2017 IEEE 31st International Conference on Advanced Information Networking and Applications (AINA), 2017
This paper proposes a k-nearest neighbour search method inspired by grid space partitioning and t... more This paper proposes a k-nearest neighbour search method inspired by grid space partitioning and the compact-trie structure. A compact trie structure, and a k-nearest neighbour search strategy are presented. Then, a k-nearest neighbour search performance comparison is carried out against two well-known methods, using one million two-dimensional spatial points and finding up to 1000 nearest neighbours. The results of the comparison shows that the proposed compact-trie method can perform up to 300 times better when k is small, and up to 25 times better for larger k values, suggesting that the proposed method is suitable for low dimensions and location-dependent spatial queries in applications such as mobile computing.
Kathy Crewdson, Ian Dew NextLibrary Inc. Wendy Osborn, Steve Fox, Attila Aros Southern Alberta Di... more Kathy Crewdson, Ian Dew NextLibrary Inc. Wendy Osborn, Steve Fox, Attila Aros Southern Alberta Digital Library University of Lethbridge The Southern Alberta Information Resources (SAIR) Project [1,2] is an online bibliography of resources significant to southern Alberta. SAIR contains bibliographic information on the region and its communities such as Lethbridge, Claresholm, Waterton and Medicine Hat, environmental topics such as water and land, economy, and international border issues, among others. SAIR is a collaborative effort between NextLibrary Inc. [4] and the Southern Alberta Digital Library (SADL) [5] at the University of Lethbridge. NextLibrary Inc. is a consulting company that provides digitization solutions. Their mandate is to promote lifelong learning through digitization. SADL is a site for developing online bibliographic and multimedia digital libraries. It runs Greenstone [3], which is open-source, and provides functionality for creating and customizing a collection...
In this paper, we propose the mqr-tree, a two-dimensional spatial access method that organizes sp... more In this paper, we propose the mqr-tree, a two-dimensional spatial access method that organizes spatial objects in a two-dimensional node and based on their spatial relationships. Previously proposed spatial access methods that attempt to maintain spatial relationships between objects in their structures are limited in their incorporation of existing one-dimensional spatial access methods, or have lower space utilization in its nodes, and higher tree height, overcoverage and overlap than is necessary. The mqr-tree utilizes a node organization, set of spatial relationship rules and insertion strategy in order to gain significant improvements in overlap and overcoverage. In addition, other desirable properties are identified as a result of the chosen node organization and insertion strategies. In particular, zero overlap is achieved when the mqr-tree is used to index point data. A comparison of the mqr-tree insertion strategy versus the R-tree shows significant improvements in overlap ...
This paper proposes several strategies for continuous k-nearest neighbour query processing in loc... more This paper proposes several strategies for continuous k-nearest neighbour query processing in location-based services. All utilize a clustered dataset to create safe regions. The only previously proposed strategy to do so has limited validation of the safe region on the client, which leads to only approximate k-nearest neighbour results being obtained. This work improves upon both the creation of the safe region and the validation of the safe region on the client. An evaluation of our strategy, including a comparison versus the existing strategy, show significant improvements in accuracy, and up to over 90% accuracy. When compared to repeated k-nearest neighbour search, our strategy is computationally significantly faster for larger datasets. Obtaining significantly high accuracy at low computational costs is significant for our strategies.
Strategies for Alternate Group Trip Planning Queries in Location-Based Services
This paper explores the concept of an Alternate Group Trip Planning Query for Location Based Serv... more This paper explores the concept of an Alternate Group Trip Planning Query for Location Based Services. A Sequenced Group Trip Planning Query is a set of users’ start and destination locations, and a sequence of Categories of Interest for a trip they wish to travel together. The outcome is a trip through a sequence of Points of Interest that minimizes the travel distance for all. Previously proposed strategies for processing Sequenced Group Trip Planning Queries do not consider other alternative orderings of the sequence of Categories of Interest, in order to find a shorter trip. In this paper, we propose three strategies for processing Alternate Group Trip Planning Queries - Permutation, Random, and Greedy. Experimental results show that improvements in trip travel distance are possible with our strategies, with some accomplishing it in a short running time.
