scholarly journals A New Method to Compare the Interpretability of Rule-Based Algorithms

AI ◽  
2021 ◽  
Vol 2 (4) ◽  
pp. 621-635
Author(s):  
Vincent Margot ◽  
George Luta

Interpretability is becoming increasingly important for predictive model analysis. Unfortunately, as remarked by many authors, there is still no consensus regarding this notion. The goal of this paper is to propose the definition of a score that allows for quickly comparing interpretable algorithms. This definition consists of three terms, each one being quantitatively measured with a simple formula: predictivity, stability and simplicity. While predictivity has been extensively studied to measure the accuracy of predictive algorithms, stability is based on the Dice-Sorensen index for comparing two rule sets generated by an algorithm using two independent samples. The simplicity is based on the sum of the lengths of the rules derived from the predictive model. The proposed score is a weighted sum of the three terms mentioned above. We use this score to compare the interpretability of a set of rule-based algorithms and tree-based algorithms for the regression case and for the classification case.

Author(s):  
Carlos Pinheiro ◽  
Fernando Gomide ◽  
Otávio Carpinteiro ◽  
Isaías Lima

This chapter suggests a new method to develop rule-based models using concepts about rough sets. The rules encapsulate relations among variables and give a mechanism to link granular descriptions of the models with their computational procedures. An estimation procedure is suggested to compute values from granular representations encoded by rule sets. The method is useful to develop granular models of static and dynamic nonlinear systems and processes. Numerical examples illustrate the main features and the usefulness of the method.


Author(s):  
Maria J. Perez-Villadóniga ◽  
Ana Rodriguez-Alvarez ◽  
David Roibas

AbstractResident physicians play a double role in hospital activity. They participate in medical practices and thus, on the one hand, they should be considered as an input. Also, they are medical staff in training and, on the other hand, must be considered as an output. The net effect on hospital activities should therefore be empirically determined. Additionally, when considering their role as active physicians, a natural hypothesis is that resident physicians are not more productive than senior ones. This is a property that standard logarithmic production functions (including Cobb–Douglas and Translog functional forms) cannot verify for the whole technology set. Our main contribution is the development of a Translog modification, which implies the definition of the input “doctors” as a weighted sum of senior and resident physicians, where the weights are estimated from the empirical application. This modification of the standard Translog is able, under suitable parameter restrictions, to verify our main hypothesis across the whole technology set while determining if the net effect of resident physicians in hospitals’ production should be associated to an output or to an input. We estimate the resulting output distance function frontier with a sample of Spanish hospitals. Our findings show that the overall contribution of resident physicians to hospitals’ production allows considering them as an input in most cases. In particular, their average productivity is around 37% of that corresponding to senior physicians.


2021 ◽  
Vol 2 (3) ◽  
pp. 431-441
Author(s):  
Odysseas Kosmas

In previous works we developed a methodology of deriving variational integrators to provide numerical solutions of systems having oscillatory behavior. These schemes use exponential functions to approximate the intermediate configurations and velocities, which are then placed into the discrete Lagrangian function characterizing the physical system. We afterwards proved that, higher order schemes can be obtained through the corresponding discrete Euler–Lagrange equations and the definition of a weighted sum of “continuous intermediate Lagrangians” each of them evaluated at an intermediate time node. In the present article, we extend these methods so as to include Lagrangians of split potential systems, namely, to address cases when the potential function can be decomposed into several components. Rather than using many intermediate points for the complete Lagrangian, in this work we introduce different numbers of intermediate points, resulting within the context of various reliable quadrature rules, for the various potentials. Finally, we assess the accuracy, convergence and computational time of the proposed technique by testing and comparing them with well known standards.


2011 ◽  
Vol 4 (1) ◽  
pp. 76 ◽  
Author(s):  
Na Kong

Based on the current contradiction between the grammar-translation method and the communicative teaching method in English teaching, this paper, starting with clarifying the task of comprehensive English as well as the definition of the two teaching methods, objectively analyzes their advantages and disadvantages and proposes establishing a new method by fusing them with an elaboration on the reasonability of combining them in the practical teaching of comprehensive English with their complementary advantages.


2010 ◽  
Vol 118-120 ◽  
pp. 601-605
Author(s):  
Han Ming

Evaluation method of reliability parameter estimation needs to be improved effectively with the advance of science and technology. This paper develops a new method of parameter estimation, which is named E-Bayesian estimation method. In the case one hyper-parameter, the definition of E-Bayesian estimation of the failure probability is provided, moreover, the formulas of E-Bayesian estimation and hierarchical Bayesian estimation, and the property of E-Bayesian estimation of the failure probability are also provided. Finally, calculation on practical problems shows that the provided method is feasible and easy to perform.


2014 ◽  
Vol 532 ◽  
pp. 113-117
Author(s):  
Zhou Jin ◽  
Ru Jing Wang ◽  
Jie Zhang

The rotating machineries in a factory usually have the characteristics of complex structure and highly automated logic, which generated a large amounts of monitoring data. It is an infeasible task for uses to deal with the massive data and locate fault timely. In this paper, we explore the causality between symptom and fault in the context of fault diagnosis in rotating machinery. We introduce data mining into fault diagnosis and provide a formal definition of causal diagnosis rule based on statistic test. A general framework for diagnosis rule discovery based on causality is provided and a simple implementation is explored with the purpose of providing some enlightenment to the application of causality discovery in fault diagnosis of rotating machinery.


1973 ◽  
Vol 63 ◽  
pp. 50-67 ◽  
Author(s):  
Fergus Millar

More than thirty years after its publication The Roman Revolution still stands unrivalled, not as the ‘definitive’ account of the emergence of a monarch from the ruins of the Republic but as something far more than that, the demonstration of a new method in the presentation of historical change. The aspect of this method, which has found most imitation, is of course prosopography; and it is indeed essential to it. But far more important is the use made of contemporary literature to mirror events, and to analyse and define the concepts and the terms in which the events were seen by those who lived through them.It is the common characteristic, perhaps even the definition, of great works of history that they invite imitation and offer a challenge, not just to apply their methods and standards to other areas, but to pursue their own conclusions further. The present paper is gratefully offered as an attempt to portray with a different emphasis some aspects of the establishment of Octavian as a monarch, first by demonstrating the extent to which the institutions of the res publica remained active in the Triumviral period, and secondly by redefining the change which culminated in 27 B.C., precisely by asking again in what terms it and the novus status which emerged from it were seen by contemporaries.


2011 ◽  
Vol 26 (1) ◽  
pp. 117-128 ◽  
Author(s):  
Ilya B. Gertsbakh ◽  
Yoseph Shpungin

We consider binary coherent systems with independent binary components having equal failure probability q. The system DOWN probability is expressed via its signature's combinatorial analogue, the so-called D-spectrum. Using the definition of the Birnbaum importance measure (BIM), we introduce for each component a new combinatorial parameter, so-called BIM-spectrum, and develop a simple formula expressing component BIM via the component BIM-spectrum. Further extension of this approach allows obtaining a combinatorial representation for the joint reliability importance (JRI) of two components. To estimate component BIMs and JRIs, there is no need to know the analytic formula for system reliability. We demonstrate how our method works using the Monte Carlo approach. We present several examples of estimating component importance measures in a network when the DOWN state is defined as the loss of terminal connectivity.


Sign in / Sign up

Export Citation Format

Share Document