Supervised Aggregative Feature Extraction for Big Data Time Series Regression

In many applications, and especially those where batch processes are involved, a target scalar output of interest is often dependent on one or more time series of data. With the exponential growth in data logging in modern industries such time series are increasingly available for statistical modeling in soft sensing applications. In order to exploit time series data for predictive modelling, it is necessary to summarise the information they contain as a set of features to use as model regressors. Typically this is done in an unsupervised fashion using simple techniques such as computing statistical moments, principal components or wavelet decompositions, often leading to significant information loss and hence suboptimal predictive models. In this paper, a functional learning paradigm is exploited in a supervised fashion to derive continuous, smooth estimates of time series data (yielding aggregated local information), while simultaneously estimating a continuous shape function yielding optimal predictions. The proposed Supervised Aggregative Feature Extraction (SAFE) methodology can be extended to support nonlinear predictive models by embedding the functional learning framework in a Reproducing Kernel Hilbert Spaces setting. SAFE has a number of attractive features including closed form solution and the ability to explicitly incorporate first and second order derivative information. Using simulation studies and a practical semiconductor manufacturing case study we highlight the strengths of the new methodology with respect to standard unsupervised feature extraction approaches.

Ultracold fluorine production via Doppler cooled BeF

Large parts of the periodic table cannot be cooled by current laser-based methods. We investigate whether zero energy fragmentation of laser cooled fluorides is a potential source of ultracold fluorine atoms. We report new ab initio calculations on the lowest electronic states of the BeF diatomic molecule including spin-orbit coupling, the calculated minima for the valence electronic states being within 1 pm of the spectroscopic values. A four colour cooling scheme based on the A2? ? X2S+ transition is shown to be feasible for this molecule. Multi-Reference Configuration Interaction (MRCI) potentials of the lowest energy Rydberg states are reported for the first time and found to be in good agreement with experimental data. A series of multi-pulse excitation schemes from a single rovibrational level of the cooled molecule are proposed to produce cold fluorine atoms.

Soil conditioning of sand for EPB applications:A laboratory research

EPB tunnelling requires the application of soil conditioning to increase its field of applicability particularly for cohesionless soils. Choosing the most suitable conditioning set for the various soils requires the use of a feasible laboratory test which can permit to define the characteristics of the conditioned soils and provide measurable data. A series of tests has been carried out using a laboratory screw conveyor device which was designed for this purpose and which simulates the extraction of the spoil from a pressure chamber in a similar way as in EPB tunnelling. The tested soils were medium-grain sands with varying amounts of silt and the tested conditioned mixtures were obtained with different water contents and amounts of foam. A simple slump test was also used to analyze the global characteristics of the conditioned soils. The test has shown that the proposed laboratory procedure permits a quantitative comparison to be made between different conditioning amounts and agents on the basis of measurable parameters. �� 2007 Elsevier Ltd. All rights reserved.

Error sensitive historical GIS: Identifying areal interpolation errors in time series data

Historical GIS has the potential to re-invigorate our use of statistics from historical censuses and related sources. In particular, areal interpolation can be used to create long-run time-series of spatially detailed data that will enable us to enhance significantly our understanding of geographical change over periods of a century or more. The difficulty with areal interpolation, however, is that the data that it generates are estimates which will inevitably contain some error. This paper describes a technique that allows the automated identification of possible errors at the level of the individual data values.

Limitations of instantaneous water quality sampling in surface-water catchments: Comparison with near-continuous phosphorus time-series data.

The validity of load estimates from intermittent, instantaneous grab sampling is dependent on adequate spatial coverage by monitoring networks and a sampling frequency that re?ects the variability in the system under study. Catchments with a ?ashy hydrology due to surface runoff pose a particular challenge as intense short duration rainfall events may account for a signi?cant portion of the total diffuse transfer of pollution from soil to water in any hydrological year. This can also be exacerbated by the presence of strong background pollution signals from point sources during low flows. In this paper, a range of sampling methodologies and load estimation techniques are applied to phosphorus data from such a surface water dominated river system, instrumented at three sub-catchments (ranging from 3 to 5 km2 in area) with near-continuous monitoring stations. Systematic and Monte Carlo approaches were applied to simulate grab sampling using multiple strategies and to calculate an estimated load, Le based on established load estimation methods. Comparison with the actual load, Lt, revealed signi?cant average underestimation, of up to 60%, and high variability for all feasible sampling approaches. Further analysis of the time series provides an insight into these observations; revealing peak frequencies and power-law scaling in the distributions of P concentration, discharge and load associated with surface runoff and background transfers. Results indicate that only near-continuous monitoring that re?ects the rapid temporal changes in these river systems is adequate for comparative monitoring and evaluation purposes. While the implications of this analysis may be more tenable to small scale ?ashy systems, this represents an appropriate scale in terms of evaluating catchment mitigation strategies such as agri-environmental policies for managing diffuse P transfers in complex landscapes.