The 2DR-tree is a novel approach for accessing spatial data, particularly when spatial relationsh... more The 2DR-tree is a novel approach for accessing spatial data, particularly when spatial relationships are defined among objects. The 2DR-tree uses nodes that are the same dimensionality as the data space. Therefore, all relationships between objects are preserved and different binary search strategies are supported. This paper presents the 2DR-tree binary search strategy. A preliminary performance evaluation identifies the advantages of the binary search strategy.
Database system design for energy economization of smart home energy services
This paper proposes the practical implementation of a balanced electrical supply among the grid, ... more This paper proposes the practical implementation of a balanced electrical supply among the grid, residential units and shared facility controllers for designing future smart home energy services. To achieve energy economization, the project involves design and analysis of a database management system incorporated with different web applications. The main purpose for using these technologies is to inspect, measure and control the energy flow within a specific area. For condition monitoring and long distance work controlling, PostgreSQL version 9.3 for database management and a cloud based platform for website support have been used. The main goal of this project is to process a program by developing a database management system and bring it to the users accessible levels.
Exploring Bit Arrays for Join Processing in Spatial Data Streams
In this paper, the use of bit arrays for processing spatial joins in spatial data streams is expl... more In this paper, the use of bit arrays for processing spatial joins in spatial data streams is explored. Although spatial joins between objects have been explored in other contexts, such as centralized and distributed systems, they have not been explored in great detail in spatial data streams. This work explores the use of a Bloom-filter (i.e., bit array) inspired representation of a spatial object. Strategies for both mapping objects to bit arrays, and processing spatial joins using the bit arrays in a data stream environment are presented. The strategies are evaluated and compared with spatial (non-bit) join approaches. Performance improvements are identified, and areas of improvement are also identified.
Nowadays, many mobile applications provide location-based services that allow users to access loc... more Nowadays, many mobile applications provide location-based services that allow users to access location-related information from anywhere, whenever they desire. A moving user can issue queries to access information about moving or static objects. Continuous spatial query processing systems are used for this type of application. We propose two continuous query processing strategies moving point sets. The objectives of our strategies are to reduce: (1) the server workload, (2) the data transmission cost and (3) the query response time, for location-based services while providing an answer for a continuous region query. We compare our strategies with a brute-force strategy. Results show that our strategies can significantly reduce the server workload, data transmission cost, and response time over the brute-force method in many cases.
Nowadays, data management and processing systems exist that handle streaming data. Streaming data... more Nowadays, data management and processing systems exist that handle streaming data. Streaming data is data that is very large and cannot all be stored, and also may only be valid for a certain period of time. In existing stream data managements systems, data streams are transmitted to a server where the management and processing of queries take place, before the results are transmitted on an outbound stream possibly to another mobile device. However, no strategies exist that receive, manage and process stream data directly on a mobile device, without having to rely on the services of a server. This paper proposes an architecture and framework for stream data management and query processing on a mobile device. The architecture only keeps a minimal number of features. All database operations are handled with minimal data storage requirements. It also presents details on an implementation of this framework for a currency conversion application, which demonstrates the utility of our fram...
A k-nearest Neighbour Query Processing Strategy Using the mqr-tree
This paper presents a k-nearest neighbour query processing strategy that utilizes a recently prop... more This paper presents a k-nearest neighbour query processing strategy that utilizes a recently proposed spatial access method, the mqr-tree, to narrow down the search for candidate nearest neighbours. It requires the traversal of only one path of the tree, and in the majority of cases, does not require backtracking all the way to the root. The mqr-tree-based k-nearest neighbour strategy is evaluated both individually and compared against the brute-force method for running time. It is shown that the proposed approach performs better in many cases. In addition, the running time is not affected by the number of points being indexed or the number of nearest neighbour being sought.
