Probabilistic models over ordered partitions with applications in document ranking and collaborative filtering

Ranking is an important task for handling a large amount of content. Ideally, training data for supervised ranking would include a complete rank of documents (or other objects such as images or videos) for a particular query. However, this is only possible for small sets of documents. In practice, one often resorts to document rating, in that a subset of documents is assigned with a small number indicating the degree of relevance. This poses a general problem of modelling and learning rank data with ties. In this paper, we propose a probabilistic generative model, that models the process as permutations over partitions. This results in super-exponential combinatorial state space with unknown numbers of partitions and unknown ordering among them. We approach the problem from the discrete choice theory, where subsets are chosen in a stagewise manner, reducing the state space per each stage significantly. Further, we show that with suitable parameterisation, we can still learn the models in linear time. We evaluate the proposed models on two application areas: (i) document ranking with the data from the recently held Yahoo! challenge, and (ii) collaborative filtering with movie data. The results demonstrate that the models are competitive against well-known rivals.

Learning from ordered sets and applications in collaborative ranking

Ranking over sets arise when users choose between groups of items. For example, a group may be of those movies deemed 5 stars to them, or a customized tour package. It turns out, to model this data type properly, we need to investigate the general combinatorics problem of partitioning a set and ordering the subsets. Here we construct a probabilistic log-linear model over a set of ordered subsets. Inference in this combinatorial space is highly challenging: The space size approaches (N!/2)6.93145N+1 as N approaches infinity. We propose a split-and-merge Metropolis-Hastings procedure that can explore the state-space efficiently. For discovering hidden aspects in the data, we enrich the model with latent binary variables so that the posteriors can be efficiently evaluated. Finally, we evaluate the proposed model on large-scale collaborative filtering tasks and demonstrate that it is competitive against state-of-the-art methods.

Intelligent traffic light control of isolated intersections using machine learning methods

Traffic congestion is one of the major problems in modern cities. This study applies machine learning methods to determine green times in order to minimize in an isolated intersection. Q-learning and neural networks are applied here to set signal light times and minimize total delays. It is assumed that an intersection behaves in a similar fashion to an intelligent agent learning how to set green times in each cycle based on traffic information. Here, a comparison between Q-learning and neural network is presented. In Q-learning, considering continuous green time requires a large state space, making the learning process practically impossible. In contrast to Q-learning methods, the neural network model can easily set the appropriate green time to fit the traffic demand. The performance of the proposed neural network is compared with two traditional alternatives for controlling traffic lights. Simulation results indicate that the application of the proposed method greatly reduces the total delay in the network compared to the alternative methods.

The use of ground rules in investigative interviews with children : a synthesis and call for research

Guidelines for conducting investigative interviews with children often include instructions that explain the conversational rules of the interview. Despite the widespread and international use of such instructions (also referred to as "ground rules"), the body of research characterizing children's understanding of these rules and documenting the impact of instruction on memory reports is relatively small. We review the use of ground rules in investigative interviews, the developmental differences that likely underlie children's ability to make sense of these rules, and research pertaining to the effects of the ground rules commonly included in interview guidelines on the reports of 3- to 13-year-old children. We then present a study space analysis concerning the five ground rules reviewed: (a) a statement about interviewer naïveté regarding the target events, (b) instructions to tell the interviewer when a mistake has been made, (c) cautions that some questions may be repeated, and instructions to say (d) "I don't understand" and (e) "I don't know." The results demonstrate obvious gaps in this body of literature, with only the "I don't know" ground rule having received significant attention. In addition to exploring how individual rules impact interview performance, we encourage more process-oriented studies that relate developmental differences in ground rules benefits to the cognitive processes that underlie rule understanding and implementation. Optimally, this research should identify the most suitable format and placement of instruction in interviews and broaden to more often include field studies of child witnesses.

Modelling human preferences for ranking and collaborative filtering: a probabilistic ordered partition approach

Learning preference models from human generated data is an important task in modern information processing systems. Its popular setting consists of simple input ratings, assigned with numerical values to indicate their relevancy with respect to a specific query. Since ratings are often specified within a small range, several objects may have the same ratings, thus creating ties among objects for a given query. Dealing with this phenomena presents a general problem of modelling preferences in the presence of ties and being query-specific. To this end, we present in this paper a novel approach by constructing probabilistic models directly on the collection of objects exploiting the combinatorial structure induced by the ties among them. The proposed probabilistic setting allows exploration of a super-exponential combinatorial state-space with unknown numbers of partitions and unknown order among them. Learning and inference in such a large state-space are challenging, and yet we present in this paper efficient algorithms to perform these tasks. Our approach exploits discrete choice theory, imposing generative process such that the finite set of objects is partitioned into subsets in a stagewise procedure, and thus reducing the state-space at each stage significantly. Efficient Markov chain Monte Carlo algorithms are then presented for the proposed models. We demonstrate that the model can potentially be trained in a large-scale setting of hundreds of thousands objects using an ordinary computer. In fact, in some special cases with appropriate model specification, our models can be learned in linear time. We evaluate the models on two application areas: (i) document ranking with the data from the Yahoo! challenge and (ii) collaborative filtering with movie data. We demonstrate that the models are competitive against state-of-the-arts.

