Raphaella Carvalho Diniz, Pedro O. S. Vaz de Melo, R. Assunção
We show that the usual evaluation metrics used in machine learning are not appropriate to measure the performance of spatial disease cluster detection algorithms. We demonstrate that the usual recall and precision metrics give a distorted evaluation of the algorithms. To solve this problem, we propose new metrics based on probability predictive rules. We evaluate the performance of the main spatial disease cluster algorithms with these new metrics. Our analysis and experiments offer insights into when the usual metrics are not appropriate and also show that our proposal is very effective at eliminating the bias from the usual metrics.
{"title":"Evaluating the Evaluation Metrics for Spatial Disease Cluster Detection Algorithms","authors":"Raphaella Carvalho Diniz, Pedro O. S. Vaz de Melo, R. Assunção","doi":"10.1145/3397536.3422251","DOIUrl":"https://doi.org/10.1145/3397536.3422251","url":null,"abstract":"We show that the usual evaluation metrics used in machine learning are not appropriate to measure the performance of spatial disease cluster detection algorithms. We demonstrate that the usual recall and precision metrics give a distorted evaluation of the algorithms. To solve this problem, we propose new metrics based on probability predictive rules. We evaluate the performance of the main spatial disease cluster algorithms with these new metrics. Our analysis and experiments offer insights into when the usual metrics are not appropriate and also show that our proposal is very effective at eliminating the bias from the usual metrics.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115166603","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}
Ratna Mandal, Prasenjit Karmakar, Abhijit Roy, Arpan Saha, S. Chatterjee, Sandip Chakraborty, Sujoy Saha, S. Nandi
Public city bus services across various developing cities inhabit multiple stay-locations on the routes due to ad-hoc bus stops to provide on-demand passenger boarding and alighting services. Characterizing these stay-locations is essential to correctly develop models for bus transit patterns used in various digital navigation services. In this poster, we create a deep learning-driven methodology to characterize ad-hoc stay-locations over bus routes based on crowd-sensing contextual information. Experiments over 720km of bus travel data in a semi-urban city in India indicate promising results from the model in terms of good detection accuracy.
{"title":"Ad-hocBusPoI: Context Analysis of Ad-hoc Stay-locations from Intra-city Bus Mobility and Smartphone Crowdsensing","authors":"Ratna Mandal, Prasenjit Karmakar, Abhijit Roy, Arpan Saha, S. Chatterjee, Sandip Chakraborty, Sujoy Saha, S. Nandi","doi":"10.1145/3397536.3422273","DOIUrl":"https://doi.org/10.1145/3397536.3422273","url":null,"abstract":"Public city bus services across various developing cities inhabit multiple stay-locations on the routes due to ad-hoc bus stops to provide on-demand passenger boarding and alighting services. Characterizing these stay-locations is essential to correctly develop models for bus transit patterns used in various digital navigation services. In this poster, we create a deep learning-driven methodology to characterize ad-hoc stay-locations over bus routes based on crowd-sensing contextual information. Experiments over 720km of bus travel data in a semi-urban city in India indicate promising results from the model in terms of good detection accuracy.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123437237","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}
Menghai Pan, Weixiao Huang, Yanhua Li, Xun Zhou, Zhenming Liu, Jie Bao, Yu Zheng, Jun Luo
Learning to make optimal decisions is a common yet complicated task. While computer agents can learn to make decisions by running reinforcement learning (RL), it remains unclear how human beings learn. In this paper, we perform the first data-driven case study on taxi drivers to validate whether humans mimic RL to learn. We categorize drivers into three groups based on their performance trends and analyze the correlations between human drivers and agents trained using RL. We discover that drivers that become more efficient at earning over time exhibit similar learning patterns to those of agents, whereas drivers that become less efficient tend to do the opposite. Our study (1) provides evidence that some human drivers do adapt RL when learning, (2) enhances the deep understanding of taxi drivers' learning strategies, (3) offers a guideline for taxi drivers to improve their earnings, and (4) develops a generic analytical framework to study and validate human learning strategies.
