When things matter

Demand-side management technology is a key element of the proposed smart grid, which will help utilities make more efficient use of their generation assets by reducing consumers’ energy demand during peak load periods. However, although some modern appliances can respond to price signals from the utility companies, there is a vast stock of older appliances that cannot. For such appliances, utilities must infer what appliances are operating in a home, given only the power signals on the main feeder to the home (i.e., the home’s power consumption must be disaggregated into individual appliances). We report on an in-depth investigation of multi-label classification algorithms for disaggregating appliances in a power signal. A systematic review of this research topic shows that this class of algorithms has received little attention in the literature, even though it is arguably a more natural fit to the disaggregation problem than the traditional single-label classifiers used to date. We examine a multi-label meta-classification framework (RA ${k}$ EL), and a bespoke multi-label classification algorithm (ML ${k}$ NN), employing both time-domain and wavelet-domain feature sets. We test these classifiers on two real houses from the Reference Energy Disaggregation Dataset. We found that the multilabel algorithms are effective and competitive with published results on the datasets.

Toward Non-Intrusive Load Monitoring via Multi-Label Classification

Various systems and methods are provided that display various interactive user interfaces containing interactive maps. The user may use the interactive map to perform searches for data objects, toggle different layers associated with different types of objects for display, generate heatmaps based upon a grid and/or data object shapes, copy data between different interactive maps, and/or define filters to filter the displayed data. In response to user interaction with the map at a client system, tile layers are generated using server-side components, assembled into map tiles, and transmitted to client-side components to be displayed to the user. The tile layers may include a base tile layer, vector tile layer, selection tile layer, and inactive tile layer.

Interactive geospatial map

Advances in hardware technology have lead to an ability to collect data with the use of a variety of sensor technologies. In particular sensor notes have become cheaper and more efficient, and have even been integrated into day-to-day devices of use, such as mobile phones.This has lead to a much larger scale of applicability and mining of sensor data sets. The human-centric aspect of sensor data hascreated tremendous opportunities in integrating social aspects of sensor data collection into the mining process. Managing and Mining Sensor Data is a contributed volume by prominent leaders in this field, targeting advanced-level students in computer science as a secondary text book or reference. Practitioners and researchers working in this field will also find this book useful.

Managing and Mining Sensor Data

In cases where the right mouse click operation to access a context menu is disabled, the techniques described in the disclosure may provide access to the context menu through a long click of the left mouse button. For example, the user may click and hold the left mouse button for at least a threshold period of time, and in response thereto, the user interface is updated to display the context menu. To indicate to the user how long it would take for the context menu to appear, an icon can be displayed next to the cursor, or as a part of the cursor. The icon can show the relative time left before the context menu is displayed (e.g., an elapsed and/or remaining portion of the threshold period of time, etc.).

Long click display of a context menu

One embodiment is a method that builds a model of multi-dimensional sequence data in real-time with cuboids that aggregate the multi-dimensional sequence data over both patterns and dimensions. The model provides search results for a query.

Modeling multi-dimensional sequence data over streams

Many modern applications, including online financial feeds, tag-based mass transit systems and RFID-based supply chain management systems transmit real-time data streams. There is a need for event stream processing technology to analyze this vast amount of sequential data to enable online operational decision making. Existing techniques such as traditional online analytical processing (OLAP) systems are not designed for real-time pattern-based operations, while state-of-the-art Complex Event Processing (CEP) systems designed for sequence detection do not support OLAP operations. We propose a novel E-Cube model which combines CEP and OLAP techniques for efficient multi-dimensional event pattern analysis at different abstraction levels. Our analysis of the interrelationships in both concept abstraction and pattern refinement among queries facilitates the composition of these queries into an integrated E-Cube hierarchy. Based on this E-Cube hierarchy, strategies of drill-down (refinement from abstract to more specific patterns) and of roll-up (generalization from specific to more abstract patterns) are developed for the efficient workload evaluation. Our proposed execution strategies reuse intermediate results along both the concept and the pattern refinement relationships between queries. Based on this foundation, we design a cost-driven adaptive optimizer called Chase, that exploits the above reuse strategies for optimal E-Cube hierarchy execution. Our experimental studies comparing alternate strategies on a real world financial data stream under different workload conditions demonstrate the superiority of the Chase method. In particular, our Chase execution in many cases performs ten fold faster than the state-of-the art strategy for real stock market query workloads.

