Generate Decision Rules Research Articles

Performance Bounds for Look-Ahead Power System Dispatch Using Generalized Multistage Policies

We present a combined look-ahead dispatch and reserve optimization formulation, which extends our recent work on time-coupled reserve policies and employs the recent notion of generalized decision rules from the robust optimization literature. This aims to improve the performance of traditional linear decision rules when applied to short-term electrical reserve operation. We derive a primal problem whose solution is a time-coupled rule for up- and down-regulating the power injections of each controllable device, such as a generator or energy storage unit, in response to discovered values of prediction errors. This rule depends on the “bin” into which measured prediction errors fall, so that, for example, up-regulation follows a different rule to down-regulation. We also derive an associated dual problem, whose solution provides a lower-bound on the best possible primal cost. This allows the suboptimality of a candidate solution based on a particular decision rule parameterization to be bounded. The primal and dual solutions are also compared to the so-called prescient case, in which the values of the uncertainty are known in advance. We demonstrate the method using numerical case studies, including the standard IEEE-118 bus network, in which a minimum possible reserve cost is identified using the dual lower-bounding approach.

Spur bevel gearbox fault diagnosis using wavelet packet transform and rough set theory

The gearbox is an important component in industrial drives, providing safe and reliable operation for industrial production. Wavelet packet transform (WPT) analysis was used to extract fault features in the vibration signals generated by a gearbox. The extracted features from the WPT were used as input in a rough set (RS) for attribute reduction and then combined with a genetic algorithm to obtain global optimal attribute reduction results. The fault features gained after the attribute reductions were used to generate decision rules. The unknown gear status signal attributes were used as input to match the generated decision rules for fault diagnosis purposes. Gearbox vibration signals contain a significant amount of gear status information; a WPT has an acute portion-locked ability to extract attribute information from the vibration signals. However, WPT frequency aliasing would lead to the generation of spurious frequency components, affecting gear fault diagnosis. In this paper, we introduce an improved WPT to eliminate frequency aliasing, thus improving the accuracy of fault diagnosis. This paper studies the use of wavelet packet for feature extraction and the RS for classification; the results demonstrate that this method can accurately and reliably detect failure modes in a gearbox.

Data Classification based on Decision Tree, Rule Generation, Bayes and Statistical Methods: An Empirical Comparison

this paper, twenty well known data mining classification methods are applied on ten UCI machine learning medical datasets and the performance of various classification methods are empirically compared while varying the number of categorical and numeric attributes, the types of attributes and the number of instances in datasets. In the performance study, Classification Accuracy (CA), Root Mean Square Error (RMSE) and Area Under Curve (AUC) of Receivers Operational Characteristics (ROC) is used as the metric and come up with some findings: (i) performance of classification methods depends upon the type of dataset variables or attributes such as categorical, numeric and both (mixed), (ii) performance of classification methods on categorical attributes is superior than on numeric attributes of a dataset, (iii) classification accuracy, RMSE and AUC of a classification method depends on the number of instances in datasets, (iv) classification performance decreases in case of instances decreases for both categorical as well as numeric datasets, (v) top three classification methods are established after comparing the performance of twenty different classification methods for the categorical, numeric and both (mixed) attribute datasets, (vi) out of these twenty different classification methods Bayes Net, Naive Bayes, Classification Via Regression, Logistic Regression and Random Forest method performs best on these medical datasets.

Hydroeconomic optimization of reservoir management under downstream water quality constraints

Summary A hydroeconomic optimization approach is used to guide water management in a Chinese river basin with the objectives of meeting water quantity and water quality constraints, in line with the China 2011 No. 1 Policy Document and 2015 Ten-point Water Plan. The proposed modeling framework couples water quantity and water quality management and minimizes the total costs over a planning period assuming stochastic future runoff. The outcome includes cost-optimal reservoir releases, groundwater pumping, water allocation, wastewater treatments and water curtailments. The optimization model uses a variant of stochastic dynamic programming known as the water value method. Nonlinearity arising from the water quality constraints is handled with an effective hybrid method combining genetic algorithms and linear programming. Untreated pollutant loads are represented by biochemical oxygen demand (BOD), and the resulting minimum dissolved oxygen (DO) concentration is computed with the Streeter–Phelps equation and constrained to match Chinese water quality targets. The baseline water scarcity and operational costs are estimated to 15.6 billion CNY/year. Compliance to water quality grade III causes a relatively low increase to 16.4 billion CNY/year. Dilution plays an important role and increases the share of surface water allocations to users situated furthest downstream in the system. The modeling framework generates decision rules that result in the economically efficient strategy for complying with both water quantity and water quality constraints.