{"title":"Is Reinforcement Learning the Choice of Human Learners?: A Case Study of Taxi Drivers","authors":"Menghai Pan, Weixiao Huang, Yanhua Li, Xun Zhou, Zhenming Liu, Jie Bao, Yu Zheng, Jun Luo","doi":"10.1145/3397536.3422246","DOIUrl":"https://doi.org/10.1145/3397536.3422246","url":null,"abstract":"Learning to make optimal decisions is a common yet complicated task. While computer agents can learn to make decisions by running reinforcement learning (RL), it remains unclear how human beings learn. In this paper, we perform the first data-driven case study on taxi drivers to validate whether humans mimic RL to learn. We categorize drivers into three groups based on their performance trends and analyze the correlations between human drivers and agents trained using RL. We discover that drivers that become more efficient at earning over time exhibit similar learning patterns to those of agents, whereas drivers that become less efficient tend to do the opposite. Our study (1) provides evidence that some human drivers do adapt RL when learning, (2) enhances the deep understanding of taxi drivers' learning strategies, (3) offers a guideline for taxi drivers to improve their earnings, and (4) develops a generic analytical framework to study and validate human learning strategies.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115108759","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}
Ayush Bandil, Vaishali Girdhar, K. Dinçer, Harsh Govind, Peiwei Cao, Ashley Song, Mohamed H. Ali
All online map service providers are working hard to maintain high-quality maps to provide high-quality services. Example inaccuracies that can be encountered in the provided maps may include missing road segments, shifted road segments, missing road connections, missing or incorrect turn restrictions, and mislabeling road attributes like marking a directional road as one-way. Maps may also be rapidly changing in some areas due to new constructions. While the accuracy of various mapping systems, as given by service providers, is known to be high, even the minor discrepancies in the underlying maps may lead to unsatisfactory user experience in routing and location-based services. In this paper, we present a system that compares the routes returned by the public APIs of some major routing engines, namely Bing Maps, Google Maps, and OpenStreetMap. The system highlights the differences in the proposed routes between these routing engines, given the same start/end points for a planned trip. The route differences are examined based on travel distance, travel duration, and route geometry. The system can also enforce a routing engine to take the same route as another routing engine to identify the possible discrepancies in the underlying mapping system of each routing engine. The system identifies and categorizes the discovered discrepancies, across various engines, in (1) the geometry of the road segments, (2) the connectivity and turn restrictions of the Road Network Graph (RNG), and (3) the attributes of the road segments. The presented system is currently in pilot use by a group of professional editors to support their daily work of identifying, visually inspecting, and interactively trying alternative corrections to the underlying RNGs in various parts of the globe. This helps us develop the system's capabilities even further based on their continuous feedback in real usage scenarios.
{"title":"An Interactive System to Compare, Explore and Identify Discrepancies across Map Providers","authors":"Ayush Bandil, Vaishali Girdhar, K. Dinçer, Harsh Govind, Peiwei Cao, Ashley Song, Mohamed H. Ali","doi":"10.1145/3397536.3422348","DOIUrl":"https://doi.org/10.1145/3397536.3422348","url":null,"abstract":"All online map service providers are working hard to maintain high-quality maps to provide high-quality services. Example inaccuracies that can be encountered in the provided maps may include missing road segments, shifted road segments, missing road connections, missing or incorrect turn restrictions, and mislabeling road attributes like marking a directional road as one-way. Maps may also be rapidly changing in some areas due to new constructions. While the accuracy of various mapping systems, as given by service providers, is known to be high, even the minor discrepancies in the underlying maps may lead to unsatisfactory user experience in routing and location-based services. In this paper, we present a system that compares the routes returned by the public APIs of some major routing engines, namely Bing Maps, Google Maps, and OpenStreetMap. The system highlights the differences in the proposed routes between these routing engines, given the same start/end points for a planned trip. The route differences are examined based on travel distance, travel duration, and route geometry. The system can also enforce a routing engine to take the same route as another routing engine to identify the possible discrepancies in the underlying mapping system of each routing engine. The system identifies and categorizes the discovered discrepancies, across various engines, in (1) the geometry of the road segments, (2) the connectivity and turn restrictions of the Road Network Graph (RNG), and (3) the attributes of the road segments. The presented system is currently in pilot use by a group of professional editors to support their daily work of identifying, visually inspecting, and interactively trying alternative corrections to the underlying RNGs in various parts of the globe. This helps us develop the system's capabilities even further based on their continuous feedback in real usage scenarios.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"71 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116378573","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}
