Download presentation
Presentation is loading. Please wait.
1
24.11.2002 The Fourth WIM Meeting 1 Active Nearest Neighbor Queries for Moving Objects Jan Kolar, Igor Timko
2
24.11.2002 The Fourth WIM Meeting 2 Outline Problem Statement System Architecture Data Model Tracking Moving Objects NNC Search & Active Result Distance between Moving Points Conclusions Proposals for Bachelor Projects
3
24.11.2002 The Fourth WIM Meeting 3 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 :
4
24.11.2002 The Fourth WIM Meeting 4 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 :
5
24.11.2002 The Fourth WIM Meeting 5 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 :
6
24.11.2002 The Fourth WIM Meeting 6 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 :
7
24.11.2002 The Fourth WIM Meeting 7 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : Distance along the roads
8
24.11.2002 The Fourth WIM Meeting 8 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? A B C
9
24.11.2002 The Fourth WIM Meeting 9 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 1 A B C
10
24.11.2002 The Fourth WIM Meeting 10 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 1 A B C
11
24.11.2002 The Fourth WIM Meeting 11 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 1 A B C
12
24.11.2002 The Fourth WIM Meeting 12 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 1 A B C
13
24.11.2002 The Fourth WIM Meeting 13 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 2 A B C
14
24.11.2002 The Fourth WIM Meeting 14 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 2 A B C
15
24.11.2002 The Fourth WIM Meeting 15 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 2 A B C
16
24.11.2002 The Fourth WIM Meeting 16 Problem Statement Road Network Copenhagen Moving Data Points Cars, pedestrians, cyclists,... Distance along the roads Query Point A shop assistant Active K-Nearest Neighbor Query Monitor 2 nearest shoppers that need a nice and cheap dress Active Query Result T 1 : T 2 : ? Time T 2 A B C
17
24.11.2002 The Fourth WIM Meeting 17 System Architecture Active Result Positioning Unit Client Position Visualization DB of Distances Road Network User Query Result NNC Search DB of Moving Points Road Network NNC Request NNC Reply NNC Refresh RN Input Position Update RN Update SERVERCLIENT
18
24.11.2002 The Fourth WIM Meeting 18 System Architecture Active Result Positioning Unit Client Position Visualization DB of Distances Road Network User Query Result NNC Search DB of Moving Points Road Network NNC Request NNC Reply NNC Refresh RN Input Position Update RN Update SERVERCLIENT
19
24.11.2002 The Fourth WIM Meeting 19 System Architecture Active Result Positioning Unit Client Position Visualization DB of Distances Road Network User Query Result NNC Search DB of Moving Points Road Network NNC Request NNC Reply NNC Refresh RN Input Position Update RN Update SERVERCLIENT
20
24.11.2002 The Fourth WIM Meeting 20 System Architecture Active Result Positioning Unit Client Position Visualization DB of Distances Road Network User Query Result NNC Search DB of Moving Points Road Network NNC Request NNC Reply NNC Refresh RN Input Position Update RN Update SERVERCLIENT
21
24.11.2002 The Fourth WIM Meeting 21 Outline Problem Statement System Architecture Data Model Tracking Moving Objects NNC Search & Active Result Distance between Moving Points Conclusions Proposals for Bachelor Projects
22
24.11.2002 The Fourth WIM Meeting 22 Data Model : Overview Problem Data Road Network (RN) Data Points (DPs) 2D Representation Captures data in native form Supports positioning and visualization Source for graph representation Graph Representation Captures data in simpler and more ”compact” form Supports algorithms for NN search
23
24.11.2002 The Fourth WIM Meeting 23 Data Model : Overview Problem Data Road Network (RN) Data Points (DPs) 2D Representation Captures data in native form Supports positioning and visualization Source for graph representation Graph Representation Captures data in simpler and more ”compact” form Supports algorithms for NN search
24
24.11.2002 The Fourth WIM Meeting 24 Data Model : Overview Problem Data Road Network (RN) Data Points (DPs) 2D Representation Captures data in native form Supports positioning and visualization Source for graph representation Graph Representation Captures data in simpler and more ”compact” form Supports algorithms for NN search
25
