In recent years, numerous neural network models have been put forth, with an emphasis on the applications of raster imagery and spatiotemporal non-imagery datasets. Implementing these models using existing deep learning frame-works, such as PyTorch and TensorFlow, requires nontrivial coding efforts from the developers although these deep learning frameworks support the implementation of various state-of-the-art machine learning models, such as neural networks, hidden Markov models, and support vector machines. This is due to the fact that the models emphasized on spatiotemporal applications differ extensively from state-of-the-art models supported by existing deep learning frameworks. Moreover, existing deep learning frameworks lack the support for scalable data preprocessing, a mandatory step for converting spatiotemporal datasets into trainable tensors. Considering the limitations of existing deep learning frameworks, we present GeoTorchAI, a framework for deep learning and scalable data processing on raster imagery and spatiotemporal non-imagery datasets. GeoTorchAI enables machine learning practitioners to implement spatiotemporal deep learning models with minimum coding efforts on top of PyTorch. It provides state-of-the-art neural network models, ready-to-use benchmark datasets, and transformation operations for raster imagery and spatiotemporal non-imagery datasets. Besides deep learning, GeoTorchAI contains a data preprocessing module and a DFtoTorch Converter module that enable the formation of trainable spatiotemporal vector datasets and the mapping of preprocessed DataFrames into PyTorch tensors, respectively.

Deep learning frameworks, such as PyTorch and TensorFlow, support the implementation of various state-of-the-art machine learning models such as neural networks, hidden Markov models, and support vector machines. In recent years, many extensions of neural network models have been proposed in the literature targeting the applications of raster and spatiotemporal datasets. Implementing these models using existing deep learning frameworks requires nontrivial coding efforts from the developers because these extensions either are hybrid combinations of various categories of neural network models or differ extensively from state-of-the-art models supported by existing deep learning frameworks. Moreover, existing deep learning frameworks lack the support for scalable data preprocessing required to form trainable tensors from raw spatiotemporal datasets. To enable easy implementation of these neural network extensions, we present GeoTorch, a framework for deep learning and scalable data processing on raster and spatiotemporal datasets. Along with the state-of-the-art spatiotemporal models and ready-to-use benchmark datasets, we propose a data preprocessing module that allows the processing and transformation of spatiotemporal datasets in a cluster computing setting.

Spatial datasets are used extensively to train machine learning (ML) models for applications such as spatial regression, classification, clustering, and deep learning. Most of the real-world spatial datasets are often too large, and many spatial ML algorithms represent the geographical region as a grid consisting of several spatial cells. If the granularity of the grid is too fine, that results in a large number of grid cells leading to long training time and high memory consumption issues during the model training. To alleviate this problem, we propose a machine learning-aware spatial data re-partitioning framework that substantially reduces the granularity of the spatial grid. Our spatial data re-partitioning approach combines fine-grained, adjacent spatial cells from a grid into coarser cells prior to training an ML model. During this re-partitioning phase, we keep the information loss within a user-defined threshold without significantly degrading the accuracy of the ML model. According to the empirical evaluation performed on several real-world datasets, the best results achieved by our spatial re-partitioning framework show that we can reduce the data volume and training time by up to 81%, while keeping the difference in prediction or classification error below 5% as compared to a model that is trained on the original input dataset, for most of the ML applications. Our re-partitioned framework also outperforms the state-of-the-art data reduction baselines by 2% to 20% w.r.t. prediction and classification errors.

Identifying various mentions of the same real-world locations is known as spatial entity matching. GEM is an end-to-end Geospatial EM framework that matches polygon geometry entities in addition to point geometry type. Blocking, feature vector creation, and classification are the core steps of our system. GEM comprises of an efficient and lightweight blocking technique, GeoPrune, that uses the geohash encoding mechanism. We re-purpose the spatial proximality operators from Apache Sedona to create semantically rich spatial feature vectors. The classification step in GEM is a pluggable component, which consumes a unique feature vector and determines whether the geolocations match or not. We conduct experiments with three classifiers upon multiple large-scale geospatial datasets consisting of both spatial and relational attributes. GEM achieves an F-measure of 1.0 for a point x point dataset with 176k total pairs, which is 42% higher than a state-of-the-art spatial EM baseline. It achieves F-measures of 0.966 and 0.993 for the point x polygon dataset with 302M total pairs, and the polygon x polygon dataset with 16M total pairs respectively.

