Interpretability assessment of fuzzy knowledge bases: A cointension based approach

Computing with words (CWW) relies on linguistic representation of knowledge that is processed by operating at the semantical level defined through fuzzy sets. Linguistic representation of knowledge is a major issue when fuzzy rule based models are acquired from data by some form of empirical learning. Indeed, these models are often requested to exhibit interpretability, which is normally evaluated in terms of structural features, such as rule complexity, properties on fuzzy sets and partitions. In this paper we propose a different approach for evaluating interpretability that is based on the notion of cointension. The interpretability of a fuzzy rule-based model is measured in terms of cointension degree between the explicit semantics, defined by the formal parameter settings of the model, and the implicit semantics conveyed to the reader by the linguistic representation of knowledge. Implicit semantics calls for a representation of user’s knowledge which is difficult to externalise. Nevertheless, we identify a set of properties - which we call “logical view” - that is expected to hold in the implicit semantics and is used in our approach to evaluate the cointension between explicit and implicit semantics. In practice, a new fuzzy rule base is obtained by minimising the fuzzy rule base through logical properties. Semantic comparison is made by evaluating the performances of the two rule bases, which are supposed to be similar when the two semantics are almost equivalent. If this is the case, we deduce that the logical view is applicable to the model, which can be tagged as interpretable from the cointension viewpoint. These ideas are then used to define a strategy for assessing interpretability of fuzzy rule-based classifiers (FRBCs). The strategy has been evaluated on a set of pre-existent FRBCs, acquired by different learning processes from a well-known benchmark dataset. Our analysis highlighted that some of them are not cointensive with user’s knowledge, hence their linguistic representation is not appropriate, even though they can be tagged as interpretable from a structural point of view.     

A Green’s function formulation of nonlocal finite-difference schemes for reaction-diffusion equations

Reaction-diffusion equations are commonly used in different science and engineering fields to describe spatial patterns arising from the interaction of chemical or biochemical reactions and diffusive transport mechanisms. The aim of this work is to show that a Green’s function formulation of reaction-diffusion PDEs is a suitable framework to derive FD schemes incorporating both O(h²) accuracy and nonlocal approximations in the whole domain (including boundary nodes). By doing so, the approach departs from a Green’s function formulation of the boundary-value problem to pose an approximation problem based on a domain decomposition. Within each subdomain, the corresponding integral equation is forced to have zero residual at given grid points. Different FD schemes are obtained depending on the numerical scheme used for computing the Green’s integral over each subdomain. Dirichlet and Neumann boundary conditions are considered, showing that the FD scheme based on the Green’s function formulation incorporates, in a natural way, the effects of boundary nodes in the discretization approximation.     

Buffer storage for continuous delivery of video data

The video communication requires considerable large bandwidth – a scarce resource and thus it needs be used efficiently. It can be traded for buffer storage, i.e. for maintaining more buffer storage the data downloading rate (bandwidth) can be reduced. In order to transmitting the videos, they are divided into segments and the bandwidth of the physical channel is divided into logical channels. A logical channel transmits only one segment of each video. When there is switching of the segments, some time delay, called jitter delay, occurs. The jitter delay, being independent, memoryless, and positive, can be modeled as exponentially distributed random variable. In this paper, we propose a stochastic model incorporating jitter delay for estimating the storage space required in the user’s system in order to maintain continuous delivery of the video data to the users.     

High school students’ understandings of geometric transformations in the context of a technological environment

This study investigated the nature of students’ understandings of geometric transformations, which included translations, reflections, rotations, and dilations, in the context of the technological tool, The Geometer’s Sketchpad. The researcher implemented a seven-week instructional unit on geometric transformations within an Honors Geometry class. Students’ conceptions of transformations as functions were analyzed using the APOS theory and were informed by an analysis of students’ interpretations and uses of representations of geometrical objects using the constructs of drawing and figure. The analysis suggests students’ understandings of key concepts including domain, variables and parameters, and relationships and properties of transformations were critical for supporting the development of deeper understandings of transformations as functions.     

Resilience and reliability analysis of P2P network systems

This paper conducts resilience and reliability analysis of P2P network by studying the isolation probability and the durable time of a single user. The network with user’s lifetime having stronger NWUE property is proved to be more resilient. Further, both graphical and nonparametric methods are developed to test the NWUE order between two data sets.     

Learning effects on maintenance outsourcing

The objective of this paper is to study learning effects on maintenance outsourcing. We consider a situation in which a manufacturer offers a short-term outsourcing contract to an external contractor who is responsible for scheduling and performing preventive maintenance and carrying out minimal repairs when the process fails. The manufacturer’s payment to the contractor consists of a fixed amount along with cost subsidization for each maintenance operation performed. We assume learning occurs when the contractor performs preventive maintenance that reduces both time and cost. Two types of learning are considered: natural learning and learning by costly efforts. We demonstrate that a well-designed payment scheme can induce the contractor to adopt the maintenance schedule that maximizes the manufacturer’s profit.     