24.11.2002 The Fourth WIM Meeting 25 Data Model : Overview Problem Data Road Network (RN) Data Points (DPs) 2D Representation Captures data in native form Supports positioning and visualization Source for graph representation Graph Representation Captures data in simpler and more ”compact” form Supports algorithms for NN search
26
24.11.2002 The Fourth WIM Meeting 26 Data Model : Road Network 2D Graph Real-World RN Road segments 2D RN Lines approximate road segments Lines start and end at vertices Vertices have coordinates Graph RN Edges are obtained from paths Edges start and end at nodes Nodes have no coordinates Road Network
27
24.11.2002 The Fourth WIM Meeting 27 Data Model : Road Network 2D Graph Real-World RN Road segments 2D RN Lines approximate road segments Lines start and end at vertices Vertices have coordinates Graph RN Edges are obtained from paths Edges start and end at nodes Nodes have no coordinates Road Network
28
24.11.2002 The Fourth WIM Meeting 28 Data Model : Road Network Graph Real-World RN Road segments 2D RN Lines approximate road segments Lines start and end at vertices Vertices have coordinates Graph RN Edges are obtained from paths Edges start and end at nodes Nodes have no coordinates 2D Road Network
29
24.11.2002 The Fourth WIM Meeting 29 Data Model : Road Network Graph 2D Real-World RN Road segments 2D RN Lines approximate road segments Lines start and end at vertices Vertices have coordinates Graph RN Edges are obtained from paths Edges start and end at nodes Nodes have no coordinates Road Network
30
24.11.2002 The Fourth WIM Meeting 30 Data Model : RN Characteristics Graph 2D Real-World RN Road segments have length, maximum speed, and width 2D RN Lines approximate road segments Lines have length and maximum speed Lines have no width Graph RN Edges are obtained from paths Edges have edge weight Edge weight is minimal travel time along the edge – distance in graph Edge weight is calculated by combining line length and maximum speed Road Network
31
24.11.2002 The Fourth WIM Meeting 31 Data Model : RN Characteristics Real-World RN Road segments have length, maximum speed, and width 2D RN Lines approximate road segments Lines have length and maximum speed Lines have no width Graph RN Edges are obtained from paths Edges have edge weight Edge weight is minimal travel time along the edge – distance in graph Edge weight is calculated by combining line length and maximum speed Road Network Graph 2D
32
24.11.2002 The Fourth WIM Meeting 32 Data Model : RN Characteristics Real-World RN Road segments have length, maximum speed, and width 2D RN Lines approximate road segments Lines have length and maximum speed Lines have no width Graph RN Edges are obtained from paths Edges have edge weight Edge weight is minimal travel time along the edge – distance in graph Edge weight is calculated by combining line length and maximum speed Road Network Graph 2D L=10 MS=2 L=12 MS=4 L=10 MS=5
33
24.11.2002 The Fourth WIM Meeting 33 Data Model : RN Characteristics Real-World RN Road segments have length, maximum speed, and width 2D RN Lines approximate road segments Lines have length and maximum speed Lines have no width Graph RN Edges are obtained from paths Edges have edge weight Edge weight is minimal travel time along the edge – distance in graph Edge weight is calculated by combining line length and maximum speed Road Network Graph 2D W=2+3+5=10 L=10 MS=2 L=12 MS=4 L=10 MS=5
34
24.11.2002 The Fourth WIM Meeting 34 Data Model : Data Points Real-World DPs Movement of a DP is a continuous function of time 2D Road DPs A DP at a reference time is given by DP characteristics (DPC): reference time coordinate speed Road Network 2D
35
24.11.2002 The Fourth WIM Meeting 35 Data Model : Data Points Real-World DPs Movement of a DP is a continuous function of time 2D Road DPs A DP at a reference time is given by DP characteristics (DPC): reference time coordinate speed Road Network 2D C(12)=(33,60)
36
24.11.2002 The Fourth WIM Meeting 36 Data Model : Data Points Real-World DPs Movement of a DP is a continuous function of time 2D Road DPs A DP at a reference time is given by DP characteristics (DPC): reference time coordinate speed Road Network C(12)=(33,60) 2D T=11 C=(34,56) S=3
37
24.11.2002 The Fourth WIM Meeting 37 Data Model : Data Points 2D Road DPs A DP at the reference time is given by DP characteristics (DPC): reference time coordinate speed Graph DPs Movement of a DP is a function of time (positioning function) Positioning function is a combination of DPC: reference time edge initial position graph speed 2D T=11 C=(34,56) S=3 Graph T=11 E=3 IP=3 GS=3