Prediction of the next SQL query from the user, given her sequence of queries until the current timestep, during an ongoing interaction session of the user with the database, can help in speculative query processing and increased interactivity. While existing machine learning-- (ML) based approaches use recommender systems to suggest relevant queries to a user, there has been no exhaustive study on applying temporal predictors to predict the next user issued query. In this work, we experimentally compare ML algorithms in predicting the immediate next future query in an interaction workload, given the current user query or the sequence of queries in a user session thus far. As a part of this, we propose the adaptation of two powerful temporal predictors: (a) Recurrent Neural Networks (RNNs) and (b) a Reinforcement Learning approach called Q-Learning that uses Markov Decision Processes. We represent each query as a comprehensive set of fragment embeddings that not only captures the SQL operators, attributes, and relations but also the arithmetic comparison operators and constants that occur in the query. Our experiments on two real-world datasets show the effectiveness of temporal predictors against the baseline recommender systems in predicting the structural fragments in a query w.r.t. both quality and time. Besides showing that RNNs can be used to synthesize novel queries, we find that exact Q-Learning outperforms RNNs despite predicting the next query entirely from the historical query logs.

In this paper, we present a middleware framework that runs on top of a SQL data system with the purpose of increasing the interactivity of geospatial visualization dashboards. The proposed system adopts a sampling cube approach that stores pre-materialized spatial samples and allows users to define their own accuracy loss function such that the produced samples can be used for various user-defined visualization tasks. The system ensures that the difference between the sample fed into the visualization dashboard and the raw query answer never exceeds the user-specified loss threshold. To reduce the number of cells in the sampling cube and hence mitigate the initialization time and memory utilization, the system employs two main strategies: (1) a partially materialized cube to only materialize local samples of those queries for which the global sample (the sample drawn from the entire dataset) exceeds the required accuracy loss threshold. (2) a sample selection technique that finds similarities between different local samples and only persists a few representative samples. Based on the extensive experimental evaluation, Tabula can bring down the total data-to-visualization time (including both data-system and visualization times) of a heat map generated over 700 million taxi rides to 600 milliseconds with 250 meters user-defined accuracy loss. Besides, Tabula costs up to two orders of magnitude less memory footprint (e.g., only 800 MB for the running example) and one order of magnitude less initialization time than the fully materialized sampling cube.

Entity Matching (EM) is a core data cleaning task, aiming to identify different mentions of the same real-world entity. Active learning is one way to address the challenge of scarce labeled data in practice, by dynamically collecting the necessary examples to be labeled by an Oracle and refining the learned model (classifier) upon them. In this paper, we build a unified active learning benchmark framework for EM that allows users to easily combine different learning algorithms with applicable example selection algorithms. The goal of the framework is to enable concrete guidelines for practitioners as to what active learning combinations will work well for EM. Towards this, we perform comprehensive experiments on publicly available EM datasets from product and publication domains to evaluate active learning methods, using a variety of metrics including EM quality, #labels and example selection latencies. Our most surprising result finds that active learning with fewer labels can learn a classifier of comparable quality as supervised learning. In fact, for several of the datasets, we show that there is an active learning combination that beats the state-of-the-art supervised learning result. Our framework also includes novel optimizations that improve the quality of the learned model by roughly 9% in terms of F1-score and reduce example selection latencies by up to 10× without affecting the quality of the model.

In this paper, we demonstrate Tabula, a middleware that sits between the data system and the geospatial visualization dashboard to increase user interactivity. The proposed system adopts a sampling cube approach that stores prematerialized spatial samples and allows data scientists to define their own accuracy loss function such that the produced samples can be used for various user-defined visualization tasks. The system ensures that the difference between the sample fed into the dashboard and the raw query answer never exceeds the user-specified loss threshold. For demonstration purposes, we connect Apache Zeppelin, a visualization dashboard, to the system and show how Tabula accelerates interactive visualizations on NYC Taxi Trip data, Yelp review data and San Diego Smart Streetlights data.

Knowledge Graphs are widely used to store facts about real-world entities and events. With the ubiquity of spatial data, vertexes or edges in knowledge graphs can possess spatial location attributes side by side with other non-spatial attributes. For instance, as of June 2018 the Wikidata knowledge graph contains 48; 547; 142 data items (i.e., vertexes) to date and ≈13% of them have spatial location attributes. The co-existence of the graph and spatial data in the same geographic knowledge graph allows users to search the graph with local intent. Many location-based services such as UberEats, GrubHub, and Yelp already employ similar knowledge graphs to enhance the location search experience for their end-users. In this paper, we demonstrate a system, namely Spindra, that provides efficient management of geographic knowledge graphs. We demonstrate the system using an interactive map-based web interface that allows users to issue location-aware search queries over the WikiData knowledge graph. The Front end will then visualize the returned geographic knowledge to the user using OpenStreetMaps.