Toward Inclusive Trial Protocols in Heterogeneous Neurological Disorders

Background. Several novel drug- and cell-based potential therapies for spinal cord injury (SCI) have either been applied or will be considered for future clinical trials. Limitations on the number of eligible patients require trials be undertaken in a highly efficient and effective manner. However, this is particularly challenging when people living with incomplete SCI (iSCI) represent a very heterogeneous population in terms of recovery patterns and can improve spontaneously over the first year after injury. Objective. The current study addresses 2 requirements for designing SCI trials: first, enrollment of as many eligible participants as possible; second, refined stratification of participants into homogeneous cohorts from a heterogeneous iSCI population. Methods. This is a retrospective, longitudinal analysis of prospectively collected SCI data from the European Multicenter study about Spinal Cord Injury (EMSCI). We applied conditional inference trees to provide a prediction-based stratification algorithm that could be used to generate decision rules for the appropriate inclusion of iSCI participants to a trial. Results. Based on baseline clinical assessments and a defined subsequent clinical endpoint, conditional inference trees partitioned iSCI participants into more homogeneous groups with regard to the illustrative endpoint, upper extremity motor score. Assuming a continuous endpoint, the conditional inference tree was validated both internally as well as externally, providing stable and generalizable results. Conclusion. The application of conditional inference trees is feasible for iSCI participants and provides easily implementable, prediction-based decision rules for inclusion and stratification. This algorithm could be utilized to model various trial endpoints and outcome thresholds.

Survey on Background Modeling and Foreground Detection for Real Time Video Surveillance

In Image Processing, a very trivial task is to detect the changes in the multiple images of the same scene of a real time instant. The task is not only trivial but also very indispensable as it brings into play a great number of diversified subject area applications such as, remote sensing, surveillance, medical diagnosis and treatment, security surveillance, and underwater sensing. The main perspective of this survey is to give a nomenclature study of the general processing steps and prime decision rules used in the advanced change detection algorithms, which are employed for the real time video surveillance. The real time video surveillance models encompass the predictive and the background modelling techniques. The survey also emphasizes on the comparison of the processing speed of the various change detection algorithms applied in real time video surveillance.

A Rough Set Based Classification Model for the Generation of Decision Rules

This paper introduces a very important classification aspect for the analysis of huge amount of data stored in databases and other repositories. Numerous classification models are available in the literature, to predict the class of objects whose class level is unknown. Literature reveals that most of the available models are not capable in handling imperfect data. In view of this, present paper proposes a new rough set based classification model to derive the classification (IF-THEN) rules. Furthermore, developed model has been applied to handle bank-loan applications database as either safe, unsafe or risky. However, proposed model can also be used for the analysis of data from other domains.

Integrating Kansei engineering with conjoint analysis to fulfil market segmentation and product customisation for digital cameras

Diverse customer desires coupled with technological advances have forced companies to manufacture products with ultimate performance, low cost, high quality and much shorter time-to-market. Recently, the popularity of smart phones has given rise to seriously declined product sales of digital cameras. In this paper, a two-phase framework is presented to offer decision supports on developing next-generation cameras. In the phase of market segmentation, Kansei engineering is employed to capture customer perceptions of affective features. Then, rough set theory is conducted to generate decision rules for partitioning the whole market into the consumer segment and the professional segment, respectively. In the phase of product customisation, conjoint analysis is applied to extract customer preferences for functional features. Furthermore, Grey relational analysis is conducted to select the top three varieties with regard to two distinct segments. In particular, this paper is capable to help brand companies or camera manufacturers better capture customer perceptions and preferences for digital cameras, effectively perform market segmentation (based on affective features) and efficiently conduct product customisation (based on functional features).

Classification Models for RFID-Based Real-Time Detection of Process Events in the Supply Chain

RFID technology allows the collecting of fine-grained real-time information on physical processes in the supply chain that often cannot be monitored using conventional approaches. However, because of the phenomenon of false-positive reads, RFID data streams resemble noisy analog measurements rather than the desired recordings of activities within a business process. The present study investigates the use of data mining techniques for filtering and aggregating raw RFID data. We consider classifiers based on logistic regression, decision trees, and artificial neural networks using attributes derived from low-level reader data. In addition, we present a custom-made algorithm for generating decision rules using artificial attributes and an iterative training procedure. We evaluate the classifiers using a massive set of data on pallet movements collected under real-world conditions at one of the largest retailers worldwide. The results clearly indicate high classification performance of the classification models, with the rule-based classifier outperforming all others. Moreover, we show that utilizing the full spectrum of data generated by the reader hardware leads to superior performance compared with the approaches based on timestamp and antenna information proposed in prior research.

