Geotagged data (e.g. images or news items) have empowered various important applications, e.g., search engines and news agencies. However, the lack of available geotagged data significantly reduces the impact of such applications. Meanwhile, existing geotagging approaches rely on the existence of prior knowledge, e.g., accurate training dataset for machine learning techniques. This paper presents Stella; a crowdsourcing framework for image geotagging. The high accuracy of Stella is resulted by being able to recruit workers near the image location even without knowing its location. In addition, Stella also return its confidence about the reported location to help users in understanding the result quality. Experimental evaluation shows that Stella consistently geotags an image with an average of 95% accuracy and 90% of confidence.

Database systems use many pointer-based data structures, including hash tables and B+-trees, which require extensive "pointer-chasing." Each pointer dereference, e.g., during a hash probe or a B+-tree traversal, can result in a CPU cache miss, stalling the CPU. Recent work has shown that CPU stalls due to main memory accesses are a significant source of overhead, even for cache-conscious data structures, and has proposed techniques to reduce this overhead, by hiding memory-stall latency. In this work, we compare and contrast the state-of-the-art approaches to reduce CPU stalls due to cache misses for pointer-intensive data structures. We present an in-depth experimental evaluation and a detailed analysis using four popular data structures: hash table, binary search, Masstree, and Bw-tree. Our focus is on understanding the practicality of using coroutines to improve throughput of such data structures. The implementation, experiments, and analysis presented in this paper promote a deeper understanding of how to exploit coroutines-based approaches to build highly efficient systems.

This paper demonstrates Stella ; an efficient crowdsourcing-based geotagging framework for any types of objects. In this demonstration, we showcase the effectiveness of Stella in geotagging images via two different scenarios: (1) we provide a graphical interface to show the process of a geotagging process that have been done by using Amazon Mechanical Turk, (2) we seek help from the conference attendees to propose an image to be geotagged or to help us geotag an image by using our application during the demonstration period. At the end of the demonstration period, we will show the geotagging result.

This paper introduces HadoopViz; a MapReduce-based framework for visualizing big spatial data. HadoopViz has three unique features that distinguish it from other techniques. (1) It exposes an extensible interface which allows users to define a new visualization types, e.g., scatter plot, road network, or heat map, by defining five abstract functions, without delving into the implementation details of the MapReduce algorithms. As it is open source, HadoopViz allows algorithm designers to focus on how the data should be visualized rather than performance or scalability issues. (2) HadoopViz is capable of generating big images with giga-pixel resolution by employing a three-phase technique, partition-plot-merge. (3) HadoopViz provides a smoothing functionality which can fuse nearby records together as the image is plotted. This makes it capable of generating more types of images with high quality as compared to existing work. Experimental results on real datasets of up to 14 Billion points show the extensibility, scalability, and efficiency of HadoopViz to handle different visualization types of spatial big data.

The recent wide popularity of microblogs (e.g., tweets, online comments) has empowered various important applications, including, news delivery, event detection, market analysis, and target advertising. A core module in all these applications is a frequent/trending query processor that aims to find out those topics that are highly frequent or trending in the social media through posted microblogs. Unfortunately current attempts for such core module suffer from several drawbacks. Most importantly, their narrow scope, as they focus only on solving trending queries for a very special case of localized and very recent microblogs. This paper presents GARNET; a holistic system equipped with one-stop efficient and scalable solution for supporting a generic form of context-aware frequent and trending queries on microblogs. GARNET supports both frequent and trending queries, any arbitrary time interval either current, recent, or past, of fixed granularity, and having a set of arbitrary filters over contextual attributes. From a system point of view, GARNET is very appealing and industry-friendly, as one needs to realize it once in the system. Then, a myriad of various forms of trending and frequent queries are immediately supported. Experimental evidence based on a real system prototype of GARNET and billions of real Twitter data show the scalability and efficiency of GARNET for various query types.

This demonstration presents HadoopViz; an extensible MapReduce-based system for visualizing Big Spatial Data. HadoopViz has two main unique features that distinguish it from other techniques. (1) It provides an extensible interface that allows users to visualize various types of data by defining five abstract functions, without delving into the details of the MapReduce algorithms. We show how it is used to create four types of visualizations, namely, scatter plot, road network, frequency heat map , and temperature heat map. (2) HadoopViz is capable of generating big images with giga-pixel resolution by employing a three-phase approach of partitioning, rasterize , and merging. HadoopViz generates single and multi-level images, where the latter allows users to zoom in/out to get more/less details. Both types of images are generated with a very high resolution using the extensible and scalable framework of HadoopViz.