Adaptive random forests for evolving data stream classification

Random forests is currently one of the most used machine learning algorithms in the non-streaming (batch) setting. This preference is attributable to its high learning performance and low demands with respect to input preparation and hyper-parameter tuning. However, in the challenging context of evolving data streams, there is no random forests algorithm that can be considered state-of-the-art in comparison to bagging and boosting based algorithms. In this work, we present the adaptive random forest (ARF) algorithm for classification of evolving data streams. In contrast to previous attempts of replicating random forests for data stream learning, ARF includes an effective resampling method and adaptive operators that can cope with different types of concept drifts without complex optimizations for different data sets. We present experiments with a parallel implementation of ARF which has no degradation in terms of classification performance in comparison to a serial implementation, since trees and adaptive operators are independent from one another. Finally, we compare ARF with state-of-the-art algorithms in a traditional test-then-train evaluation and a novel delayed labelling evaluation, and show that ARF is accurate and uses a feasible amount of resources.     

ONTRACK: Dynamically adapting music playback to support navigation

Listening to music on personal, digital devices whilst mobile is an enjoyable, everyday activity. We explore a scheme for exploiting this practice to immerse listeners in navigation cues. Our prototype, ONTRACK, continuously adapts audio, modifying the spatial balance and volume to lead listeners to their target destination. First we report on an initial lab-based evaluation that demonstrated the approach’s efficacy: users were able to complete tasks within a reasonable time and their subjective feedback was positive. Encouraged by these results we constructed a handheld prototype. Here, we discuss this implementation and the results of field-trials. These indicate that even with a low-fidelity realisation of the concept, users can quite effectively navigate complicated routes.

---

Using the MC68901 multifunction peripheral

First-generation microprocessor peripheral chips were each dedicated to specific functions (e.g. serial interface, parallel interface, timer). Some of today's peripherals are designed to carry out a wide range of functions. The 68901 multifunction peripheral (MFP), for example, combines the functions of a serial interface, a parallel interface, a timer and an interrupt generator. It is not possible to go into detail about the MFP here — its data manual is the size of a small book. Instead, this note illustrates the design of a complete single-board microcomputer built around the 68901 and the 68008 microprocessor.     

Aquatic passive sampling of herbicides on naked particle loaded membranes: accelerated measurement and empirical estimation of kinetic parameters

Water-sampler equilibrium partitioning coefficients and aqueous boundary layer mass transfer coefficients for atrazine, diuron, hexazionone and fluometuron onto C18 and SDB-RPS Empore disk-based aquatic passive samplers have been determined experimentally under a laminar flow regime (Re = 5400). The method involved accelerating the time to equilibrium of the samplers by exposing them to three water concentrations, decreasing stepwise to 50% and then 25% of the original concentration. Assuming first-order Fickian kinetics across a rate-limiting aqueous boundary layer, both parameters are determined computationally by unconstrained nonlinear optimization. In addition, a method of estimation of mass transfer coefficients--therefore sampling rates--using the dimension-less Sherwood correlation developed for laminar flow over a flat plate is applied. For each of the herbicides, this correlation is validated to within 40% of the experimental data.The study demonstrates that for trace concentrations (sub 0.1 microg/L) and these flow conditions, a naked Empore disk performs well as an integrative sampler over short deployments (up to 7 days) for the range of polar herbicides investigated. The SDB-RPS disk allows a longer integrative period than the C18 disk due to its higher sorbent mass and/or its more polar sorbent chemistry. This work also suggests that for certain passive sampler designs, empirical estimation of sampling rates may be possible using correlations that have been available in the chemical engineering literature for some time.     

Sustainability at home: Policy measures for energy-efficient appliances

Residential energy consumption is now an important component of total energy consumption and its related emissions. In addition, this consumption has significant potential for growth in both developed and developing countries, as average incomes increase, and domestic appliance numbers rise. Reduction in this energy use may be achieved both through conservation measures, and through increased efficiency in its use. This paper focuses on energy efficiency in appliances—those many devices through which household energy is consumed. The policy instruments available to promote the uptake of more efficient devices, and issues associated with their use, are reviewed, drawing on developed country experiences to date. The instruments available are more limited than in other energy applications, and largely comprise information dissemination, forms of subsidy, and regulation. The last is commonly used in the form of minimum energy performance standards. Assessment of the three instrument types and issues with their use suggests that regulation is the most important measure—albeit with several qualifications on the manner of its use.     

Buying Power

The global push toward finalizing and implementing the next generation of connected smart machinery—the Industry 4.0 protocols—begs the question: how will these new standards influence equipment purchasing decisions for plastics manufacturers?     

The role of multi-tenure reserve networks in improving reserve design and connectivity

Multi-tenure reserve networks aim to connect areas managed for biodiversity conservation across public and private land. This paper seeks to determine to what extent multi-tenure reserve networks improve the reserve design and connectivity of the public protected area estate, using three networks in southeastern Australia as case studies. Network configuration varied considerably and those networks with generally larger parcels tended to be better connected. On average, public land components were larger than private land components in all networks. Two networks had 18 components physically adjoining other network components while another had only 6 components adjoining. Importantly for two of the networks, the average distance between the nearest neighbouring component was significantly less than average distances between public protected areas in the subregion. Thus these multi-tenure reserve networks acted to enhance the existing public protected area estate by increasing the potential linkages in the landscape and therefore the viability of individual public protected areas.     

The TPTP Problem Library and Associated Infrastructure

This paper describes the First-Order Form (FOF) and Clause Normal Form (CNF) parts of the TPTP problem library, and the associated infrastructure. TPTP v3.5.0 was the last release containing only FOF and CNF problems, and thus serves as the exemplar. This paper summarizes the history and development of the TPTP, describes the structure and contents of the TPTP, and gives an overview of TPTP related projects and tools.     

A modularized framework for solving an economic–environmental power generation mix problem

This paper presents a modularized simulation modelling framework for evaluating the impacts on economic cost and CO₂ emissions resulting from the introduction of a solid oxide fuel cell (SOFC) system into the existing mix of centralized power generation technologies in Japan. The framework is comprised of three parts: a dual‐objective linear programming model that solves the generation best‐mix problem for the existing power generation technologies; a nonlinear SOFC system model in which the economic cost and CO₂ emissions by the SOFC system can be calculated; and the Queuing Multi‐Objective Optimizer (QMOO), a multi‐objective evolutionary algorithm (MOEA) developed at the EPFL in Switzerland as the overall optimizer of the combined power supply system. Thus, the framework integrates an evolutionary algorithm that is more suitable for handling nonlinearities with a calculus‐based method that is more efficient in solving linear programming problems. Simulation experiments show that the framework is successful in solving the stated problem. Moreover, the three components of the modularized framework can be interconnected through a platform‐independent model integration environment. As a result, the framework is flexible and scalable, and can potentially be modified and/or integrated with other models to study more complex problems. Copyright © 2004 John Wiley & Sons, Ltd.