Searching for similar entities within a database is a common and a daily, billions of times, perf... more Searching for similar entities within a database is a common and a daily, billions of times, performed task. Generally, similarities are calculated using common distance measures like Manhattan, Euclidean, Levenshtein, Mahalanobis or Dynamic Time Warping (DTW). In this paper, we present a similarity measure for time interval data, which allows searching for similar sets of time interval records bounded by a time window (e.g., a day, a week, or a month). We introduce three different groups of distance measures i.e., temporal order, temporal measure, and temporal relation distances. In addition, we present bitmap-based implementations for algorithms of each of the three types. We designed our solutions to perform well on large datasets and support distributed calculations. Evaluations show the out-standing performance regarding other interval related similarity measures, i.e., ARTEMIS and IBSM.
On-line analytical processing is in the focus of research over the last couple decades. Several p... more On-line analytical processing is in the focus of research over the last couple decades. Several papers dealing with summarizability problems, cube computations, query languages, fact-dimension relationships or different types of hierarchies have been published. Nowadays, analyzing time interval data became ubiquitous. Nevertheless, the use of established, reliable, and proven technologies like OLAP is desirable in this respect.
In this paper, we present an OLAP system capable to process time interval data. The system is based on bitmaps, enabling performant selection and fast aggregation. Moreover, we introduce a two-step aggregation technique, which enables the calculation of relevant measures in the context of time interval data. We evaluate the performance of our system using different bitmap implementations and a real-world data set. To our knowledge, there are no other systems available enabling OLAP and providing correct results considering the summarizability of time interval data.
Nowadays, time interval data is ubiquitous. The requirement of analyzing such data using known te... more Nowadays, time interval data is ubiquitous. The requirement of analyzing such data using known techniques like on-line analytical processing arises more and more frequently. Nevertheless, the usage of approved multi-dimensional models and established systems is not sufficient, because of modeling, querying and processing limitations. Even though recent research and requests from various types of industry indicate that the handling and analyzing of time interval data is an important task, a definition of a query language to enable on-line analytical processing and a suitable implementation are, to the best of our knowledge, neither introduced nor realized. In this paper, we present a query language based on requirements stated by business analysts from different domains that enables the analysis of time interval data in an on-line analytical manner. In addition, we introduce our query
processing, established using a bitmap-based and interval-tree based implementation. Finally, we present a performance analysis and discuss the language, the processing as well as the results critically.
Searching for similar entities within a database is a common and a daily, billions of times, perf... more Searching for similar entities within a database is a common and a daily, billions of times, performed task. Generally, similarities are calculated using common distance measures like Manhattan, Euclidean, Levenshtein, Mahalanobis or Dynamic Time Warping (DTW). In this paper, we present a similarity measure for time interval data, which allows searching for similar sets of time interval records bounded by a time window (e.g., a day, a week, or a month). We introduce three different groups of distance measures i.e., temporal order, temporal measure, and temporal relation distances. In addition, we present bitmap-based implementations for algorithms of each of the three types. We designed our solutions to perform well on large datasets and support distributed calculations. Evaluations show the out-standing performance regarding other interval related similarity measures, i.e., ARTEMIS and IBSM.
On-line analytical processing is in the focus of research over the last couple decades. Several p... more On-line analytical processing is in the focus of research over the last couple decades. Several papers dealing with summarizability problems, cube computations, query languages, fact-dimension relationships or different types of hierarchies have been published. Nowadays, analyzing time interval data became ubiquitous. Nevertheless, the use of established, reliable, and proven technologies like OLAP is desirable in this respect.
In this paper, we present an OLAP system capable to process time interval data. The system is based on bitmaps, enabling performant selection and fast aggregation. Moreover, we introduce a two-step aggregation technique, which enables the calculation of relevant measures in the context of time interval data. We evaluate the performance of our system using different bitmap implementations and a real-world data set. To our knowledge, there are no other systems available enabling OLAP and providing correct results considering the summarizability of time interval data.
Nowadays, time interval data is ubiquitous. The requirement of analyzing such data using known te... more Nowadays, time interval data is ubiquitous. The requirement of analyzing such data using known techniques like on-line analytical processing arises more and more frequently. Nevertheless, the usage of approved multi-dimensional models and established systems is not sufficient, because of modeling, querying and processing limitations. Even though recent research and requests from various types of industry indicate that the handling and analyzing of time interval data is an important task, a definition of a query language to enable on-line analytical processing and a suitable implementation are, to the best of our knowledge, neither introduced nor realized. In this paper, we present a query language based on requirements stated by business analysts from different domains that enables the analysis of time interval data in an on-line analytical manner. In addition, we introduce our query
processing, established using a bitmap-based and interval-tree based implementation. Finally, we present a performance analysis and discuss the language, the processing as well as the results critically.
Uploads
Papers by Diane Keng
In this paper, we present an OLAP system capable to process time interval data. The system is based on bitmaps, enabling performant selection and fast aggregation. Moreover, we introduce a two-step aggregation technique, which enables the calculation of relevant measures in the context of time interval data. We evaluate the performance of our system using different bitmap implementations and a real-world data set. To our knowledge, there are no other systems available enabling OLAP and providing correct results considering the summarizability of time interval data.
processing, established using a bitmap-based and interval-tree based implementation. Finally, we present a performance analysis and discuss the language, the processing as well as the results critically.
In this paper, we present an OLAP system capable to process time interval data. The system is based on bitmaps, enabling performant selection and fast aggregation. Moreover, we introduce a two-step aggregation technique, which enables the calculation of relevant measures in the context of time interval data. We evaluate the performance of our system using different bitmap implementations and a real-world data set. To our knowledge, there are no other systems available enabling OLAP and providing correct results considering the summarizability of time interval data.
processing, established using a bitmap-based and interval-tree based implementation. Finally, we present a performance analysis and discuss the language, the processing as well as the results critically.