38
24.11.2002 The Fourth WIM Meeting 38 Data Model : Data Points 2D Road DPs A DP at the reference time is given by DP characteristics (DPC): reference time coordinate speed Graph DPs Movement of a DP is a function of time (positioning function) Positioning function is a combination of DPC: reference time edge initial position graph speed 2D T=11 C=(34,56) S=3 Graph T=11 E=3 IP=3 GS=3 P(12)=3+3 = 6
39
24.11.2002 The Fourth WIM Meeting 39 Outline Problem Statement System Architecture Data Model Tracking Moving Objects NNC Search & Active Result Distance between Moving Points Conclusions Proposals for Bachelor Projects
40
24.11.2002 The Fourth WIM Meeting 40 System Architecture Active Result Positioning Unit Client Position Visualization DB of Distances Road Network User Query Result NNC Search DB of Moving Points Road Network NNC Request NNC Reply NNC Refresh RN Input Position Update RN Update SERVERCLIENT
41
24.11.2002 The Fourth WIM Meeting 41 For a DP, its Client DPC are obtained from the Positioning Unit on the Client For a DP, its Server DPC reside in the DB of Moving Points on the Server Update Policy Threshold is a maximum allowed deviation between the positions given by the Client DPC and by the Server DPC Start Node End Deviation P(S) Th P(C) Tracking Moving Points P(C)=P(S) Th P(S) Th P(C) Deviation P(C)=P(S) Th
42
24.11.2002 The Fourth WIM Meeting 42 After a DP traverses to a new edge, its Server DPC expires Update policies Update immediately after traversing to a new edge Update after the threshold is exceeded Tracking Moving Points – Passing a Node P(C) P(S) Th Deviation P(S) P(C) Deviation Th P(S) P(C) Deviation Th DPC Update P(C) Th P(S)
43
24.11.2002 The Fourth WIM Meeting 43 Outline Problem Statement Data Model System Architecture Tracking Moving Objects NNC Search & Active Result Distance between Moving Points Conclusions Proposals for Bachelor Projects
44
24.11.2002 The Fourth WIM Meeting 44 System Architecture Active Result Positioning Unit Client Position Visualization DB of Distances Road Network User Query Result NNC Search DB of Moving Points Road Network NNC Request NNC Reply NNC Refresh RN Input Position Update RN Update SERVERCLIENT
45
24.11.2002 The Fourth WIM Meeting 45 NNC Search Searches for some number of DPs that are nearest to the QP Application of the Best First Search in graphs Extended with “reading” DPs from edges During the search, all the DPs are fixed at the time when the search starts
46
24.11.2002 The Fourth WIM Meeting 46 NNC Search Searches for some number of DPs that are nearest to the QP Application of the Best First Search in graphs Extended with “reading” DPs from edges During the search, all the DPs are fixed at the time when the search starts
47
24.11.2002 The Fourth WIM Meeting 47 NNC Search Searches for some number of DPs that are nearest to the QP Application of the Best First Search in graphs Extended with “reading” DPs from edges During the search, all the DPs are fixed at the time when the search starts
48
24.11.2002 The Fourth WIM Meeting 48 NNC Search Searches for some number of DPs that are nearest to the QP Application of the Best First Search in graphs Extended with “reading” DPs from edges During the search, all the DPs are fixed at the time when the search starts
49
24.11.2002 The Fourth WIM Meeting 49 Active Result Distance between QP and NNCs Sorting NNCs with respect to the distance Estimate of imprecision of NNCs Expiration Number Distance Limit
50
24.11.2002 The Fourth WIM Meeting 50 Active Result Distance between QP and NNCs Sorting NNCs with respect to the distance Estimate of imprecision of NNCs Expiration Number Distance Limit
51
24.11.2002 The Fourth WIM Meeting 51 Active Result Distance between QP and NNCs Sorting NNCs with respect to the distance Estimate of imprecision of NNCs Expiration Number Distance Limit
52
24.11.2002 The Fourth WIM Meeting 52 Active Result Distance between QP and NNCs Sorting NNCs with respect to the distance Estimate of imprecision of NNCs Expiration Number Distance Limit
53
24.11.2002 The Fourth WIM Meeting 53 Active Result: Procedure Distance from QP13589 Expired DPfalse Distance Limit = 10 Expiration Number = 2 Number of Expired DP = 0 NNC is Valid: YES 12345 21435 Time = 1 Number of Expired DP = 1 Time = 2 1371112Distance from QP false truefalse Expired DP NNC is Valid: NO 15121613Distance from QP true false Expired DP Time = 3 Number of Expired DP = 3 21435 New NNC Request