The volume of spatial data increases at a staggering rate. This tutorial comprehensively studies how existing works extend Apache Spark to uphold massive-scale spatial data. During this 1.5 hour tutorial, we first provide a background introduction of the characteristics of spatial data and the history of distributed data management systems. A follow-up section presents the common approaches used by the practitioners to extend Spark and introduces the vital components in a generic spatial data management system. The third, fourth and fifth sections then discuss the ongoing efforts and experience in spatial-temporal data, spatial data analytics and streaming spatial data, respectively. The sixth part finally concludes this tutorial to help the audience better grasp the overall content and points out future research directions.

Visualizing data on maps is deemed a powerful tool for data scientists to make sense of geospatial data. The geospatial map visualization (abbr. MapViz) process first loads the designated geospatial data, processes the data and then applies the map visualization effect. Guaranteeing detailed and accurate geospatial MapViz (e.g., at multiple zoom levels) requires extremely high-resolution maps. Classic solutions suffer from limited computation resources while scalable MapViz system architectures are not able to co-optimize the data management and visualization phases in the same system. This paper demonstrates GeoSparkViz, a full-fledged system that allows the user to load, prepare, integrate and execute MapViz tasks in the same system. For demonstration purpose, we implemented a web interface using a node.js web server, Baidu echarts library, and MapBox on top of GeoSparkViz to visually explore patterns in the New York City Taxi Trips dataset. The demonstration scenarios show how the data preparation and map visualization phases are combined in GeoSparkViz.

Road network traffic data has been widely studied by researchers and practitioners in different areas such as urban planning, traffic prediction, and spatial-temporal databases. For instance, researchers use such data to evaluate the impact of road network changes. Unfortunately, collecting large-scale high-quality urban traffic data requires tremendous efforts because participating vehicles must install GPS receivers and administrators must continuously monitor these devices. There has been a number of urban traffic simulators trying to generate such data with different features. However, they suffer from two critical issues (1) scalability: most of them only offer single-machine solution which is not adequate to produce large-scale data. Some simulators can generate traffic in parallel but do not well balance the load among machines in a cluster. (2) granularity: many simulators do not consider microscopic traffic situations including traffic lights, lane changing, car following. In the paper, we propose GeoSparkSim, a scalable traffic simulator which extends Apache Spark to generate large-scale road network traffic datasets with microscopic traffic simulation. The proposed system seamlessly integrates with a Spark-based spatial data management system, GeoSpark, to deliver a holistic approach that allows data scientists to simulate, analyze and visualize largescale urban traffic data. To implement microscopic traffic models, GeoSparkSim employs a simulation-aware vehicle partitioning method to partition vehicles among different machines such that each machine has a balanced workload. The experimental analysis shows that GeoSparkSim can simulate the movements of 200 thousand vehicles over a very large road network (250 thousand road junctions and 300 thousand road segments).

When a data scientist analyzes mobility data (e.g., using a data visualization tool), she may find out some interesting facts in the dataset. An example of a fact can be: "The number of Taxi trips in NYC on January 23, 2016, dropped drastically as compared to other days of the same month". However, the data scientist may be left clueless if they cannot find a crisp explanation to such a fact. Furthermore, the tedious task of finding an explanation by manually scraping the data becomes even impossible with big data. Existing techniques are designed for non-spatial data which cannot be applied to spatial data because it does not consider the spatial proximity. In this paper, we propose an automatic framework which guides the data scientist to explain the fact discovered from mobility data. Our approach expands on the aggravation and intervention techniques while using spatial partitioning/clustering to improve explanations for spatial data. Experiments show that the proposed approach outperforms the state-of-the-art approaches in finding the explanation for facts extracted from NYC taxi real mobility dataset.

Road network traffic data has been widely studied by researchers and practitioners in different areas such as urban planning, traffic prediction and spatial-temporal databases. The existing urban traffic simulators suffer from two critical issues (1) scalability: most of them only offer single-machine solutions which are not adequate to produce large-scale data. Some simulators can generate traffic in parallel but do not well balance the load among machines in a cluster. (2) granularity: many simulators do not consider microscopic traffic situations including traffic lights, lane changing, and car following. In the paper, we propose GeoSparkSim, a scalable traffic simulator which extends Apache Spark to generate large-scale road network traffic datasets with microscopic traffic simulation. The proposed system seamlessly integrates with a Spark-based spatial data management system, GeoSpark, to deliver a holistic approach that allows data scientists to simulate, analyze and visualize large-scale urban traffic data. To implement microscopic traffic models, GeoSparkSim employs a simulation-aware vehicle partitioning method to partition vehicles among different machines such that each machine has a balanced workload. A full-fledged prototype of GeoSparkSim is implemented in Apache Spark. In this demonstration, we will show the attendees how to issue GeoSparkSim simulation tasks via the user interface, visualize simulated vehicle movements, and monitor the backend Spark cluster status.