Income ineqaulity from 1960-2012: a brief time-series history of capital and labour

This commentary examines two principal forms of inequality and their evolution since the 1960s: the division of national income between capital and labour, and the share of total income held by the top 1 per cent of earners. Trends are linked to current discussions of inequality drivers such as financialisation, and a brief time-series analysis of the effects of trade and financial sector growth on top incomes is presented.

���If I Were Nick���: Men���s Responses to an Interactive Video Drama Series���� to Support Smoking Cessation

Background: Men continue to smoke in greater numbers than women; however, few interventions have been developed and tested to support men���s cessation. Men also tend to rely on quitting strategies associated with stereotypical manliness, such as willpower, stoicism and independence, but may lack the self���efficacy skills required to sustain a quit. In this article we describe the development of and reception to an interactive video drama (IVD) series, composed of 7 brief scenarios, to support and strengthen men���s smoking cessation efforts. The value of IVD in health promotion is predicated on the evidence that viewers engage with the material when they are presented characters with whom they can personally identify. The video dramatizes the challenges unfolding in the life of the main character, Nick, on the first day of his quit and models the skills necessary to embark upon a sustainable quit.��<br/>Objective: The objective was to describe men���s responses to the If I were Nick IVD series as part of a pilot study of QuitNow MenTM, an innovative smoking cessation website designed for men. Specific objectives were to explore the resonance of the main character of the IVD series with end���users, and men���s perceptions of the effectiveness of the IVD series for supporting their quit self���management.��<br/>Methods: Seven brief IVD scenarios were developed, filmed with a professional actor and uploaded to a new online smoking cessation website, QuitNow MenTM.����A sample of 117 men who smoked were recruited into the study and provided baseline data prior to access to the QuitNow MenTM website for a 6 month period. During this time, 47 men chose to view the IVDs. Their responses to questions about the IVDs were collected in 3���month and 6���month online follow���up surveys and analyzed using descriptive statistics.��<br/>Findings: The majority of participants indicated they related to the main character, Nick. Participants who ���strongly agreed��� they could relate to Nick perceived significantly higher levels of support from the IVDs than the ���neutral��� and ���disagree��� groups (P &lt;.001, d =2.0, P &lt;.001 d =3.1). The ���agree��� and ���neutral��� groups were significantly higher on rated support from the videos than the ���disagree��� (P &lt;.001 d =2.2, P =.01 d = 1.5). Participants��� perception of the main character was independent of participant age, education attainment or previous quit attempts.��<br/>Conclusions: The findings suggest that IVD interventions may be an important addition to men���s smoking cessation programs. Given that the use of IVD scenarios in health promotion is in its infancy, the positive outcomes from this pilot study signal the potential for IVD and warrant ongoing evaluation in smoking cessation and, more generally, men���s health promotion.����

Open Source Telecommunications Framework for Time-Series Data from Smart Grid Sensors.

This paper presents a framework for a telecommunications interface which allows data from sensors embedded in Smart Grid applications to reliably archive data in an appropriate time-series database. The challenge in doing so is two-fold, firstly the various formats in which sensor data is represented, secondly the problems of telecoms reliability. A prototype of the authors' framework is detailed which showcases the main features of the framework in a case study featuring Phasor Measurement Units (PMU) as the application. Useful analysis of PMU data is achieved whenever data from multiple locations can be compared on a common time axis. The prototype developed highlights its reliability, extensibility and adoptability; features which are largely deferred from industry standards for data representation to proprietary database solutions. The open source framework presented provides link reliability for any type of Smart Grid sensor and is interoperable with existing proprietary database systems, and open database systems. The features of the authors' framework allow for researchers and developers to focus on the core of their real-time or historical analysis applications, rather than having to spend time interfacing with complex protocols.