Tatsuhiro Sakai, Keiichi Tamura, H. Kitakami, T. Takezawa
Users on social networking services (SNSs) have been transmitting information about events they witnessed themselves in their daily life through geo-social data as geo-tagged texts and photos. Geo-social data are usually related to not only personal topics but also local topics and events. Therefore, extracting local topics and events in geo-social data is one of the most important challenges in many application domains. In this study, to extract local topics in geo-social data, we propose a new method based on a density-based multimodal spatial clustering algorithm called the (ϵ, σ)-density-based multimodal spatial clustering, which can extract multimodal spatial clusters that are spatially and semantically separated from other spatial clusters. Moreover, to present the main topics of each multimodal spatial cluster, representative photos are detected using network-based importance analysis. The proposed method utilizes a pre-trained deep network for extracting feature vectors of photos, and feature vectors are utilized to calculate the similarity between two geo-social data. To evaluate our new local topic extraction method, we conducted experiments using actual geo-tagged tweets that include photos. The experimental results show that the proposed method can extract local topics as multimodal spatial clusters more sensitively than our previous method.
{"title":"Density-based Multimodal Spatial Clustering using Pre-trained Deep Network for Extracting Local Topics","authors":"Tatsuhiro Sakai, Keiichi Tamura, H. Kitakami, T. Takezawa","doi":"10.1145/3210272.3210274","DOIUrl":"https://doi.org/10.1145/3210272.3210274","url":null,"abstract":"Users on social networking services (SNSs) have been transmitting information about events they witnessed themselves in their daily life through geo-social data as geo-tagged texts and photos. Geo-social data are usually related to not only personal topics but also local topics and events. Therefore, extracting local topics and events in geo-social data is one of the most important challenges in many application domains. In this study, to extract local topics in geo-social data, we propose a new method based on a density-based multimodal spatial clustering algorithm called the (ϵ, σ)-density-based multimodal spatial clustering, which can extract multimodal spatial clusters that are spatially and semantically separated from other spatial clusters. Moreover, to present the main topics of each multimodal spatial cluster, representative photos are detected using network-based importance analysis. The proposed method utilizes a pre-trained deep network for extracting feature vectors of photos, and feature vectors are utilized to calculate the similarity between two geo-social data. To evaluate our new local topic extraction method, we conducted experiments using actual geo-tagged tweets that include photos. The experimental results show that the proposed method can extract local topics as multimodal spatial clusters more sensitively than our previous method.","PeriodicalId":106620,"journal":{"name":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","volume":"43 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116643658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A range of GPS, social network and transportation applications have been developed, targetting to improve the quality of life of the user. Furthermore, the development of smart devices allows the user to use the applications any time, while also providing the location of the user. As a result, a range of datasets of different nature has been created, describing events that are related to the location. Regardless the great volume of these datasets, their different nature (i.e. schema) deters the analysts from combining the datasets, losing insights of a location that could be important. In this study, we propose a framework that targets to achieve a knowledge fusion by connecting datasets of different nature. In order to achieve the fusion, we initially transform the datasets into graph bases. Afterwards, we import the graph bases into a knowledge base represented as Heterogeneous Information Network (HIN), using the location as the main node type that connects the datasets. This knowledge base provides to the user a bigger picture of the real world, is able to connect information across domains that initially seemed unconnected and provides a semantically rich data basis that is useful to answer many types of questions.