54
24.11.2002 The Fourth WIM Meeting 54 Outline Problem Statement System Architecture Data Model Tracking Moving Objects NNC Search & Active Result Distance between Moving Points Conclusions Proposals for Bachelor Projects
55
24.11.2002 The Fourth WIM Meeting 55 Definition Distance between two DPs is the shortest path between the DPs Difficulty The shortest path between two DPs changes as the DPs move Distance between Moving Points
56
24.11.2002 The Fourth WIM Meeting 56 Definition Distance between two DPs is the shortest path between the DPs Difficulty The shortest path between two DPs changes as the DPs move Distance between Moving Points
57
24.11.2002 The Fourth WIM Meeting 57 Definition Distance between two DPs is the shortest path between the DPs Difficulty The shortest path between two DPs changes as the DPs move Distance between Moving Points
58
24.11.2002 The Fourth WIM Meeting 58 5 1 1 3 2 3 6 1 4 6 7 Distance between Moving Points Q D Q D Q D
59
24.11.2002 The Fourth WIM Meeting 59 Distance between Moving Points Requirement Find the shortest path quickly Idea DB of Distances: pre-compute shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
60
24.11.2002 The Fourth WIM Meeting 60 Distance between Moving Points Requirement Find the shortest path quickly Idea DB of Distances: pre-compute shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
61
24.11.2002 The Fourth WIM Meeting 61 Distance between Moving Points Requirement Find the shortest path quickly Idea DB of Distances: pre-compute shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
62
24.11.2002 The Fourth WIM Meeting 62 Distance between moving DP: Procedure 5 1 2 1 6 4 6 Q D A B X Y |AX| |AY| |BX| |BY| D = min
63
24.11.2002 The Fourth WIM Meeting 63 Outline Problem Statement System Architecture Data Model Tracking Moving Objects NNC Search & Active Result Distance between Moving Points Conclusions Proposals for Bachelor Projects
64
24.11.2002 The Fourth WIM Meeting 64 Conclusions Reusable data model Applicable for other NN, and non-NN problems Classical algorithm for the NNC search An extension of the Best First Search Simple idea for maintaining the active result Sort NNCs with respect to the distance to the QP Ask for new NNCs, if the current ones get too far from the QP Efficient algorithm for the distance calculation Uses the pre-computed shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
65
24.11.2002 The Fourth WIM Meeting 65 Conclusions Reusable data model Applicable for other NN, and non-NN problems Classical algorithm for the NNC search An extension of the Best First Search Simple idea for maintaining the active result Sort NNCs with respect to the distance to the QP Ask for new NNCs, if the current ones get too far from the QP Efficient algorithm for the distance calculation Uses the pre-computed shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
66
24.11.2002 The Fourth WIM Meeting 66 Conclusions Reusable data model Applicable for other NN, and non-NN problems Classical algorithm for the NNC search An extension of the Best First Search Simple idea for maintaining the active result Sort NNCs with respect to the distance to the QP Ask for new NNCs, if the current ones get too far from the QP Efficient algorithm for the distance calculation Uses the pre-computed shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
67
24.11.2002 The Fourth WIM Meeting 67 Conclusions Reusable data model Applicable for other NN, and non-NN problems Classical algorithm for the NNC search An extension of the Best First Search Simple idea for maintaining the active result Sort NNCs with respect to the distance to the QP Ask for new NNCs, if the current ones get too far from the QP Efficient algorithm for the distance calculation Uses the pre-computed shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
68
24.11.2002 The Fourth WIM Meeting 68 Conclusions Reusable data model Applicable for other NN, and non-NN problems Classical algorithm for the NNC search An extension of the Best First Search Simple idea for maintaining the active result Sort NNCs with respect to the distance to the QP Ask for new NNCs, if the current ones get too far from the QP Efficient algorithm for the distance calculation Uses the pre-computed shortest distances between each pair of nodes Reduces the distance calculation to several arithmetic operations
