brown University Providence RI USA tigecs




Download Pdf - The PPT/PDF document "Top Queries on Uncertain Data On Score D..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.



Presentation on theme: "Top Queries on Uncertain Data On Score Distribution and Typical Answers Tingjian Ge Computer Science Department Brown University Providence RI USA tigecs"— Presentation transcript:


Page 1
Top- Queries on Uncertain Data: On Score Distribution and Typical Answers Tingjian Ge Computer Science Department Brown University Providence, RI, USA tige@cs.brown.edu Stan Zdonik Computer Science Department Brown University Providence, RI, USA sbz@cs.brown.edu Samuel Madden CSAIL MIT Cambridge, MA, USA madden@csail.mit.edu 0.3 125 (11, 19) 10:50 T7 0.5 58 (9, 25) 10:50 T6 1.0 56 (12, 7) 10:49 T5 0.3 80 (10, 19) 10:50 T4 0.4 110 (9, 25) 10:51 T3 0.4 60 (10, 19) 10:49 T2 0.4 49 (10, 20) 10:50 T1 Conf. Score for Medical Needs Location Time Soldier ID Tuple ID 0.3 125 (11, 19)

10:50 T7 0.5 58 (9, 25) 10:50 T6 1.0 56 (12, 7) 10:49 T5 0.3 80 (10, 19) 10:50 T4 0.4 110 (9, 25) 10:51 T3 0.4 60 (10, 19) 10:49 T2 0.4 49 (10, 20) 10:50 T1 Conf. Score for Medical Needs Location Time Soldier ID Tuple ID
Page 2
100 120 140 160 180 200 220 240 0.05 0.1 0.15 0.2 Top-2 total scores Probability U-Topk returns this line only T4, T6 0.06 9 = {T1, T4, T5, T6} T3, T4 0.072 = {T3, T4, T5} T3, T4 0.048 W7 = {T1, T3, T4, T5} T2, T5 0.024 = {T2, T5} T2, T5 0.016 W5 = {T1, T2, T5} T2, T6 0.12 = {T2, T5, T6} T2, T6 0.08 3 = {T1, T2, T5, T6} T3, T2 0.096 = {T2, T3, T5} T3, T2 0.064

1 = {T1, T2, T3, T5} Top-2 Prob. ossible world W8 W6 W4 W2 T4, T6 0.06 9 = {T1, T4, T5, T6} T3, T4 0.072 = {T3, T4, T5} T3, T4 0.048 W7 = {T1, T3, T4, T5} T2, T5 0.024 = {T2, T5} T2, T5 0.016 W5 = {T1, T2, T5} T2, T6 0.12 = {T2, T5, T6} T2, T6 0.08 3 = {T1, T2, T5, T6} T3, T2 0.096 = {T2, T3, T5} T3, T2 0.064 1 = {T1, T2, T3, T5} Top-2 Prob. ossible world W8 W6 W4 W2 T7, T5 0.018 W18 = {T5, T7} T7, T5 0.012 W17 = {T1, T5, T7} T7, T6 0.09 W16 = {T5, T6, T7} T7, T6 0.06 W15 = {T1, T5, T6, T7} T7, T3 0.072 W14 = {T3, T5, T7} T7, T3 0.048 W13 = {T1, T3, T5, T7} T4, T5 0.018 W12 = {T4, T5} T4, T5

0.012 W11 = {T1, T4, T5} T4, T6 0.09 W10 = {T4, T5, T6} Top-2 Prob. Possible world T7, T5 0.018 W18 = {T5, T7} T7, T5 0.012 W17 = {T1, T5, T7} T7, T6 0.09 W16 = {T5, T6, T7} T7, T6 0.06 W15 = {T1, T5, T6, T7} T7, T3 0.072 W14 = {T3, T5, T7} T7, T3 0.048 W13 = {T1, T3, T5, T7} T4, T5 0.018 W12 = {T4, T5} T4, T5 0.012 W11 = {T1, T4, T5} T4, T6 0.09 W10 = {T4, T5, T6} Top-2 Prob. Possible world
Page 3
|u |
Page 4
d
Page 5
k WW t t StateExpansion
Page 6
T1 T2 Ti Ti+1 Tn kk-1 j-1 j1 ss Tn-k T1 T2 Ti Ti+1 Tn kk-1 j-1 j1 ss Tn-k
Page 7


Page 8
T1 T2 Ti kk-1 j-1 j1 Lead tuple region (0, 1) (0, 0) auxiliary column block exit points enable exit points T1 T2 Ti kk-1 j-1 j1 Lead tuple region (0, 1) (0, 0) auxiliary column block exit points enable exit points
Page 9
dd d   t
Page 10
dd d
Page 11
50 100 150 200 250 300 350 400 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 Congestion scores of top-5 Probability U-Topk 3-Typical 50 100 150 200 250 300 350 400 45 0.005 0.01 0.015 0.02 0.025 0.03 Congestion scores of top-5 Probability U-Topk 3-Typical 150 200 250 300

350 400 450 500 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 Congestion scores of top-10 Probability U-Topk 3-Typical 10 20 30 40 50 60 50 100 150 200 250 scan depth (n) 10 20 30 40 50 60 10 -1 10 10 10 10 Execution time (seconds) Main algorithm StateExpansion k-Combo 0.1 0.2 0.3 0.4 0.5 0.5 1.5 2.5 ME tuple portion Execution time (seconds) 100 200 300 400 500 Maximum number of lines Execution time (seconds)
Page 12
800 1000 1200 1400 1600 1800 0.02 0.04 0.06 0.08 0.1 0.12 0.14 Scores of top-k vectors Probability 800 1000 1200 1400 1600 1800 0.02 0.04 0.06 0.08 0.1 0.12 Scores

of top-k vectors Probability 800 1000 1200 1400 1600 1800 0.02 0.04 0.06 0.08 0.1 Scores of top-k vectors Probability U-Topk 3-Typical U-Topk 3-Typical U-Topk 3-Typical 0.12 0.09 0.08 800 1000 1200 1400 1600 1800 2000 2200 0.02 0.04 0.06 0.08 0.1 Scores of top-k vectors Probability 1000 1100 1200 1300 1400 1500 1600 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 Scores of top-k vectors Probability 200 400 600 800 1000 1200 1400 0.01 0.02 0.03 0.04 0.05 0.06 0.07 Scores of top-k vectors Probability U-Topk 3-Typical U-Topk 3-Typical 3-Typical U-Topk
Page 13