A few pedagogical designs in linear algebra with Cabri and Maple

Since the 90’s, with the creation of new electronic environments for learning and teaching, several research groups in Mathematics Education have been emerging and developing. This article elaborates few pedagogical designs in Linear Algebra supported by both the geometrical micro-world Cabri and the computer algebra system Maple. Stumbling blocks in the learning of Linear Algebra are examined, more exactly linear transformations, eigenvectors, quadratic forms, conics with changes of bases and finally singular values. Encountering a special group of students very eager to explore the world of linear algebra, we initiated a classification of linear transformations of the Euclidean plane R² via ellipses.     

Ostrowski’s fourth-order iterative method speedily solves cubic equations of state

Pressure-volume-temperature (P-V-T) data are required in simulating chemical plants because the latter usually involve production, separation, transportation, and storage of fluids. In the absence of actual experimental data, the pertinent mathematical model must rely on phase behaviour prediction by the so-called equations of state (EOS). When the plant model is a combination of differential and algebraic equations, simulation generally relies on numerical integration which proceeds in a piecewise fashion unless an approximate solution is needed at a single point. Needless to say, the constituent algebraic equations must be efficiently re-solved before each update of derivatives. Now, Ostrowski’s fourth-order iterative technique is a partial substitution variant of Newton’s popular second-order method. Although simple and powerful, this two-point variant has been utilised very little since its publication over forty years ago. After a brief introduction to cubic equations of state and their solution, this paper solves five of them. The results clearly demonstrate the superiority of Ostrowski’s method over Newton’s, Halley’s, and Chebyshev’s solvers.     

Comparison of three semiparametric methods for estimating dependence parameters in copula models

Three semiparametric methods for estimating dependence parameters in copula models are compared, namely maximum pseudo-likelihood estimation and the two method-of-moment approaches based on the inversion of Spearman’s rho and Kendall’s tau. For each of these three asymptotically normal estimators, an estimator of their asymptotic (co)variance is stated in three different situations, namely the bivariate one-parameter case, the multivariate one-parameter case and the multivariate multiparameter case. An extensive Monte Carlo study is carried out to compare the finite-sample performance of the three estimators under consideration in these three situations. In the one-parameter case, it involves up to six bivariate and four-variate copula families, and up to five levels of dependence. In the multiparameter case, attention is restricted to trivariate and four-variate normal and t copulas. The maximum pseudo-likelihood estimator appears as the best choice in terms of mean square error in all situations except for small and weakly dependent samples. It is followed by the method-of-moment estimator based on Kendall’s tau, which overall appears to be significantly better than its analogue based on Spearman’s rho. The simulation results are complemented by asymptotic relative efficiency calculations. The numerical computation of Spearman’s rho, Kendall’s tau and their derivatives in the case of copula families for which explicit expressions are not available is also investigated.     

A simulation optimization method that considers uncertainty and multiple performance measures

Simulation optimization provides a structured approach to system design and configuration when analytical expressions for input/output relationships are unavailable. This research focuses on the development of a new simulation optimization technique applicable to systems having multiple performance measures. The aim of this research is to incorporate a simulation end user’s preference towards risk and uncertainty into the search process for the best decision alternative. Automation of the optimization procedure is a necessity. Therefore, this paper proposes a simulation optimization method that involves a preference model, specifically adapted for decision making with simulation models.     

An empirical study of ICASE learning curves and probability bounds for software development effort

In this paper, we investigate the existence of learning curves in software development. Under the assumption of independent and identical distribution (iid) of programmer’s experience and identical effort-experience learning curve relationship for different programmers, we illustrate the existence of an exponentially decreasing learning curve relationship between a programmer’s effort and his/her ICASE tool experience, and show that the effort-experience relationship is inelastic when a programmer’s ICASE tool experience is low. We analyze the impact of our assumptions on actual software development effort, and propose a tight probability upper bound and a central-limit theorem based probability estimator for estimating the approximate probability that the software development effort will be less than or equal to a certain number. Examples to illustrate the use of the probability estimator are also provided.     

A general age-replacement model with minimal repair under renewing free-replacement warranty

A general age-replacement model in which incorporates minimal repair, planned and unplanned replacement, is considered in this paper for products under a renewing free-replacement warranty policy. For both warranted and non-warranted products, cost models from the user’s perspective are developed, and the corresponding optimal replacement ages are derived such that the long-run expected cost rate is minimized. The impacts of a product warranty on the optimal replacement model are investigated analytically. Furthermore, we show that the optimal replacement age for a warranted product is closer to the end of the warranty period than for a non-warranted product. Finally, numerical examples are given for illustration.     