{"title":"Data Fusion of Diverse Data Sources: Enrich Spatial Data Knowledge Using HINs","authors":"Hardik Patel, P. Paraskevopoulos, M. Renz","doi":"10.1145/3210272.3210275","DOIUrl":"https://doi.org/10.1145/3210272.3210275","url":null,"abstract":"A range of GPS, social network and transportation applications have been developed, targetting to improve the quality of life of the user. Furthermore, the development of smart devices allows the user to use the applications any time, while also providing the location of the user. As a result, a range of datasets of different nature has been created, describing events that are related to the location. Regardless the great volume of these datasets, their different nature (i.e. schema) deters the analysts from combining the datasets, losing insights of a location that could be important. In this study, we propose a framework that targets to achieve a knowledge fusion by connecting datasets of different nature. In order to achieve the fusion, we initially transform the datasets into graph bases. Afterwards, we import the graph bases into a knowledge base represented as Heterogeneous Information Network (HIN), using the location as the main node type that connects the datasets. This knowledge base provides to the user a bigger picture of the real world, is able to connect information across domains that initially seemed unconnected and provides a semantically rich data basis that is useful to answer many types of questions.","PeriodicalId":106620,"journal":{"name":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127423655","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the advent of reliable positioning technologies and prevalence of location-based services, it is now feasible to accurately study the propagation of items such as infectious viruses, sensitive information pieces, and malwares through a population of moving objects, e.g., individuals, vehicles, and mobile devices. In such application scenarios, an item passes between two objects when the objects are sufficiently close (i.e., when they are, so-called, in contact), and hence once an item is initiated, it can propagate in the object population through the evolving network of contacts among objects, termed contact network. In this paper, for the first time we define and study probabilistic reachability queries in large uncertain contact networks, where propagation of items through contacts are uncertain. A probabilistic reachability query verifies whether two objects are "reachable" through the evolving uncertain contact network with a probability greater than a threshold η. For efficient processing of probabilistic queries, we propose a novel index structure, termed spatiotemporal tree cover (STC), which leverages the spatiotemporal properties of the contact network for efficient processing of the queries. Our experiments with real data demonstrate superiority of our proposed solution versus the only other existing solution (based on Monte Carlo sampling) for processing probabilistic reachability queries in generic uncertain graphs, with 300% improvement in query processing time on average.
{"title":"Reach Me If You Can: Reachability Query in Uncertain Contact Networks","authors":"Zohreh Raghebi, F. Kashani","doi":"10.1145/3210272.3210276","DOIUrl":"https://doi.org/10.1145/3210272.3210276","url":null,"abstract":"With the advent of reliable positioning technologies and prevalence of location-based services, it is now feasible to accurately study the propagation of items such as infectious viruses, sensitive information pieces, and malwares through a population of moving objects, e.g., individuals, vehicles, and mobile devices. In such application scenarios, an item passes between two objects when the objects are sufficiently close (i.e., when they are, so-called, in contact), and hence once an item is initiated, it can propagate in the object population through the evolving network of contacts among objects, termed contact network. In this paper, for the first time we define and study probabilistic reachability queries in large uncertain contact networks, where propagation of items through contacts are uncertain. A probabilistic reachability query verifies whether two objects are \"reachable\" through the evolving uncertain contact network with a probability greater than a threshold η. For efficient processing of probabilistic queries, we propose a novel index structure, termed spatiotemporal tree cover (STC), which leverages the spatiotemporal properties of the contact network for efficient processing of the queries. Our experiments with real data demonstrate superiority of our proposed solution versus the only other existing solution (based on Monte Carlo sampling) for processing probabilistic reachability queries in generic uncertain graphs, with 300% improvement in query processing time on average.","PeriodicalId":106620,"journal":{"name":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131164117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this seminar, we will explore how processing rich spatial data is not the only practical (and research-wise promising) application domain for traditional spatial database techniques. An equally promising direction, possibly with low-hanging fruits for research innovation, may be to apply the spatial data management expertise of our community to non-spatial types of queries, and to extend standard, more theoretical operators to large scale datasets with the objective of practical solutions (as opposed to favorable asymptotic complexity alone). As a case study, we will review spatial database work on top-k-related operators (i.e., non-spatial problems) and how it integrates fundamental computational geometric operators with spatial indexing/pruning to produce efficient solutions to practical problems.
{"title":"Applying Spatial Database Techniques to Other Domains: a Case Study on Top-k and Computational Geometric Operators","authors":"K. Mouratidis","doi":"10.1145/3210272.3226094","DOIUrl":"https://doi.org/10.1145/3210272.3226094","url":null,"abstract":"In this seminar, we will explore how processing rich spatial data is not the only practical (and research-wise promising) application domain for traditional spatial database techniques. An equally promising direction, possibly with low-hanging fruits for research innovation, may be to apply the spatial data management expertise of our community to non-spatial types of queries, and to extend standard, more theoretical operators to large scale datasets with the objective of practical solutions (as opposed to favorable asymptotic complexity alone). As a case study, we will review spatial database work on top-k-related operators (i.e., non-spatial problems) and how it integrates fundamental computational geometric operators with spatial indexing/pruning to produce efficient solutions to practical problems.","PeriodicalId":106620,"journal":{"name":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","volume":"427 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133267039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Snowdon, Olga Gkountouna, Andreas Züfle, D. Pfoser
Effective road traffic assessment and estimation is crucial not only for traffic management applications, but also for long-term transportation and, more generally, urban planning. Traditionally, this task has been achieved by using a network of stationary traffic count sensors. These costly and unreliable sensors have been replaced with so-called Probe Vehicle Data (PVD), which relies on sampling individual vehicles in traffic using for example smartphones to assess the overall traffic condition. While PVD provides uniform road network coverage, it does not capture the actual traffic flow. On the other hand, stationary sensors capture the absolute traffic flow only at discrete locations. Furthermore, these sensors are often unreliable; temporary malfunctions create gaps in their time-series of measurements. This work bridges the gap between these two data sources by learning the time-dependent fraction of vehicles captured by GPS-based probe data at discrete stationary sensor locations. We can then account for the gaps of the traffic-loop measurements by using the PVD data to estimate the actual total flow. In this work, we show that the PVD flow capture changes significantly over time in the Washington DC area. Exploiting this information, we are able to derive tight confidence intervals of the traffic volume for areas with no stationary sensor coverage.