Load frequency control of power systems with electric vehicles and diverse transmission links using distributed functional observers

This paper presents a load frequency control scheme using electric vehicles (EVs) to help thermal turbine units to provide the stability fluctuated by load demands. First, a general framework for deriving a state-space model for general power system topologies is given. Then, a detailed model of a four-area power system incorporating a smart and renewable discharged EVs system is presented. The areas within the system are interconnected via a combination of alternating current/high voltage direct current links and thyristor controlled phase shifters. Based on some recent development on functional observers, novel distributed functional observers are designed, one at each local area, to implement any given global state feedback controller. The designed observers are of reduced order and dynamically decoupled from others in contrast to conventional centralized observer (CO)-based controllers. The proposed scheme can cope better against accidental failures than those CO-based controllers. Extensive simulations and comparisons are given to show the effectiveness of the proposed control scheme.

Characteristics evaluation of a μ-Synthesis H∞ controller for a grid-connected three-phase photovoltaic system

This paper presents a μ-Synthesis H∞ Controller for regulating the switching signal of the inverter connected with a three-phase photovoltaic (PV) system. To facilitate the control design, the system is represented in terms of state space realization with uncertainties. The control design involves selecting proper weighting functions and performing synthesis. The controller order is reduced by Henkel-norm method. Simulations are carried out to evaluate the characteristics of the controller under parametric uncertainties. It is found out that the proposed controller is inherently stable, possesses significantly small tracking error, and preserves nominal performance, robust stability and robust performance for the grid-connected three-phase PV system.

Characteristics evaluation of an LMI-synthesis H∞ controller for a superconducting magnetic energy systems applied in power systems

© 2015 IEEE.This paper presents an H« controller synthesised based on linear matrix inequalities (LMI) for a current source converter based superconducting magnetic energy systems (SMESs) connected to a node of power systems where the regulation of grid current has considered as a control objective. To facilitate the control design, the system is represented in terms of state space realization with uncertainties. The control design involves selecting proper weighting functions and performing LMI-synthesis. The controller order is reduced by Henkel-norm method. Simulations are carried out to evaluate the characteristics of the controller under parametric uncertainties. It is found out that the proposed controller is inherently stable, possesses significantly small tracking error, and preserves robust performance for the SMES.

Enhancement of ion dynamics in organic ionic plastic crystal/PVDF composite electrolytes prepared by co-electrospinning

Electrospun fibers are widely used in composite material design and fabrication due to their high aspect ratio, high surface area and favorable mechanical properties. In this report, novel organic ionic plastic crystal (OIPC) modified poly(vinylidene difluoride) (PVDF) composite fiber membranes were prepared by electrospinning. These composite materials are of interest for application as solid electrolytes in devices including lithium and sodium batteries. The influence of the OIPC, N-ethyl-N-methylpyrrolidinium tetrafluoroborate [C2mpyr][BF4], on the morphology and phase behavior of the composite fibers was investigated by scanning electron microscopy and Fourier transform infrared spectroscopy. Compared with pure electrospun PVDF fibers, which have an electroactive β phase and a small amount of non-polar α phase, the ion-dipole interaction between OIPC and the polymer in the co-electrospun composite system can reduce the non-polar α phase PVDF, resulting in almost entirely electroactive β phase PVDF. Differential scanning calorimetry shows that the ion-dipole interaction between the OIPC and PVDF can also interrupt the crystalline structure of the OIPC. Solid state NMR analysis also reveals different molecular dynamics of the [C2mpyr][BF4] in co-electrospun fibers compared with pure OIPC. Thus, electrospun [C2mpyr][BF4]/PVDF composite fibers that combine both increased ionic conductivity and almost pure β phase PVDF are demonstrated.

An analysis of factors influencing waste minimisation and use of recycled materials for the construction of residential buildings

Residential building construction activities, whether it is new build, repair or maintenance, consumes a large amount of natural resources. This has a negative impact on the environment in the form depleting natural resources, increasing waste production and pollution. Previous research has identified the benefits of preventing or reducing material waste, mainly in terms of the limited available space for waste disposal, and escalating costs associated with landfills, waste management and disposal and their impact on a  building company's profitability. There has however been little development internationally of innovative waste management strategies aimed at reducing the resource requirement of the construction process. The authors contend that embodied energy is a useful indicator of resource value. Using data provided by a regional high-volume residential builder in the State of Victoria, Australia, this paper identifies the various types of waste that are generated from the construction of a typical standard house. It was found that in this particular case, wasted amounts of materials were less than those found previously by others for cases in capital cities (5-10 per cent), suggesting that waste minimisation strategies are successfully being implemented. Cost and embodied energy savings from using materials with recycled content are potentially more beneficial in terms of embodied energy and resource depletion than waste minimisation strategies.