Saeed Arasteh, M. A. Tayebi, Zahra Zohrevand, U. Glässer, A. Shahir, Parvaneh Saeedi, H. Wehn
The impact of marine life on the oceans of our planet is undeniable and overfishing is a serious threat to marine ecosystems worldwide. Maritime domain awareness calls for continuous monitoring and tracking of fisheries using data from maritime intelligence sources to detect illegal fishing activities. Marine traffic data from vessel tracking services is a promising source for identifying, locating, and capturing vessel information. Given the volume of such data, manual processing is impossible, raising an immediate need for autonomous and smart systems to follow the footprints of vessels and detect their activity types in near real-time. To achieve this goal, we propose FishNET, a simple yet effective convolutional neural network (CNN) model for vessel trajectory classification. The model is trained using a set of invariant spatiotemporal feature sequences extracted from the behavioral characteristics of vessel movements. While existing approaches present point-based classification models, in this paper we not only discuss that a segment-based classification model has more realistic real-world applications but also show, by using expert-labelled data, that FishNET outperforms state-of-the-art fishing activity detection models. Our method does not require information about the fishing vessels type or type of fishing gear which is deployed. To show applications in taking action against illegal fishing, we apply the trained model on large real-world but unlabelled fishing vessel data from the U.S. and Denmark gathered over a period of four years. In this analysis, we show how FishNET can contribute to managing fisheries by learning more about spatiotemporal fishing effort distribution, and to law enforcement agencies by detecting unreported and underreported fishing effort of individual vessels.
{"title":"Fishing Vessels Activity Detection from Longitudinal AIS Data","authors":"Saeed Arasteh, M. A. Tayebi, Zahra Zohrevand, U. Glässer, A. Shahir, Parvaneh Saeedi, H. Wehn","doi":"10.1145/3397536.3422267","DOIUrl":"https://doi.org/10.1145/3397536.3422267","url":null,"abstract":"The impact of marine life on the oceans of our planet is undeniable and overfishing is a serious threat to marine ecosystems worldwide. Maritime domain awareness calls for continuous monitoring and tracking of fisheries using data from maritime intelligence sources to detect illegal fishing activities. Marine traffic data from vessel tracking services is a promising source for identifying, locating, and capturing vessel information. Given the volume of such data, manual processing is impossible, raising an immediate need for autonomous and smart systems to follow the footprints of vessels and detect their activity types in near real-time. To achieve this goal, we propose FishNET, a simple yet effective convolutional neural network (CNN) model for vessel trajectory classification. The model is trained using a set of invariant spatiotemporal feature sequences extracted from the behavioral characteristics of vessel movements. While existing approaches present point-based classification models, in this paper we not only discuss that a segment-based classification model has more realistic real-world applications but also show, by using expert-labelled data, that FishNET outperforms state-of-the-art fishing activity detection models. Our method does not require information about the fishing vessels type or type of fishing gear which is deployed. To show applications in taking action against illegal fishing, we apply the trained model on large real-world but unlabelled fishing vessel data from the U.S. and Denmark gathered over a period of four years. In this analysis, we show how FishNET can contribute to managing fisheries by learning more about spatiotemporal fishing effort distribution, and to law enforcement agencies by detecting unreported and underreported fishing effort of individual vessels.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123569171","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}
Zhou Fang, Jiaxin Qi, Tianren Yang, L. Wan, Ying Jin
This paper builds on the proven track record of CNN-based pattern recognition and feature extraction methods, and reports a novel model that classifies urban fabric samples of metropolitan areas in terms of (1) which city they belong to, (2) what types of urban fabric they belong to, and (3) which historic period they originate from. Currently, such tasks require intensive manual work by senior professionals, and even then, inconsistencies and errors occur. Our work is based on a novel urban fabric dataset of four metropolitan areas with distinct typologies (linear development, open block, gated compound, medieval region, irregular grid and orthogonal gird), which consist of high resolution 3-dimensional built form data and hierarchical street networks. The classification model presented in this paper is the first that is capable of predicting the city origin, urban fabric pattern type and construction period. The novelty is also characterised by jointly considering urban fabric features across multiple spatial scales. The experiments demonstrate that this multi-scale approach can capture a wide range of urban fabric features across cities, urban fabric pattern types and development periods. We further find that the effectiveness can be enhanced by appending an auxiliary network for identifying the most appropriate combinations of the multiple spatial scales in line with the classification task. The dataset and model can massively scale up the productivity of researchers and professionals working on cities.