The paper proposes Riso-Tree, a generic indexing framework for geographic knowledge graphs. Riso-Tree enables fast execution of graph queries that involve spatial predicates (aka. GraSp). The proposed framework augments the classic R-Tree structure with pre-materialized sub-graph entries. Riso-Tree first partitions the graph into sub-graphs based on their connectivity to the spatial sub-regions. The proposed index allows for fast execution of GraSp queries by efficiently pruning the traversed vertexes/edges based upon the materialized sub-graph information. The experiments show that the proposed Riso-Tree achieves up to two orders magnitude faster execution time than its counterparts when executing GraSp queries on real knowledge graphs (e.g., WikiData).

There has been an increasing interest into blurring the line between human-interaction and database systems. Several research papers tackled the Human-Database Interaction (HDI) challenge, yet none of them provides a holistic HDI approach. In this talk, we briefly describe dbTinder, a database engine that bridges the gap between the conversation approach humans use to interact and theQuery→Answerapproach used in classic database systems. To achieve that, dbTinder turns HDI into an intent discovery process where the system pro-actively converses with the user to guide her towards potentiallyinterestingtuples in the database.

Data Visualization allows users to summarize, analyze and reason about data. A map visualization tool first loads the designated geospatial data, processes the data and then applies the map visualization effect. Guaranteeing detailed and accurate geospatial map visualization (e.g., at multiple zoom levels) requires extremely high-resolution maps. Classic solutions suffer from limited computation resources and hence take a tremendous amount of time to generate maps for large-scale geospatial data.

Personalized recommendation has become popular in modern web services. For instance, Amazon recommends new items to shoppers. Also, Netflix recommends shows to viewers, and Facebook recommends friends to its users. Despite the ubiquity of recommendation applications, classic database management systems still do not provide in-house support for recommending data stored in the database. In this paper, we present the anatomy of RecDB an open source PostgreSQLbased system that provides a unified approach for declarative data recommendation inside the database engine. RecDB realizes the personalized recommendation functionality as query operators inside the database kernel. That facilitates applying the recommendation functionality and typical database operations (e.g., Selection, Join, Top-k) side-by-side. To further reduce the application latency, RecDB pre-computes and caches the generated recommendation in the database. In the paper, we present extensive experiments that study the performance of personalized recommendation applications based on an actual implementation inside PostgreSQL 9.2 using real Movie recommendation and location-aware recommendation scenarios. The results show that a recommendation-aware database engine, i.e., RecDB, outperforms the classic approach that implements the recommendation logic on-top of the database engine in various recommendation applications.

Thanks to the wide spread use of mobile and wearable devices, popular social networks, e.g., Facebook, prompts users to add spatial attributes to social entities, e.g., check-ins, traveling posts, and geotagged photos, leading to what is known as, The GeoSocial Graph. In such graph, usersmay issue a Reachability Query with Spatial Range Predicate (abbr. RangeReach). RangeReach finds whether an input vertex can reach any spatial vertex that lies within an input spatial range. The paper proposes GEOREACH, an approach that adds spatial data awareness to a graph database management system. GEOREACH allows efficient execution of RangeReach queries, yet without compromising a lot on the overall system scalability. Experiments based on system implementation inside Neo4j prove that GEOREACH exhibits up to two orders of magnitude better query performance and up to four times less storage than the state-of-the-art spatial and reachability indexing approaches.

The paper demonstrates Hippo a lightweight database indexing scheme that significantly reduces the storage and maintenance overhead without compromising much on the query execution performance. Hippo stores disk page ranges instead of tuple pointers in the indexed table to reduce the storage space occupied by the index. It maintains simplified histograms that represent the data distribution and adopts a page grouping technique that groups contiguous pages into page ranges based on the similarity of their index key attribute distributions. When a query is issued, Hippo leverages the page ranges and histogram-based page summaries to recognize those pages such that their tuples are guaranteed not to satisfy the query predicates and then inspects the remaining pages. We demonstrate Hippo using a billion NYC taxi trip records. Video: http://www.youtube.com/watch?v=wWaOK2-9k9A.