Behind the headlines : an analysis of Australian Commonwealth, State and Territory budget balance numbers

The introduction of accrual accounting principles for government reporting in recent years has been complicated by the presence of two alternative financial reporting frameworks, the Government Finance Statistics (GFS) framework and the Australian professional accounting standard rules. This paper presents the findings from a study of the 2004-05 and 2005-06 annual budgets prepared by the Australian Commonwealth government and the governments of the six Australian States and the two Australian Territories. The study examined the basis for the budget balance numbers (government surplus or deficit) headlined in the budgets of each of the nine governments over the two year period. Findings indicate the adoption of varying measurement bases and a resultant lack of inter-governmental and inter temporal comparability. A number of departures from the measurements
prescribed in the reporting frameworks were also observed.

Stereo correspondence estimation using multiwavelets scale-space representation-based multiresolution analysis

A multiresolution technique based on multiwavelets scale-space representation for stereo correspondence estimation is presented. The technique uses the well-known coarse-to-fine strategy, involving the calculation of stereo correspondences at the coarsest resolution level with consequent refinement up to the finest level. Vector coefficients of the multiwavelets transform modulus are used as corresponding features, where modulus maxima defines the shift invariant high-level features (multiscale edges) with phase pointing to the normal of the feature surface. The technique addresses the estimation of optimal corresponding points and the corresponding 2D disparity maps. Illuminative variation that can exist between the perspective views of the same scene is controlled using scale normalization at each decomposition level by dividing the details space coefficients with approximation space. The problems of ambiguity, explicitly, and occlusion, implicitly, are addressed by using a geometric topological refinement procedure. Geometric refinement is based on a symbolic tagging procedure introduced to keep only the most consistent matches in consideration. Symbolic tagging is performed based on probability of occurrence and multiple thresholds. The whole procedure is constrained by the uniqueness and continuity of the corresponding stereo features. The comparative performance of the proposed algorithm with eight famous existing algorithms, presented in the literature, is shown to validate the claims of promising performance of the proposed algorithm.

Advanced graph mining methods for protein analysis

As one of the primary substances in a living organism, protein defines the character of each cell by interacting with the cellular environment to promote the cell’s growth and function [1]. Previous studies on proteomics indicate that the functions of different proteins could be assigned based upon protein structures [2,3]. The knowledge on protein structures gives us an overview of protein fold space and is helpful for the understanding of the evolutionary principles behind structure. By observing the architectures and topologies of the protein families, biological processes can be investigated more directly with much higher resolution and finer detail. For this reason, the analysis of protein, its structure and the interaction with the other materials is emerging as an important problem in bioinformatics. However, the determination of protein structures is experimentally expensive and time consuming, this makes scientists largely dependent on sequence rather than more general structure to infer the function of the protein at the present time. For this reason, data mining technology is introduced into this area to provide more efficient data processing and knowledge discovery approaches.

Unlike many data mining applications which lack available data, the protein structure determination problem and its interaction study, on the contrary, could utilize a vast amount of biologically relevant information on protein and its interaction, such as the protein data bank (PDB) [4], the structural classification of proteins (SCOP) databases [5], CATH databases [6], UniProt [7], and others. The difficulty of predicting protein structures, specially its 3D structures, and the interactions between proteins as shown in Figure 6.1, lies in the computational complexity of the data. Although a large number of approaches have been developed to determine the protein structures such as ab initio modelling [8], homology modelling [9] and threading [10], more efficient and reliable methods are still greatly needed.

In this chapter, we will introduce a state-of-the-art data mining technique, graph mining, which is good at defining and discovering interesting structural patterns in graphical data sets, and take advantage of its expressive power to study protein structures, including protein structure prediction and comparison, and protein-protein interaction (PPI). The current graph pattern mining methods will be described, and typical algorithms will be presented, together with their applications in the protein structure analysis.

The rest of the chapter is organized as follows: Section 6.2 will give a brief introduction of the fundamental knowledge of protein, the publicly accessible protein data resources and the current research status of protein analysis; in Section 6.3, we will pay attention to one of the state-of-the-art data mining methods, graph mining; then Section 6.4 surveys several existing work for protein structure analysis using advanced graph mining methods in the recent decade; finally, in Section 6.5, a conclusion with potential further work will be summarized.