{"title":"\"Reading\" cities with computer vision: a new multi-spatial scale urban fabric dataset and a novel convolutional neural network solution for urban fabric classification tasks","authors":"Zhou Fang, Jiaxin Qi, Tianren Yang, L. Wan, Ying Jin","doi":"10.1145/3397536.3422240","DOIUrl":"https://doi.org/10.1145/3397536.3422240","url":null,"abstract":"This paper builds on the proven track record of CNN-based pattern recognition and feature extraction methods, and reports a novel model that classifies urban fabric samples of metropolitan areas in terms of (1) which city they belong to, (2) what types of urban fabric they belong to, and (3) which historic period they originate from. Currently, such tasks require intensive manual work by senior professionals, and even then, inconsistencies and errors occur. Our work is based on a novel urban fabric dataset of four metropolitan areas with distinct typologies (linear development, open block, gated compound, medieval region, irregular grid and orthogonal gird), which consist of high resolution 3-dimensional built form data and hierarchical street networks. The classification model presented in this paper is the first that is capable of predicting the city origin, urban fabric pattern type and construction period. The novelty is also characterised by jointly considering urban fabric features across multiple spatial scales. The experiments demonstrate that this multi-scale approach can capture a wide range of urban fabric features across cities, urban fabric pattern types and development periods. We further find that the effectiveness can be enhanced by appending an auxiliary network for identifying the most appropriate combinations of the multiple spatial scales in line with the classification task. The dataset and model can massively scale up the productivity of researchers and professionals working on cities.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"338 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116208827","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}
Spatio-textual similarity join is an operation for finding documents, which are both spatially close and textually relevant. Joins in databases are considered to be the most expensive operation; similarly spatio-textual similarity join is a resource intensive operation. Therefore, it is natural to consider approaches to parallelize this operation. Many modern multi-core systems adopt a NUMA-based memory architecture. NUMA systems entail varying memory access latencies across nodes, which may adversely affect overall query latency. Recent work on spatio-textual similarity join have not addressed the effects of non-uniform access latencies in multi-node NUMA systems. In this paper, we propose a NUMA-aware parallel spatio-textual similarity join algorithm NA-STSJ-WS. It exploits topology-aware work-stealing with adaptive data placement. Experimental evaluation demonstrates that NA-STSJ-WS performs significantly better than existing approaches that are not NUMA-aware, and in the best case we observe 82× speedup over the sequential baseline.