E-Cube: multi-dimensional event sequence analysis using hierarchical pattern query sharing

A wide range of Web applications retrieve desired information from remote XML data sources across the Internet, which is usually costly due to transmission delays for large volumes of data. Therefore we propose to apply the ideas of semantic caching to XML query processing systems [2], in particular the XQuery engine. Semantic caching [3] implies view-based query answering and cache management. While it is well studied in the traditional database context, query containment for XQuery is left unexplored due to its complexity coming with the powerful expressiveness of hierarchy, recursion and result construction. We hence have developed the first solution for XQuery processing using cached views.We exploit the connections between XML and tree automata, and use subtype relations between two regular expression types to tackle the XQuery containment mapping problem. Inspired by XDuce [1], which explores the use of tree-automata-based regular expression types for XML processing, we have designed a containment mapping process to incorporate type inference and subtyping mechanisms provided by XDuce to establish containment mappings between regular-expression-type-based pattern variables of two queries. We have implemented a semantic caching system called XCache (see Figure 1), to realize the proposed containment and rewriting techniques for XQuery.The main modules of XCache include: (1) Query Decomposer. An input query is is decomposed into source-specific subqueries explicitly represented by matching patterns and return structures. (2) Query Pattern Register. By registering a few queries into semantic regions, we warm up XCache at its initialization phase. (3) Query Containment Mapper. The XDuce subtyper is incorporated into the containment mapper for establishing query containment mappings between variables of a new query and each cached query. (4) Query Rewriter. We implement the classical bucket algorithm and further apply heuristics to decide on an "optimal" rewriting plan if several valid ones exist. (5) Replacement Manager. We free space for new regions by both complete and partial replacement. (6) Region Coalescer. We apply a coalescing strategy to control the region granularity over time.

XCache: a semantic caching system for XML queries

A method for sharing window-based joins includes slicing window states of a join operator into smaller window slices, forming a chain of sliced window joins from the smaller window slices, and reducing by pipelining a number of the sliced window joins. The method further includes pushing selections down into chain of sliced window joins for computation sharing among queries with different window sizes. The chain buildup of the sliced window joins includes finding a chain of the sliced window joins with respect to one of memory usage or processing usage.

Multi-Query Optimization of Window-Based Stream Queries

Modern stream applications such as sensor monitoring systems and publish/subscription services necessitate the handling of large numbers of continuous queries specified over high volume data streams. Efficient sharing of computations among multiple continuous queries, especially for the memory- and CPU-intensive window-based operations, is critical. A novel challenge in this scenario is to allow resource sharing among similar queries, even if they employ windows of different lengths. This paper first reviews the existing sharing methods in the literature, and then illustrates the significant performance shortcomings of these methods.This paper then presents a novel paradigm for the sharing of window join queries. Namely we slice window states of a join operator into fine-grained window slices and form a chain of sliced window joins. By using an elaborate pipelining methodology, the number of joins after state slicing is reduced from quadratic to linear. This novel sharing paradigm enables us to push selections down into the chain and flexibly select subsequences of such sliced window joins for computation sharing among queries with different window sizes. Based on the state-slice sharing paradigm, two algorithms are proposed for the chain buildup. One minimizes the memory consumption while the other minimizes the CPU usage. The algorithms are proven to find the optimal chain with respect to memory or CPU usage for a given query workload. We have implemented the slice-share paradigm within the data stream management system CAPE. The experimental results show that our strategy provides the best performance over a diverse range of workload settings among all alternate solutions in the literature.

/pdf/state-slice-new-paradigm-of-multi-query-optimization-of-4ubofy49os.pdf

Song Wang

Papers

Modeling multi-dimensional sequence data over streams

E-Cube: multi-dimensional event sequence analysis using hierarchical pattern query sharing

XCache: a semantic caching system for XML queries

Multi-Query Optimization of Window-Based Stream Queries

State-slice: new paradigm of multi-query optimization of window-based stream queries