Gengchen Mai

From ITDL to Place2Vec: Reasoning About Place Type Similarity and Relatedness by Learning Embeddings From Augmented Spatial Contexts

Understanding, representing, and reasoning about Points Of Interest (POI) types such as Auto Repair, Body Shop, Gas Stations, or Planetarium, is a key aspect of geographic information retrieval, recommender systems, geographic knowledge graphs, as well as studying urban spaces in general, e.g., for extracting functional or vague cognitive regions from user-generated content. One prerequisite to these tasks is the ability to capture the similarity and relatedness between POI types. Intuitively, a spatial search that returns body shops or even gas stations in the absence of auto repair places is still likely to satisfy some user needs while returning planetariums will not. Place hierarchies are frequently used for query expansion, but most of the existing hierarchies are relatively shallow and structured from a single perspective, thereby putting POI types that may be closely related regarding some characteristics far apart from another. This leads to the question of how to learn POI type representations from data. Models such as Word2Vec that produces word embeddings from linguistic contexts are a novel and promising approach as they come with an intuitive notion of similarity. However, the structure of geographic space, e.g., the interactions between POI types, differs substantially from linguistics. In this work, we present a novel method to augment the spatial contexts of POI types using a distance-binned, information-theoretic approach to generate embeddings. We demonstrate that our work outperforms Word2Vec and other models using three different evaluation tasks and strongly correlates with human assessments of POI type similarity. We published the resulting embeddings for 570 place types as well as a collection of human similarity assessments online for others to use.

Moon Landing or Safari? A Study of Systematic Errors and Their Causes in Geographic Linked Data

While the adoption of Linked Data technologies has grown dramatically over the past few years, it has not come without its own set of growing challenges. The triplification of domain data into Linked Data has not only given rise to a leading role of places and positioning information for the dense interlinkage of data about actors, objects, and events, but also led to massive errors in the generation, transformation, and semantic annotation of data. In a global and densely interlinked graph of data, even seemingly minor error can have far reaching consequences as different datasets make statements about the same resources. In this work we present the first comprehensive study of systematic errors and their potential causes. We also discuss lessons learned and means to avoid some of the introduced pitfalls in the future.

ADCN: An anisotropic density-based clustering algorithm for discovering spatial point patterns with noise

Author(s): Mai, Gengchen | Advisor(s): Janowicz, Krzysztof | Abstract: Density-based clustering algorithms such as DBSCAN have been widely used for spatial knowledge discovery as they offer several key advantages compared to other clustering algorithms. They can discover clusters with arbitrary shapes, are robust to noise and do not require prior knowledge (or estimation) of the number of clusters. The idea of using a scan circle centered at each point with a search radius Eps to find at least MinPts points as a criterion for deriving local density is easily understandable and sufficient for exploring isotropic spatial point patterns. However, there are many cases that cannot be adequately captured this way, particularly if they involve linear features or shapes with a continuously changing density such as a spiral. In such cases, DBSCAN tends to either create an increasing number of small clusters or add noise points into large clusters. Therefore, in this paper, we propose a novel anisotropic density-based clustering algorithm (ADCN). To motivate our work, we introduce synthetic and real-world cases that cannot be sufficiently handled by DBSCAN (and OPTICS). We then present our clustering algorithm and test it with a wide range of cases. We demonstrate that our algorithm can perform as equally well as DBSCAN in cases that do not explicitly benefit from an anisotropic perspective and that it outperforms DBSCAN in cases that do. We show that our approach has the same time complexity as DBSCAN and OPTICS, namely O(n log n) when using a spatial index and O(n 2 ) otherwise. We provide an implementation and test the runtime over multiple cases. Finally, we apply DBSCAN, OPTICS, and ADCN to the task of extracting urban areas of interest (AOI) from geotagged photos in six cities. Visual comparison shows that, comparing to DBSCAN and OPTICS, ADCN is inclined to extract AOIs with linear shapes which follow the underline road networks. ADCN also turns out to connect clusters when the spatial distribution of them shows similar directions.

ADCN: an anisotropic density-based clustering algorithm

In this work we introduce an anisotropic density-based clustering algorithm. It outperforms DBSCAN and OPTICS for the detection of anisotropic spatial point patterns and performs equally well in cases that do not explicitly benefit from an anisotropic perspective. ADCN has the same time complexity as DBSCAN and OPTICS, namely O(n log n) when using a spatial index, O(n2) otherwise.

GNIS-LD: Serving and Visualizing the Geographic Names Information System Gazetteer as Linked Data

In this dataset description paper we introduce the GNIS-LD, an authoritative and public domain Linked Dataset derived from the Geographic Names Information System (GNIS) which was developed by the U.S. Geological Survey (USGS) and the U.S. Board on Geographic Names. GNIS provides data about current, as well as historical, physical, and cultural geographic features in the United States. We describe the dataset, introduce an ontology for geographic feature types, and demonstrate the utility of recent linked geographic data contributions made in conjunction with the development of this resource. Co-reference resolution links to GeoNames.org and DBpedia are provided in the form of owl:sameAs relations. Finally, we point out how the adapted workflow is foundational for complex Digital Line Graph (DLG) data from the USGS National Map and how the GNIS-LD data can be integrated with DLG and other data sources such as sensor observations.

Mobile GIS and Location-Based Services

With the fast development of mobile Web and computing technologies, as well as increasingly availability of mobile devices, mobile information technologies have revolutionary influence on the human society. In this article, we present a comprehensive review of mobile geographic information systems (GIS) and location-based services (LBS) concepts, core components and characteristics, technology stack, a variety of applications, and research frontiers. The applications of mobile GIS and LBS face several challenges such as small screen for data visualization, limited bandwidth and high costs of networks for transferring data, battery consuming for positioning and computing capabilities, heterogeneous types, and multilevel spatiotemporal resolutions of datasets. This subject itself is also fast developing and advancing. We do hope this article not only educates the next-generation of geographic information systems/science major students with the knowledge but also inspire them to dive into this challenging research areas and make their contributions in the future.