Spatial range queries are used on a daily basis in real-life applications, such as Google Maps and Yelp. Many of those applications may need to rank the spatial objects, enclosed by the spatial range, before presenting the result to the end-user. Existing systems rank spatial objects according to different rules, such as average user rating, distance to the user's location, etc... The popularity of social networks allowed many applications to leverage the social graph in delivering a social-proximity aware ranked list of spatial objects. In this paper, we formally define a query that returns the top-k spatial objects in a given spatial region and rank them according to the social proximity of these objects to the querying user (SKNNGEO). Furthermore, a framework that integrates a joint search on both the social and spatial domains is proposed to efficiently solve the SKNNGEO query. The paper evaluates the proposed approach using real dataset extracted from the Yelp application. Extensive experiments show that the proposed approach outperform existing baseline approaches in processing SKNNGEO query.

This paper demonstrates GEOSPARK a cluster computing framework for developing and processing large-scale spatial data analytics programs. GEOSPARK consists of three main layers: Apache Spark Layer, Spatial RDD Layer and Spatial Query Processing Layer. Apache Spark Layer provides basic Apache Spark functionalities as regular RDD operations. Spatial RDD Layer consists of three novel Spatial Resilient Distributed Datasets (SRDDs) which extend regular Apache Spark RDD to support geometrical and spatial objects with data partitioning and indexing. Spatial Query Processing Layer executes spatial queries (e.g., Spatial Join) on SRDDs. The dynamic status of SRDDs and spatial operations are visualized by GEOSPARK monitoring map interface. We demonstrate GEOSPARK using three spatial analytics applications (spatial aggregation, autocorrelation and co-location) to show how users can easily define their spatial analytics tasks and efficiently process such tasks on large-scale spatial data at interactive performance.

Classic database indexes (e.g., B + -Tree), though speed up queries, suffer from two main drawbacks: (1) An index usually yields 5% to 15% additional storage overhead which results in non-ignorable dollar cost in big data scenarios especially when deployed on modern storage devices. (2) Maintaining an index incurs high latency because the DBMS has to locate and update those index pages affected by the underlying table changes. This paper proposes Hippo a fast, yet scalable, database indexing approach. It significantly shrinks the index storage and mitigates maintenance overhead without compromising much on the query execution performance. Hippo stores disk page ranges instead of tuple pointers in the indexed table to reduce the storage space occupied by the index. It maintains simplified histograms that represent the data distribution and adopts a page grouping technique that groups contiguous pages into page ranges based on the similarity of their index key attribute distributions. When a query is issued, Hippo leverages the page ranges and histogram-based page summaries to recognize those pages such that their tuples are guaranteed not to satisfy the query predicates and inspects the remaining pages. Experiments based on real and synthetic datasets show that Hippo occupies up to two orders of magnitude less storage space than that of the B + -Tree while still achieving comparable query execution performance to that of the B + -Tree for 0.1% -- 1% selectivity factors. Also, the experiments show that Hippo outperforms BRIN (Block Range Index) in executing queries with various selectivity factors. Furthermore, Hippo achieves up to three orders of magnitude less maintenance overhead and up to an order of magnitude higher throughput (for hybrid query/update workloads) than its counterparts.

This paper introduces GeoSpark an in-memory cluster computing framework for processing large-scale spatial data. GeoSpark consists of three layers: Apache Spark Layer, Spatial RDD Layer and Spatial Query Processing Layer. Apache Spark Layer provides basic Spark functionalities that include loading / storing data to disk as well as regular RDD operations. Spatial RDD Layer consists of three novel Spatial Resilient Distributed Datasets (SRDDs) which extend regular Apache Spark RDDs to support geometrical and spatial objects. GeoSpark provides a geometrical operations library that accesses Spatial RDDs to perform basic geometrical operations (e.g., Overlap, Intersect). System users can leverage the newly defined SRDDs to effectively develop spatial data processing programs in Spark. The Spatial Query Processing Layer efficiently executes spatial query processing algorithms (e.g., Spatial Range, Join, KNN query) on SRDDs. GeoSpark also allows users to create a spatial index (e.g., R-tree, Quad-tree) that boosts spatial data processing performance in each SRDD partition. Preliminary experiments show that GeoSpark achieves better run time performance than its Hadoop-based counterparts (e.g., SpatialHadoop).