{"title":"NUMA-Aware Spatio-Textual Similarity Join","authors":"Saransh Gautam, S. Ray, B. Nickerson","doi":"10.1145/3397536.3422227","DOIUrl":"https://doi.org/10.1145/3397536.3422227","url":null,"abstract":"Spatio-textual similarity join is an operation for finding documents, which are both spatially close and textually relevant. Joins in databases are considered to be the most expensive operation; similarly spatio-textual similarity join is a resource intensive operation. Therefore, it is natural to consider approaches to parallelize this operation. Many modern multi-core systems adopt a NUMA-based memory architecture. NUMA systems entail varying memory access latencies across nodes, which may adversely affect overall query latency. Recent work on spatio-textual similarity join have not addressed the effects of non-uniform access latencies in multi-node NUMA systems. In this paper, we propose a NUMA-aware parallel spatio-textual similarity join algorithm NA-STSJ-WS. It exploits topology-aware work-stealing with adaptive data placement. Experimental evaluation demonstrates that NA-STSJ-WS performs significantly better than existing approaches that are not NUMA-aware, and in the best case we observe 82× speedup over the sequential baseline.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"116 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124220853","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}
Philip E. Brown, Krystian Czapiga, Arun Jotshi, Y. Kanza, Velin Kounev
The growing demand for high-speed networks is increasing the use of high-frequency electromagnetic waves in wireless networks, including in microwave backhaul links and 5G. The relative higher frequency provides a high bandwidth, but it is very sensitive to obstructions and interference. Hence, when positioning a transmitter-receiver pair, the line-of-sight between them should be free of obstacles. Furthermore, the Fresnel zone around the line-of-sight should be clear of obstructions, to guarantee effective transmission. When deploying microwave backhaul links or a cellular network there is a need to select the locations of the antennas accordingly. To help network planners, we developed an interactive tool that allows users to position antennas in different locations over a 3D model of the world. Users can interactively change antenna locations and other parameters, to examine clearance of Fresnel zones. In this paper we illustrate the interactive tool and the ability to test clearance in real-time, to support interactive network planning.
{"title":"Interactive Testing of Line-of-Sight and Fresnel Zone Clearance for Planning Microwave Backhaul Links and 5G Networks","authors":"Philip E. Brown, Krystian Czapiga, Arun Jotshi, Y. Kanza, Velin Kounev","doi":"10.1145/3397536.3422332","DOIUrl":"https://doi.org/10.1145/3397536.3422332","url":null,"abstract":"The growing demand for high-speed networks is increasing the use of high-frequency electromagnetic waves in wireless networks, including in microwave backhaul links and 5G. The relative higher frequency provides a high bandwidth, but it is very sensitive to obstructions and interference. Hence, when positioning a transmitter-receiver pair, the line-of-sight between them should be free of obstacles. Furthermore, the Fresnel zone around the line-of-sight should be clear of obstructions, to guarantee effective transmission. When deploying microwave backhaul links or a cellular network there is a need to select the locations of the antennas accordingly. To help network planners, we developed an interactive tool that allows users to position antennas in different locations over a 3D model of the world. Users can interactively change antenna locations and other parameters, to examine clearance of Fresnel zones. In this paper we illustrate the interactive tool and the ability to test clearance in real-time, to support interactive network planning.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131250910","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}
Quanjun Chen, Renhe Jiang, Chuang Yang, Z. Cai, Z. Fan, K. Tsubouchi, R. Shibasaki, Xuan Song
Nowadays, GPS devices have increased explosively and produced huge amounts of trajectory data related to people's outgoing. Through those big location data, many researches aim to analyze human mobility for urban development, such as human movement prediction/modeling, POI (Point-Of-Interest) recommendation. However, trajectory data only contains timestamp and location information. The intention of human movement is not explicit so that it is hard to understand why people go to somewhere. The intention prior to the activity could be of great significance for analyzing and predicting human mobility, which has not been taken into consideration by the existing researches until the present. Thus, in this study, we propose a brand-new concept called human intentional mobility, aiming to employ intention information to predict people's outgoing. We carefully utilize user's search query to sense his intention as well as the intensity. For instance, if a user searches a certain POI for many times in a short period, it will represent a relatively high intention to go there. Then, to fully utilize this intention representation for predicting whether user will visit searched POI or not, we specially design Dual Sequential Interaction Network (DualSIN) as a novel and unique deep-learning model, which can effectively capture the sophisticated interactions among two kinds of sequential information (i.e., search sequence and mobility sequence) and typical categorical information (i.e., user attributes). Last, we evaluate our model on real-world dataset collected from Yahoo! Japan portal application, and demonstrate that it can achieve superior satisfactory performances to the-state-of-the-art models on multiple POI search queries.