69
24.11.2002 The Fourth WIM Meeting 69 Conclusions Reasonable system architecture Based on the client-server architecture Distributes the tasks in an efficient way “Balanced” handling of position updates Updates are not performed continuously Threshold controls precision Prototype Single-process system that simulates the real application Experiment results show that the solutions are reasonable
70
24.11.2002 The Fourth WIM Meeting 70 Conclusions Reasonable system architecture Based on the client-server architecture Distributes the tasks in an efficient way “Balanced” handling of position updates Updates are not performed continuously Threshold controls precision Prototype Single-process system that simulates the real application Experiment results show that the solutions are reasonable
71
24.11.2002 The Fourth WIM Meeting 71 Conclusions Reasonable system architecture Based on the client-server architecture Distributes the tasks in an efficient way “Balanced” handling of position updates Updates are not performed continuously Threshold controls precision Prototype Single-process system that simulates the real application Experiment results show that the solutions are reasonable
72
24.11.2002 The Fourth WIM Meeting 72 Conclusions Reasonable system architecture Based on the client-server architecture Distributes the tasks in an efficient way “Balanced” handling of position updates Updates are not performed continuously Threshold controls precision Prototype Single-process system that simulates the real application Experiment results show that the solutions are reasonable
73
24.11.2002 The Fourth WIM Meeting 73 Outline Problem Statement Data Model NNC Search & Active Result Distance between Moving Points System Architecture Conclusions Proposals for Bachelor Projects
74
24.11.2002 The Fourth WIM Meeting 74 Proposals for Bachelor Projects Implementation, experiments, and improvements NNC search Active result Distance calculation Complete architecture Extending the settings: influence on the algorithms, the architecture, and the data model “Richer” model Uncertainty of Query Results Pre-Defined Routes Dynamic Weights
75
24.11.2002 The Fourth WIM Meeting 75 Proposals for Bachelor Projects Implementation, experiments, and improvements NNC search Active result Distance calculation Complete architecture Extending the settings: influence on the algorithms, the architecture, and the data model “Richer” model Uncertainty of Query Results Pre-Defined Routes Dynamic Weights
76
24.11.2002 The Fourth WIM Meeting 76 Proposals for Bachelor Projects Implementation, experiments, and improvements NNC search Active result Distance calculation Complete architecture Extending the settings: influence on the algorithms, the architecture, and the data model “Richer” model Uncertainty of Query Results Pre-Defined Routes Dynamic Weights
77
24.11.2002 The Fourth WIM Meeting 77 Active Nearest Neighbor Queries for Moving Objects Jan Kolar, Igor Timko
78
24.11.2002 The Fourth WIM Meeting 78 Uncertainty in the NN problem Igor Timko
79
24.11.2002 The Fourth WIM Meeting 79 Outline Uncertainty Handling Location Uncertainty Conclusions
80
24.11.2002 The Fourth WIM Meeting 80 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
81
24.11.2002 The Fourth WIM Meeting 81 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
82
24.11.2002 The Fourth WIM Meeting 82 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
83
24.11.2002 The Fourth WIM Meeting 83 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
84
24.11.2002 The Fourth WIM Meeting 84 Outline Uncertainty Handling Location Uncertainty Conclusions
85
24.11.2002 The Fourth WIM Meeting 85 Handling Location Uncertainty Old active result procedure Obtain NNCs Calculate distances between the NNCs and the QP Sort the NNCs Identifying the uncertainty Calculated distances are uncertain, because locations of NNCs are uncertain
86
24.11.2002 The Fourth WIM Meeting 86 Handling Location Uncertainty Old active result procedure Obtain NNCs Calculate distances between the NNCs and the QP Sort the NNCs Identifying the uncertainty Calculated distances are uncertain, because locations of NNCs are uncertain
87
24.11.2002 The Fourth WIM Meeting 87 Handling Location Uncertainty Old active result procedure Obtain NNCs Calculate distances between the NNCs and the QP Sort the NNCs Identifying the uncertainty Calculated distances are uncertain, because locations of NNCs are uncertain