Recently, the volume of available spatial data increased tremendously. For instance, in November 2013 NASA announced the release of hundreds of Terabytes of its earth remote sensing dataset. Such data includes but not limited to: weather maps, socioeconomic data, vegetation indices, geological maps, and more. Making sense of such spatial data will be beneficial for several applications that may transform science and society -- For example: (1) Space Science: that allows astronomers to study and probably discover new features of both the earth and the outer space, (2) Socio-Economic Analysis: that includes for example climate change analysis, study of deforestation, population migration, and variation in sea levels, (3) Urban Planning: assisting government in city planning, road network design, and transportation engineering, (4) Disaster Planning: that helps in assessing the impact of natural disasters. The main aim of this paper is to investigate novel data management techniques that enable interactive and scalable exploration of big spatial data. The paper envisions novel system architectures that provide support for interactive and spatial data exploration, as follows: (1) The paper suggests extending data analytics frameworks, e.g., Apache Spark, to support spatial data types and operations at scale. The resulting framework will serve as a scalable backbone for processing spatial data exploration tasks. (2) It also sketches novel structures and algorithms that leverage modern hardware, e.g., SSDs, and in-memory data processing techniques to efficiently store and access spatial data. Second, the paper proposes extending spatial database systems to support an exploration-aware spatial query evaluation paradigm through three novel components: (1) Spatial Query Steering: that allows the user to slightly modify the query conditions online (zooming in/out) and retrieve the new results in very low latency. (2) Recommendation-Aware Spatial Querying: that injects the recommendation functionality inside classical spatial query executors to support spatial data recommendation. It leverages recommendation algorithms to predict what spatial objects/areas the user would like based on her past interactions with the system. (3) Spatial Query Approximation: That aims at achieving interactive performance by studying the tradeoff between approximate spatial data exploration and query response time.

This paper presents RECATHON, a context-aware recommender system built entirely inside a database system. Unlike traditional recommender systems that are context-free where they support the general query of Recommend movies for a certain user, RECATHON users can request recommendations based on their age, location, gender, or any other contextual/ demographical/preferential user attribute. A main challenge of supporting such kind of recommenders is the difficulty of deciding what attributes to build recommenders on. RECATHON addresses this challenge as it supports building recommenders in database systems in an analogous way to building index structures. Users can decide to create recommenders on selected attributes, e.g., Age and/or gender, and then entertain efficient support of multidimensional recommenders on the selected attributes. RECATHON employs a multi-dimensional index structure for each built recommender that can be accessed using novel query execution algorithms to support efficient retrieval for recommender queries. Experimental results based on an actual prototype of RECATHON, built inside Postgre SQL, using real Movie Lens and Foursquare data show that RECATHON exhibits real time performance for large-scale multidimensional recommendation.

We identify an opportunity to share data among mobile apps hosted in the cloud, thus helping users improve their mobile experience, while resulting in cost savings for the cloud provider. In this work, we propose a platform for sharing data among mobile apps hosted in the cloud. A “sharing ” is specified by a triple consisting of: (a) a set of data sources to be shared, (b) a set of specified transforma-tions on the shared data, and (c) a staleness (freshness) requirement on the shared data. The platform addresses the following two main challenges: What sharings to admit into the system under a set of specified constraints, how to implement a sharing at a low cost while maintaining the desired level of staleness. We show that reductions in costs are achievable by exploiting the commonalities between the different sharings in the platform. Experimental evaluation is per-formed with a cloud platform containing 25 sharings among mo-bile apps with realistic datasets containing user, social, location and checkin data. Our platform is able to maintain the sharings with very few violations, even under a very high update rate. Our results show that our method results in a cost savings of over 35 % for the cloud provider, while enabling an improved mobile experience for users. 1.

This demo presents Minnesota Traffic Generator (MNTG); an extensible web-based road network traffic generator. MNTG enables its users to generate traffic data at any arbitrary road networks with different traffic generators. Unlike existing traffic generators that require a lot of time/effort to install, configure, and run, MNTG is a web service with a user-friendly interface where users can specify an arbitrary spatial region, select a traffic generator, and submit their traffic generation request. Once the traffic data is generated by MNTG, users can then download and/or visualize the generated data. MNTG can be extended to support: (1) various traffic generators. It is already shipped with the two most common traffic generators, Brinkhoff and BerlinMOD, but other generators can be easily added. (2) various road network sources. It is shipped with U.S. Tiger files and OpenStreetMap, but other sources can be also added. A beta version of MNTG is launched at: http://mntg.cs.umn.edu.

This paper proposes LARS*, a location-aware recommender system that uses location-based ratings to produce recommendations. Traditional recommender systems do not consider spatial properties of users nor items; LARS*, on the other hand, supports a taxonomy of three novel classes of location-based ratings, namely, spatial ratings for non-spatial items, non-spatial ratings for spatial items, and spatial ratings for spatial items. LARS* exploits user rating locations through user partitioning, a technique that influences recommendations with ratings spatially close to querying users in a manner that maximizes system scalability while not sacrificing recommendation quality. LARS* exploits item locations using travel penalty, a technique that favors recommendation candidates closer in travel distance to querying users in a way that avoids exhaustive access to all spatial items. LARS* can apply these techniques separately, or together, depending on the type of location-based rating available. Experimental evidence using large-scale real-world data from both the Foursquare location-based social network and the MovieLens movie recommendation system reveals that LARS* is efficient, scalable, and capable of producing recommendations twice as accurate compared to existing recommendation approaches.