Proceedings of the 1st ACM SIGSPATIAL International Workshop on Animal Movement Ecology and Human Mobility
Measuring characteristics of animals in the wild is not always possible, due to their demeanour a... more Measuring characteristics of animals in the wild is not always possible, due to their demeanour and lack of human contact. Remote capture and processing methods, including the segmentation of animal data into relevant body parts, are required. Existing solutions are either costly or too cumbersome to use in the wild. This study explores the use of RGB depth (RGB-D) cameras for data capture of a target animal from a distance. In addition, this study explores the extraction and segmentation of the resulting animal data into point clouds, and the creation of machine learning models for the automated segmentation of this data. Results of this study, including an experimental evaluation, demonstrate the feasibility of utilizing RGB-D cameras for animal data capture, and that classification outperformed clustering for automated animal data segmentation.
A spatial access method approach to continuous k-nearest neighbour processing for location-based services
International Journal of Grid and Utility Computing
In this paper, two strategies for handling continuous k-nearest neighbour queries for location-ba... more In this paper, two strategies for handling continuous k-nearest neighbour queries for location-based services are proposed. CKNN1 and CKNN2 utilise a validity (i.e., safe) region approach for minimising the number of query requests that need to be send to the server. They also use a two-dimensional spatial access method for both validity region selection and in-structure searching. The latter feature ensures that new searches for a validity region are not required to begin from the root. An evaluation and comparison of both strategies is performed against repeated nearest neighbour search. Both random and exponentially distributed point sets are utilised. Results show that both approaches achieve significant performance gains, especially with respect to reducing the number of queries that must be sent from the client to the server.
Proceedings of the 17th International Conference on Enterprise Information Systems, 2015
This paper presents an evaluation of the mqr-tree for indexing a database containing a very large... more This paper presents an evaluation of the mqr-tree for indexing a database containing a very large number of objects. Many spatial access methods have been proposed for handling either point and/or region data, with the vast majority able to handle a limited number of instances of these data types efficiently. However, many established and emerging application areas, such as recommender systems, require the management and indexing of very large object sets, such as a million places of interest that are each represented with a point. Using between one and five million points and objects, a comparison of both index construction and spatial query evaluation is performed versus a benchmark spatial indexing strategy. We show that the mqr-tree achieves significantly lower overlap and overcoverage when used to index a very large collection of objects. Also, the mqr-tree achieves significantly improved query processing performance in many cases. Therefore, the mqr-tree is a significant candidate for handling very large object sets for emerging applications.
In this paper, we present a novel strategy for distributed spatial query optimization that involv... more In this paper, we present a novel strategy for distributed spatial query optimization that involves multiple sites. Most previous work in the area of distributed spatial query processing and optimization focuses only on strategies for performing spatial joins and spatial semijoins, and distributed spatial queries that only involve two sites. We propose a new strategy, called the Restricted strategy, for optimizing a distributed spatial query. It uses spatial semijoins and can involve any number of sites in a distributed spatial database. The Restricted strategy improves upon an existing strategy by sending group approximations, instead of sending approximations for all objects, in order to reduce the number of comparisons between objects and thereby minimize the CPU and data transmission cost. A performance evaluation of our strategy finds that it significantly minimizes the number of data comparisons and CPU time of distributed spatial queries.
21st International Conference on Advanced Information Networking and Applications Workshops (AINAW'07), 2007
This paper explores the use of spatial indices for the modeling and retrieval of protein structur... more This paper explores the use of spatial indices for the modeling and retrieval of protein structures. With two existing spatial indices, a preliminary framework for protein structure modeling that uses a spatial index is proposed. It provides a novel technique for modeling. In addition, it provides additional flexibility with respect to modeling granularity and structure manipulation. It is expected that this modeling approach will lead to new ways of analyzing protein structures.
2014 28th International Conference on Advanced Information Networking and Applications Workshops, 2014
This paper explores the issue of caching during TIPtree navigation for location-based services. T... more This paper explores the issue of caching during TIPtree navigation for location-based services. The TIP-tree is a recently proposed spatial access method for supporting continuous point queries in a tourist information system. In addition, its navigation strategy avoids repeating searching, thus providing efficient processing of user trajectories that are not known in advance. This paper proposes a caching strategy that further improves upon the efficiency of the TIP-tree navigation algorithm. A performance evaluation shows that significant improvement is possible when navigation and caching are applied together.