Spaces of orders and their Turing degree spectra

We investigate computability theoretic and topological properties of spaces of orders on computable orderable groups. A left order on a group G is a linear order of the domain of G, which is left-invariant under the group operation. Right orders and bi-orders are defined similarly. In particular, we study groups for which the spaces of left orders are homeomorphic to the Cantor set, and their Turing degree spectra contain certain upper cones of degrees. Our approach unifies and extends Sikora’s (2004) [28] investigation of orders on groups in topology and Solomon’s (2002) [31] investigation of these orders in computable algebra. Furthermore, we establish that a computable free group F_n of rank n>1 has a bi-order in every Turing degree.     

Valuing the flexibility of investing in security process innovations

In this paper, we develop a decision model of a firm’s optimal strategy for investment in security process innovations (SPIs) when confronted with a sequence of malicious attacks. The model incorporates real options as a methodology to capture the flexibility embedded in such investment decisions. SPIs, when seamlessly integrated with the organization’s overall business dynamics, induce organizational learning and provide the flexibility of switching to more suitable technologies as the environment of malicious attacks changes. The theoretical contribution of this paper is a mathematical model of the invest-to-learn and switching options generated upon early investment in flexible SPIs. The practical significance of the paper is the application of a binomial lattice model to approximate the continuous-time model, resulting in an easy to use decision aid for managers.     

A dialogue on the use of arithmetic in geometry: Van Ceulen’s and Snellius’s Fundamenta Arithmetica et Geometrica

Snellius’s Fundamenta Arithmetica et Geometrica (1615) is much more than a Latin translation of Ludolph van Ceulen’s Arithmetische en Geometrische Fondamenten. Willebrord Snellius both adapted and commented on the Dutch original in his Fundamenta, and thus his Latin version can be read as a dialogue between representatives of two different approaches to mathematics in the early modern period: Snellius’s humanist approach and Van Ceulen’s practitioner’s approach. This article considers the relationship between the Dutch and Latin versions of the text and, in particular, puts some of their statements on the use of numbers in geometry under the microscope. In addition, Snellius’s use of the Fundamenta as an instrument to further his career is explained.     

Optimal benchmarking for active portfolio managers

Within an agency theoretic framework adapted to the portfolio delegation issue, we show how to construct optimal benchmarks. In accordance with US regulations, the benchmark-adjusted compensation scheme is taken to be symmetric. The investor’s control consists in forcing the manager to adopt the appropriate benchmark so that his first-best optimum is attained. Solving simultaneously the manager’s and the investor’s dynamic optimization programs in a fairly general framework, we characterize the optimal benchmark. We then provide completely explicit solutions when the investor’s and the manager’s utility functions exhibit different CRRA parameters. We find that, even under optimal benchmarking, it is never optimal for the manager, and therefore for the investor, to follow exactly the benchmark, except in a very restrictive case. We finally assess by simulation the practical importance, in particular in terms of the investor’s welfare, of selecting a sub-optimal benchmark.     

Remarks on Caristi’s fixed point theorem and Kirk’s problem

Without assumptions on the continuity and the subadditivity of η, by means of Caristi’s fixed point theorem, we investigated the existence of fixed points for a Caristi type mapping which partially answered Kirk’s problem and improved Caristi’s fixed point theorem, Jachymski’s fixed point theorem and Khamsi’s fixed point theorem since φ is not necessarily assumed to be bounded below on X.     

On iterative methods for the quadratic matrix equation with M-matrix

In this paper, we consider Newton’s method and Bernoulli’s method for a quadratic matrix equation arising from an overdamped vibrating system. By introducing M-matrix to this equation, we provide a sufficient condition for the existence of the primary solution. Moreover, we show that Newton’s method and Bernoulli’s method with an initial zero matrix converge to the primary solvent under the proposed sufficient condition.     

1-transitive cyclic orderings

We give a classification of all the countable 1-transitive cyclic orderings, being those on which the automorphism group acts singly transitively. We also classify all the countable 1-transitive coloured cyclic orderings, where these are countable cyclic orderings in which each point is assigned a member of a set C, thought of as its ‘colour’, and by ‘1-transitivity’ we now mean that the automorphism group acts singly transitively on each set of points coloured by a fixed colour. We conclude by giving constructions of some uncountable cyclic orderings whose automorphism groups enjoy certain special properties.     

On expected utility for financial insurance portfolios with stochastic dependencies

The effect of background risks as human capital, market risks and catastrophic events has been considered in the literature in different contexts. In this note, we consider financial insurance portfolios with insurable risks and one background risk (uninsurable financial asset), such that the random losses and the background risk depend on environmental parameters. We study how dependencies between the risks influence the expected utility of the portfolio’s wealth distribution under risk aversion, when the environmental parameters are random. Stochastic bounds for the expected wealth are given from modeling the dependence between the parameters by different notions. Similar results are given for multivariate portfolios with n groups and multivariate risk aversion, besides an expected utility comparison result for the minimum and the total portfolio’s wealth.