With the advent of online social networks, recommender systems have became crucial for the success of many online applications/services due to their significance role in tailoring these applications to user-specific needs or preferences. Despite their increasing popularity, in general, recommender systems suffer from data sparsity and cold-start problems. To alleviate these issues, in recent years, there has been an upsurge of interest in exploiting social information such as trust relations among users along with the rating data to improve the performance of recommender systems. The main motivation for exploiting trust information in the recommendation process stems from the observation that the ideas we are exposed to and the choices we make are significantly influenced by our social context. However, in large user communities, in addition to trust relations, distrust relations also exist between users. For instance, in Epinions, the concepts of personal “web of trust” and personal “block list” allow users to categorize their friends based on the quality of reviews into trusted and distrusted friends, respectively. Hence, it will be interesting to incorporate this new source of information in recommendation as well. In contrast to the incorporation of trust information in recommendation which is thriving, the potential of explicitly incorporating distrust relations is almost unexplored. In this article, we propose a matrix factorization-based model for recommendation in social rating networks that properly incorporates both trust and distrust relationships aiming to improve the quality of recommendations and mitigate the data sparsity and cold-start users issues. Through experiments on the Epinions dataset, we show that our new algorithm outperforms its standard trust-enhanced or distrust-enhanced counterparts with respect to accuracy, thereby demonstrating the positive effect that incorporation of explicit distrust information can have on recommender systems.

The 21st ACM SIGSPATIAL Conference on Advances in Geographic Information Systems (GIS) was held in November of 2013 in Orlando, Florida. Following the success of last year's event, we organized the second programming contest associated with the conference, called the SIGSPATIAL GIS Cup 2013. The subject of the competition was Geo-fencing, which identifies the qualified point and area pairs using a virtual perimeter for a real-world geographic area. We describe the contest details, and the results, as well as the lessons learned during the process.

In a business setting, the customer value is crucial as it determines how much it is worth spending to acquire a particular customer. Viral marketing techniques leverages social ties among users to help advertising a particular product. Recently, as mobile devices (e.g., smart phones, GPS devices) became ubiquitous, location-based social networking websites (e.g., Gowalla, BrightKite, Foursquare) are getting more and more popular. Along with location-based social networking services being prominent, new kind of data came into play besides the traditional social networking data: (1) Spatial data: represents the users geo-locations, venues geo-locations and information about users visiting different venues. (2) Users Opinions data: represents how much a user likes the venues she visits (e.g., Alice visited restaurant A and gave it a rating of five over five). In this paper, we present PLUTUS; a framework that assists venues (e.g., restaurant, gym, shopping mall) owners in growing their business. To recommend the best set of customers, PLUTUS takes three main aspects into consideration: (1) Social aspect, (2) Spatial aspect, and (3) Users opinions aspect. To this end, PLUTUS proposes two main algorithms: (1) Profit Calculation: It is responsible of calculating the total profit that a user u may add to a venue v taking into account the social, spatial, and user opinions aspects. (2) Profit Maximization: This algorithm is used to maximize the total profit of a given venue. We evaluated PLUTUS using real data set extracted from an existing Location-based Social Networking website, Foursquare. The results show that Plutus achieves higher estimated profit and more efficient profit calculation than naive marketing algorithms.

In the first part, we start by giving a quick overview of social networking services (e.g., Facebook, Twitter), their evolution, and how they impact the society. Similarly, we explain, through examples and case studies, how the widespread of mobile devices changed the computing paradigm in a way that impacted our daily life. In the second part, we present state-of-the-art research that lies within the intersection of both social networking and mobility, from the following perspectives: 1) geosocial queries; 2) geosocial recommendation; 3) geosocial analytics; 4) geosocial visualization; and 5) geosocial crowdsourcing.

This tutorial presents the state-of-the-art research that lies at the intersection of two hot topics in the data management community: (1) social networking and (2) mobility. In this tutorial, we give an overview of existing research work, systems, and applications related to both social networking and mobility. In addition, we introduce several resources (i.e., datasets, software tools) as well as a list of promising research directions.