2017 IEEE 31st International Conference on Advanced Information Networking and Applications (AINA), 2017
This paper proposes a k-nearest neighbour search method inspired by grid space partitioning and t... more This paper proposes a k-nearest neighbour search method inspired by grid space partitioning and the compact-trie structure. A compact trie structure, and a k-nearest neighbour search strategy are presented. Then, a k-nearest neighbour search performance comparison is carried out against two well-known methods, using one million two-dimensional spatial points and finding up to 1000 nearest neighbours. The results of the comparison shows that the proposed compact-trie method can perform up to 300 times better when k is small, and up to 25 times better for larger k values, suggesting that the proposed method is suitable for low dimensions and location-dependent spatial queries in applications such as mobile computing.
Kathy Crewdson, Ian Dew NextLibrary Inc. Wendy Osborn, Steve Fox, Attila Aros Southern Alberta Di... more Kathy Crewdson, Ian Dew NextLibrary Inc. Wendy Osborn, Steve Fox, Attila Aros Southern Alberta Digital Library University of Lethbridge The Southern Alberta Information Resources (SAIR) Project [1,2] is an online bibliography of resources significant to southern Alberta. SAIR contains bibliographic information on the region and its communities such as Lethbridge, Claresholm, Waterton and Medicine Hat, environmental topics such as water and land, economy, and international border issues, among others. SAIR is a collaborative effort between NextLibrary Inc. [4] and the Southern Alberta Digital Library (SADL) [5] at the University of Lethbridge. NextLibrary Inc. is a consulting company that provides digitization solutions. Their mandate is to promote lifelong learning through digitization. SADL is a site for developing online bibliographic and multimedia digital libraries. It runs Greenstone [3], which is open-source, and provides functionality for creating and customizing a collection...
In this paper, we propose the mqr-tree, a two-dimensional spatial access method that organizes sp... more In this paper, we propose the mqr-tree, a two-dimensional spatial access method that organizes spatial objects in a two-dimensional node and based on their spatial relationships. Previously proposed spatial access methods that attempt to maintain spatial relationships between objects in their structures are limited in their incorporation of existing one-dimensional spatial access methods, or have lower space utilization in its nodes, and higher tree height, overcoverage and overlap than is necessary. The mqr-tree utilizes a node organization, set of spatial relationship rules and insertion strategy in order to gain significant improvements in overlap and overcoverage. In addition, other desirable properties are identified as a result of the chosen node organization and insertion strategies. In particular, zero overlap is achieved when the mqr-tree is used to index point data. A comparison of the mqr-tree insertion strategy versus the R-tree shows significant improvements in overlap ...
This paper proposes several strategies for continuous k-nearest neighbour query processing in loc... more This paper proposes several strategies for continuous k-nearest neighbour query processing in location-based services. All utilize a clustered dataset to create safe regions. The only previously proposed strategy to do so has limited validation of the safe region on the client, which leads to only approximate k-nearest neighbour results being obtained. This work improves upon both the creation of the safe region and the validation of the safe region on the client. An evaluation of our strategy, including a comparison versus the existing strategy, show significant improvements in accuracy, and up to over 90% accuracy. When compared to repeated k-nearest neighbour search, our strategy is computationally significantly faster for larger datasets. Obtaining significantly high accuracy at low computational costs is significant for our strategies.
Strategies for Alternate Group Trip Planning Queries in Location-Based Services
This paper explores the concept of an Alternate Group Trip Planning Query for Location Based Serv... more This paper explores the concept of an Alternate Group Trip Planning Query for Location Based Services. A Sequenced Group Trip Planning Query is a set of users’ start and destination locations, and a sequence of Categories of Interest for a trip they wish to travel together. The outcome is a trip through a sequence of Points of Interest that minimizes the travel distance for all. Previously proposed strategies for processing Sequenced Group Trip Planning Queries do not consider other alternative orderings of the sequence of Categories of Interest, in order to find a shorter trip. In this paper, we propose three strategies for processing Alternate Group Trip Planning Queries - Permutation, Random, and Greedy. Experimental results show that improvements in trip travel distance are possible with our strategies, with some accomplishing it in a short running time.