{"title":"Spatiotemporal Traffic Volume Estimation Model Based on GPS Samples","authors":"J. Snowdon, Olga Gkountouna, Andreas Züfle, D. Pfoser","doi":"10.1145/3210272.3210273","DOIUrl":"https://doi.org/10.1145/3210272.3210273","url":null,"abstract":"Effective road traffic assessment and estimation is crucial not only for traffic management applications, but also for long-term transportation and, more generally, urban planning. Traditionally, this task has been achieved by using a network of stationary traffic count sensors. These costly and unreliable sensors have been replaced with so-called Probe Vehicle Data (PVD), which relies on sampling individual vehicles in traffic using for example smartphones to assess the overall traffic condition. While PVD provides uniform road network coverage, it does not capture the actual traffic flow. On the other hand, stationary sensors capture the absolute traffic flow only at discrete locations. Furthermore, these sensors are often unreliable; temporary malfunctions create gaps in their time-series of measurements. This work bridges the gap between these two data sources by learning the time-dependent fraction of vehicles captured by GPS-based probe data at discrete stationary sensor locations. We can then account for the gaps of the traffic-loop measurements by using the PVD data to estimate the actual total flow. In this work, we show that the PVD flow capture changes significantly over time in the Washington DC area. Exploiting this information, we are able to derive tight confidence intervals of the traffic volume for areas with no stationary sensor coverage.","PeriodicalId":106620,"journal":{"name":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","volume":"688 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133167617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aim of GeoRich is to provide a unique forum for discussing in depth the challenges, opportunities, novel techniques and applications on modeling, managing, searching and mining rich geo-spatial data, in order to fuel scientific research on big spatial data applications beyond the current research frontiers. The workshop is intended to bring together researchers from different fields of data-science and geoinformation-science that deal with the management of spatial and spatio-temporal data, social network data, textual data, multimedia data, semantic data and ontologies, uncertain data and other common types of geo-referenced data. The focus of the third GeoRich workshop is to analyze what has been achieved so far and how to further exploit the enormous potential of this data flood. This workshop brought together researchers from the fields of databases, data-science and geoinformation-science, who independently work on similar problems, but often apply different techniques to solve these problems. Focus of this workshop is to create synergies for databases, data-science and geoinformation-science, by sharing ideas and finding common solutions.
{"title":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","authors":"Andreas Züfle, B. Adams, Dingming Wu","doi":"10.1145/3210272","DOIUrl":"https://doi.org/10.1145/3210272","url":null,"abstract":"The aim of GeoRich is to provide a unique forum for discussing in depth the challenges, opportunities, novel techniques and applications on modeling, managing, searching and mining rich geo-spatial data, in order to fuel scientific research on big spatial data applications beyond the current research frontiers. The workshop is intended to bring together researchers from different fields of data-science and geoinformation-science that deal with the management of spatial and spatio-temporal data, social network data, textual data, multimedia data, semantic data and ontologies, uncertain data and other common types of geo-referenced data. The focus of the third GeoRich workshop is to analyze what has been achieved so far and how to further exploit the enormous potential of this data flood. This workshop brought together researchers from the fields of databases, data-science and geoinformation-science, who independently work on similar problems, but often apply different techniques to solve these problems. Focus of this workshop is to create synergies for databases, data-science and geoinformation-science, by sharing ideas and finding common solutions.","PeriodicalId":106620,"journal":{"name":"Proceedings of the Fifth International ACM SIGMOD Workshop on Managing and Mining Enriched Geo-Spatial Data","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133832037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：