In this paper, we demonstrate RecDB; a full-fledged database system that provides personalized recommendation to users. We implemented RecDB using an existing open source database system PostgreSQL, and we demonstrate the effectiveness of RecDB using two existing recommendation applications (1) Restaurant Recommendation, (2) Movie Recommendation. To make the demo even more interactive, we showcase a novel application that recommends research papers presented at VLDB 2013 to the conference attendees based on their publication history in DBLP.

Horton+ is a graph query processing system that executes declarative reachability queries on a partitioned attributed multi-graph. It employs a query language, query optimizer, and a distributed execution engine. The query language expresses declarative reachability queries, and supports closures and predicates on node and edge attributes to match graph paths. We introduce three algebraic operators, select, traverse, and join, and a query is compiled into an execution plan containing these operators. As reachability queries access the graph elements in a random access pattern, the graph is therefore maintained in the main memory of a cluster of servers to reduce query execution time. We develop a distributed execution engine that processes a query plan in parallel on the graph servers. Since the query language is declarative, we build a query optimizer that uses graph statistics to estimate predicate selectivity. We experimentally evaluate the system performance on a cluster of 16 graph servers using synthetic graphs as well as a real graph from an application that uses reachability queries. The evaluation shows (1) the efficiency of the optimizer in reducing query execution time, (2) system scalability with the size of the graph and with the number of servers, and (3) the convenience of using declarative queries.

Most recommendation methods (e.g., collaborative filtering) consist of (1) a computationally intense offline phase that computes a recommender model based on users' opinions of items, and (2) an online phase consisting of SQL-based queries that use the model (generated offline) to derive user preferences and provide recommendations for interesting items. Current application usage trends require a completely online recommender process, meaning the recommender model must update in real time as new opinions enter the system. To tackle this problem, we propose RecStore, a DBMS storage engine module capable of efficient online model maintenance. Externally, models managed by RecStore behave as relational tables, thus existing SQL-based recommendation queries remain unchanged while gaining online model support. RecStore maintains internal statistics and data structures aimed at providing efficient incremental updates to the recommender model, while employing an adaptive strategy for internal maintenance and load shedding to realize a balance between efficiency in updates or query processing based on system workloads. RecStore is also extensible, supporting a declarative syntax for defining recommender models. The efficacy of RecStore is demonstrated by providing the implementation details of three state-of-the-art collaborative filtering models. We provide an extensive experimental evaluation of a prototype of RecStore, built inside the storage engine of PostgreSQL, using a real-life recommender system workload.

This paper proposes LARS, a location-aware recommender system that uses location-based ratings to produce recommendations. Traditional recommender systems do not consider spatial properties of users nor items, LARS, on the other hand, supports a taxonomy of three novel classes of location-based ratings, namely, spatial ratings for non-spatial items, non-spatial ratings for spatial items, and spatial ratings for spatial items. LARS exploits user rating locations through user partitioning, a technique that influences recommendations with ratings spatially close to querying users in a manner that maximizes system scalability while not sacrificing recommendation quality. LARS exploits item locations using travel penalty, a technique that favors recommendation candidates closer in travel distance to querying users in a way that avoids exhaustive access to all spatial items. LARS can apply these techniques separately, or in concert, depending on the type of location-based rating available. Experimental evidence using large-scale real-world data from both the Foursquare location-based social network and the Movie Lens movie recommendation system reveals that LARS is efficient, scalable, and capable of producing recommendations twice as accurate compared to existing recommendation approaches.

Graphs are used in many large-scale applications, such as social networking. The management of these graphs poses new challenges as such graphs are too large for a single server to manage efficiently. Current distributed techniques such as map-reduce and Pregel are not well-suited to processing interactive ad-hoc queries against large graphs. In this paper we demonstrate Horton, a distributed interactive query execution engine for large graphs. Horton defines a query language that allows the expression of regular language reach ability queries and provides a query execution engine with a query optimizer that allows interactive execution of queries on large distributed graphs in parallel. In the demo, we show the functionality of Horton managing a large graph for a social networking application called Codebook, whose graph represents data on software components, developers, development artifacts such as bug reports, and their interactions in large software projects.

This demo presents Sindbad; a location-based social networking system. Sindbad supports three new services beyond traditional social networking services, namely, location-aware news feed, location-aware recommender, and location-aware ranking. These new services not only consider social relevance for its users, but they also consider spatial relevance. Since location-aware social networking systems have to deal with large number of users, large number of messages, and user mobility, efficiency and scalability are important issues. To this end, Sindbad encapsulates its three main services inside the query processing engine of PostgreSQL. Usage and internal functionality of Sindbad, implemented with PostgreSQL and Google Maps API, are demonstrated through user (i.e., web/phone) and system analyzer GUI interfaces, respectively.