The 2DR-tree is a novel approach for accessing spatial data, particularly when spatial relationsh... more The 2DR-tree is a novel approach for accessing spatial data, particularly when spatial relationships are defined among objects. The 2DR-tree uses nodes that are the same dimensionality as the data space. Therefore, all relationships between objects are preserved and different binary search strategies are supported. This paper presents the 2DR-tree binary search strategy. A preliminary performance evaluation identifies the advantages of the binary search strategy.
Database system design for energy economization of smart home energy services
This paper proposes the practical implementation of a balanced electrical supply among the grid, ... more This paper proposes the practical implementation of a balanced electrical supply among the grid, residential units and shared facility controllers for designing future smart home energy services. To achieve energy economization, the project involves design and analysis of a database management system incorporated with different web applications. The main purpose for using these technologies is to inspect, measure and control the energy flow within a specific area. For condition monitoring and long distance work controlling, PostgreSQL version 9.3 for database management and a cloud based platform for website support have been used. The main goal of this project is to process a program by developing a database management system and bring it to the users accessible levels.
Exploring Bit Arrays for Join Processing in Spatial Data Streams
In this paper, the use of bit arrays for processing spatial joins in spatial data streams is expl... more In this paper, the use of bit arrays for processing spatial joins in spatial data streams is explored. Although spatial joins between objects have been explored in other contexts, such as centralized and distributed systems, they have not been explored in great detail in spatial data streams. This work explores the use of a Bloom-filter (i.e., bit array) inspired representation of a spatial object. Strategies for both mapping objects to bit arrays, and processing spatial joins using the bit arrays in a data stream environment are presented. The strategies are evaluated and compared with spatial (non-bit) join approaches. Performance improvements are identified, and areas of improvement are also identified.
Nowadays, many mobile applications provide location-based services that allow users to access loc... more Nowadays, many mobile applications provide location-based services that allow users to access location-related information from anywhere, whenever they desire. A moving user can issue queries to access information about moving or static objects. Continuous spatial query processing systems are used for this type of application. We propose two continuous query processing strategies moving point sets. The objectives of our strategies are to reduce: (1) the server workload, (2) the data transmission cost and (3) the query response time, for location-based services while providing an answer for a continuous region query. We compare our strategies with a brute-force strategy. Results show that our strategies can significantly reduce the server workload, data transmission cost, and response time over the brute-force method in many cases.
Nowadays, data management and processing systems exist that handle streaming data. Streaming data... more Nowadays, data management and processing systems exist that handle streaming data. Streaming data is data that is very large and cannot all be stored, and also may only be valid for a certain period of time. In existing stream data managements systems, data streams are transmitted to a server where the management and processing of queries take place, before the results are transmitted on an outbound stream possibly to another mobile device. However, no strategies exist that receive, manage and process stream data directly on a mobile device, without having to rely on the services of a server. This paper proposes an architecture and framework for stream data management and query processing on a mobile device. The architecture only keeps a minimal number of features. All database operations are handled with minimal data storage requirements. It also presents details on an implementation of this framework for a currency conversion application, which demonstrates the utility of our fram...
A k-nearest Neighbour Query Processing Strategy Using the mqr-tree
This paper presents a k-nearest neighbour query processing strategy that utilizes a recently prop... more This paper presents a k-nearest neighbour query processing strategy that utilizes a recently proposed spatial access method, the mqr-tree, to narrow down the search for candidate nearest neighbours. It requires the traversal of only one path of the tree, and in the majority of cases, does not require backtracking all the way to the root. The mqr-tree-based k-nearest neighbour strategy is evaluated both individually and compared against the brute-force method for running time. It is shown that the proposed approach performs better in many cases. In addition, the running time is not affected by the number of points being indexed or the number of nearest neighbour being sought.