{"title":"DualSIN","authors":"Quanjun Chen, Renhe Jiang, Chuang Yang, Z. Cai, Z. Fan, K. Tsubouchi, R. Shibasaki, Xuan Song","doi":"10.1145/3397536.3422221","DOIUrl":"https://doi.org/10.1145/3397536.3422221","url":null,"abstract":"Nowadays, GPS devices have increased explosively and produced huge amounts of trajectory data related to people's outgoing. Through those big location data, many researches aim to analyze human mobility for urban development, such as human movement prediction/modeling, POI (Point-Of-Interest) recommendation. However, trajectory data only contains timestamp and location information. The intention of human movement is not explicit so that it is hard to understand why people go to somewhere. The intention prior to the activity could be of great significance for analyzing and predicting human mobility, which has not been taken into consideration by the existing researches until the present. Thus, in this study, we propose a brand-new concept called human intentional mobility, aiming to employ intention information to predict people's outgoing. We carefully utilize user's search query to sense his intention as well as the intensity. For instance, if a user searches a certain POI for many times in a short period, it will represent a relatively high intention to go there. Then, to fully utilize this intention representation for predicting whether user will visit searched POI or not, we specially design Dual Sequential Interaction Network (DualSIN) as a novel and unique deep-learning model, which can effectively capture the sophisticated interactions among two kinds of sequential information (i.e., search sequence and mobility sequence) and typical categorical information (i.e., user attributes). Last, we evaluate our model on real-world dataset collected from Yahoo! Japan portal application, and demonstrate that it can achieve superior satisfactory performances to the-state-of-the-art models on multiple POI search queries.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115357671","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}
Agent-based simulations, combined with large scale mobility data, have been an effective method for understanding urban scale human dynamics. However, collecting such large scale human mobility datasets are especially difficult during rare events (e.g., natural disasters), reducing the performance of agent-based simulations. To tackle this problem, we develop an agent-based model that can simulate urban dynamics during rare events by learning from other cities using inverse reinforcement learning. More specifically, in our framework, agents imitate real human-beings' travel behavior from areas where rare events have occurred in the past (source area) and produce synthetic people movement in different cities where such rare events have never occurred (target area). Our framework contains three main stages: 1) recovering the reward function, where the people's travel patterns and preferences are learned from the source areas; 2) transferring the model of the source area to the target areas; 3) simulating the people movement based on learned model in the target area. We apply our approach in various cities for both normal and rare situations using real-world GPS data collected from more than 1 million people in Japan, and show higher simulation performance than previous models.
{"title":"Intercity Simulation of Human Mobility at Rare Events via Reinforcement Learning","authors":"Y. Pang, K. Tsubouchi, T. Yabe, Y. Sekimoto","doi":"10.1145/3397536.3422244","DOIUrl":"https://doi.org/10.1145/3397536.3422244","url":null,"abstract":"Agent-based simulations, combined with large scale mobility data, have been an effective method for understanding urban scale human dynamics. However, collecting such large scale human mobility datasets are especially difficult during rare events (e.g., natural disasters), reducing the performance of agent-based simulations. To tackle this problem, we develop an agent-based model that can simulate urban dynamics during rare events by learning from other cities using inverse reinforcement learning. More specifically, in our framework, agents imitate real human-beings' travel behavior from areas where rare events have occurred in the past (source area) and produce synthetic people movement in different cities where such rare events have never occurred (target area). Our framework contains three main stages: 1) recovering the reward function, where the people's travel patterns and preferences are learned from the source areas; 2) transferring the model of the source area to the target areas; 3) simulating the people movement based on learned model in the target area. We apply our approach in various cities for both normal and rare situations using real-world GPS data collected from more than 1 million people in Japan, and show higher simulation performance than previous models.","PeriodicalId":233918,"journal":{"name":"Proceedings of the 28th International Conference on Advances in Geographic Information Systems","volume":"235 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116864989","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：