88
24.11.2002 The Fourth WIM Meeting 88 Measuring the Uncertainty Bounded normal distribution Mean is at the distance value Deviation is the update threshhold D1D1 D2D2 D3D3 Th
89
24.11.2002 The Fourth WIM Meeting 89 New Active Result Procedure New active result procedure Obtain NNCs For each NNC calculate distances between it and the QP construct the probability distribution calculate the probability of being NN
90
24.11.2002 The Fourth WIM Meeting 90 New Active Result Procedure New active result procedure Obtain NNCs For each NNC calculate distances between it and the QP construct the probability distribution calculate the probability of being NN
91
24.11.2002 The Fourth WIM Meeting 91 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
92
24.11.2002 The Fourth WIM Meeting 92 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
93
24.11.2002 The Fourth WIM Meeting 93 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
94
24.11.2002 The Fourth WIM Meeting 94 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
95
24.11.2002 The Fourth WIM Meeting 95 Uncertainty in the NN problem Igor Timko
96
24.11.2002 The Fourth WIM Meeting 96 Outline Uncertainty Handling Location Uncertainty Conclusions
97
24.11.2002 The Fourth WIM Meeting 97 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
98
24.11.2002 The Fourth WIM Meeting 98 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
99
24.11.2002 The Fourth WIM Meeting 99 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
100
24.11.2002 The Fourth WIM Meeting 100 Uncertainty Sources of the uncertainty Location of DPs NNCs Dynamic weights Partial NNC search Partial DB of distances Communication network Problem with the uncertainty Imprecise query result Handling the uncertainty Calculate the probabilistic NN neighbor
101
24.11.2002 The Fourth WIM Meeting 101 Outline Uncertainty Handling Location Uncertainty Conclusions
102
24.11.2002 The Fourth WIM Meeting 102 Handling Location Uncertainty Old active result procedure Obtain NNCs Calculate distances between the NNCs and the QP Sort the NNCs Identifying the uncertainty Calculated distances are uncertain, because locations of NNCs are uncertain
103
24.11.2002 The Fourth WIM Meeting 103 Handling Location Uncertainty Old active result procedure Obtain NNCs Calculate distances between the NNCs and the QP Sort the NNCs Identifying the uncertainty Calculated distances are uncertain, because locations of NNCs are uncertain
104
24.11.2002 The Fourth WIM Meeting 104 Handling Location Uncertainty Old active result procedure Obtain NNCs Calculate distances between the NNCs and the QP Sort the NNCs Identifying the uncertainty Calculated distances are uncertain, because locations of NNCs are uncertain
105
24.11.2002 The Fourth WIM Meeting 105 Measuring the Uncertainty Bounded normal distribution Mean is at the distance value Deviation is the update threshhold D1D1 D2D2 D3D3 Th
106
24.11.2002 The Fourth WIM Meeting 106 Measuring the Uncertainty Bounded normal distribution Mean is at the distance value Deviation is the update threshhold D1D1 D2D2 D3D3 Th
107
24.11.2002 The Fourth WIM Meeting 107 New Active Result Procedure New active result procedure Obtain NNCs For each NNC calculate distances between it and the QP construct the probability distribution calculate the probability of being NN
108
24.11.2002 The Fourth WIM Meeting 108 New Active Result Procedure New active result procedure Obtain NNCs For each NNC calculate distances between it and the QP construct the probability distribution calculate the probability of being NN
109
24.11.2002 The Fourth WIM Meeting 109 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
110
24.11.2002 The Fourth WIM Meeting 110 Outline Uncertainty Handling Location Uncertainty Conclusions
111
24.11.2002 The Fourth WIM Meeting 111 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
112
24.11.2002 The Fourth WIM Meeting 112 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
113
24.11.2002 The Fourth WIM Meeting 113 Conclusions There are many sources of the uncertainty in the NN problem The uncertainty makes the NN query result imprecise The uncertainty is handled by the probabilistic NN queries
114
24.11.2002 The Fourth WIM Meeting 114 Uncertainty in the NN problem Igor Timko
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.