Proceedings of the 1st ACM SIGSPATIAL International Workshop on Animal Movement Ecology and Human Mobility
Measuring characteristics of animals in the wild is not always possible, due to their demeanour a... more Measuring characteristics of animals in the wild is not always possible, due to their demeanour and lack of human contact. Remote capture and processing methods, including the segmentation of animal data into relevant body parts, are required. Existing solutions are either costly or too cumbersome to use in the wild. This study explores the use of RGB depth (RGB-D) cameras for data capture of a target animal from a distance. In addition, this study explores the extraction and segmentation of the resulting animal data into point clouds, and the creation of machine learning models for the automated segmentation of this data. Results of this study, including an experimental evaluation, demonstrate the feasibility of utilizing RGB-D cameras for animal data capture, and that classification outperformed clustering for automated animal data segmentation.
A spatial access method approach to continuous k-nearest neighbour processing for location-based services
International Journal of Grid and Utility Computing
In this paper, two strategies for handling continuous k-nearest neighbour queries for location-ba... more In this paper, two strategies for handling continuous k-nearest neighbour queries for location-based services are proposed. CKNN1 and CKNN2 utilise a validity (i.e., safe) region approach for minimising the number of query requests that need to be send to the server. They also use a two-dimensional spatial access method for both validity region selection and in-structure searching. The latter feature ensures that new searches for a validity region are not required to begin from the root. An evaluation and comparison of both strategies is performed against repeated nearest neighbour search. Both random and exponentially distributed point sets are utilised. Results show that both approaches achieve significant performance gains, especially with respect to reducing the number of queries that must be sent from the client to the server.
Proceedings of the 17th International Conference on Enterprise Information Systems, 2015
This paper presents an evaluation of the mqr-tree for indexing a database containing a very large... more This paper presents an evaluation of the mqr-tree for indexing a database containing a very large number of objects. Many spatial access methods have been proposed for handling either point and/or region data, with the vast majority able to handle a limited number of instances of these data types efficiently. However, many established and emerging application areas, such as recommender systems, require the management and indexing of very large object sets, such as a million places of interest that are each represented with a point. Using between one and five million points and objects, a comparison of both index construction and spatial query evaluation is performed versus a benchmark spatial indexing strategy. We show that the mqr-tree achieves significantly lower overlap and overcoverage when used to index a very large collection of objects. Also, the mqr-tree achieves significantly improved query processing performance in many cases. Therefore, the mqr-tree is a significant candidate for handling very large object sets for emerging applications.
In this paper, we present a novel strategy for distributed spatial query optimization that involv... more In this paper, we present a novel strategy for distributed spatial query optimization that involves multiple sites. Most previous work in the area of distributed spatial query processing and optimization focuses only on strategies for performing spatial joins and spatial semijoins, and distributed spatial queries that only involve two sites. We propose a new strategy, called the Restricted strategy, for optimizing a distributed spatial query. It uses spatial semijoins and can involve any number of sites in a distributed spatial database. The Restricted strategy improves upon an existing strategy by sending group approximations, instead of sending approximations for all objects, in order to reduce the number of comparisons between objects and thereby minimize the CPU and data transmission cost. A performance evaluation of our strategy finds that it significantly minimizes the number of data comparisons and CPU time of distributed spatial queries.
21st International Conference on Advanced Information Networking and Applications Workshops (AINAW'07), 2007
This paper explores the use of spatial indices for the modeling and retrieval of protein structur... more This paper explores the use of spatial indices for the modeling and retrieval of protein structures. With two existing spatial indices, a preliminary framework for protein structure modeling that uses a spatial index is proposed. It provides a novel technique for modeling. In addition, it provides additional flexibility with respect to modeling granularity and structure manipulation. It is expected that this modeling approach will lead to new ways of analyzing protein structures.
2014 28th International Conference on Advanced Information Networking and Applications Workshops, 2014
This paper explores the issue of caching during TIPtree navigation for location-based services. T... more This paper explores the issue of caching during TIPtree navigation for location-based services. The TIP-tree is a recently proposed spatial access method for supporting continuous point queries in a tourist information system. In addition, its navigation strategy avoids repeating searching, thus providing efficient processing of user trajectories that are not known in advance. This paper proposes a caching strategy that further improves upon the efficiency of the TIP-tree navigation algorithm. A performance evaluation shows that significant improvement is possible when navigation and caching are applied together.
Uploads
Papers by Wendy Osborn