The Effects and Value of a Resistant Perennial Variety: An Application to Pudrición del Cogollo Disease

AbstractThis article develops a forestry model to obtain the optimal control strategy and optimal rotation length after a disease attacks in a perennial variety. Three cases are considered: a benchmark consisting of a disease‐free field, an identical field with the disease present but no resistant variety with which to replant, and an identical field with the disease present and a resistant variety with which to replant. We determine general decision rules and then apply the model to the case of Pudrición del Cogollo, a major disease threat to the Colombian oil palm industry. In the application, we compare the optimal rotation length between the three scenarios and determine the optimal level of control in each period for the disease scenarios. The singular solution involves complete control of the disease, and in the absence of a resistant variety, the presence of the disease increases the rotation length. With these solutions, we then determine the value of developing a resistant variety. This value depends heavily on the age distribution of the current trees and decreases as the average tree age decreases. The value further declines when the resistance variety has negative attributes such as higher replanting and maintenance costs than the original variety.

Readmission to an intensive care unit after cardiac surgery: reasons and outcomes

Intensive care unit (ICU) readmission after cardiac surgery is believed to be associated with higher in-hospital mortality and may predict poor outcomes. ICU readmissions use resources and increase treatment costs. To determine reasons for readmission to ICU, evaluate outcomes in these patients, and identify factors predisposing to the need for readmission to ICU. We retrospectively investigated a total of 2076 consecutive adult patients who underwent either isolated coronary artery bypass grafting or a valve procedure or combination of both and were discharged from our ICU between January 2008 and December 2010. To identify the factors that increase the risk of readmission to ICU, we used the dominance-based rough set approach (DRSA) which is a methodology of knowledge discovery from data. The knowledge has the form of "if... then..." decision rules relating patient characteristics to the risk of readmission to ICU. Of 2076 patients discharged from ICU, 56 (2.7%) required a second stay in the ICU (study group) while 2020 patients needed no readmission to ICU (control group). The main causes of readmission were haemodynamic instability (28.6%, n = 16), respiratory failure (23.2%, n = 13), and cardiac tamponade or bleeding (23.2%, n = 13). The mean length of stay (LOS) in the general cardiac ward after primary discharge from ICU until readmission was 3.5 ± 4.2 days. The mean LOS in ICU after readmission was 12.5 ± 21.2 days. Postoperative complications occurred more frequently in readmitted patients (10.2% vs. 48.2%, p < 0.0001). In-hospital mortality was significantly higher in the study group (15 [26.8%] vs. 23 [1.1%] patients, p < 0.0001). As a result of applying the DRSA methodology, the algorithm generated decision rules categorizing patients into high and low ICU readmission risk. Advanced age, non-elective surgery and the length of initial ICU stay after the surgery were the factors of greatest importance for the correct categorisation of patients in the study group. The most common cause of readmission to ICU is haemodynamic instability. Postoperative complication and in-hospital mortality rates are significantly higher in patients readmitted to ICU. Factors most commonly predisposing to readmission to ICU after cardiac surgery included advanced patient age, non-elective surgery, and longer initial stay in ICU after the surgery.

Knowledge Discovery in Medical Data by using Rough Set Rule Induction Algorithms

In the performance of data mining and knowledge discovery activities, rough set theory has been regarded as a powerful, feasible and effective methodology. There is a need for analysis of medical data that deals with incomplete and inconsistent information with the tremendous manipulation at different levels. In this context, rough set rule induction algorithms are capable of generating decision rules which can potentially provide new medical insight and profound medical knowledge. By taking into consideration all the above aspects, the present investigation is carried out. The results clearly show that rough set approach is certainly a useful tool for medical applications. Relationships and patterns within this data could provide new medical knowledge. The genetic algorithms offer an attractive approach for solving the feature subset selection problem. The process of finding useful patterns or meaning in raw data has been called knowledge discovery in databases. The algorithms used for the present study are: Exhaustive search, Covering, LEM2 and Genetic algorithms. Rules are generated and improved in the case of the above mentioned four algorithms. Further, the generated rules are improved by applying the shortening ratio as 0.8. Some of the important results of the present investigation include maximum coverage of 100% is observed in the case of exhaustive and genetic algorithms.

Knowledge Discovery in Medical Data by using Rough Set Rule Induction Algorithms

In the performance of data mining and knowledge discovery activities, rough set theory has been regarded as a powerful, feasible and effective methodology. There is a need for analysis of medical data that deals with incomplete and inconsistent information with the tremendous manipulation at different levels. In this context, rough set rule induction algorithms are capable of generating decision rules which can potentially provide new medical insight and profound medical knowledge. By taking into consideration all the above aspects, the present investigation is carried out. The results clearly show that rough set approach is certainly a useful tool for medical applications. Relationships and patterns within this data could provide new medical knowledge. The genetic algorithms offer an attractive approach for solving the feature subset selection problem. The process of finding useful patterns or meaning in raw data has been called knowledge discovery in databases. The algorithms used for the present study are: Exhaustive search, Covering, LEM2 and Genetic algorithms. Rules are generated and improved in the case of the above mentioned four algorithms. Further, the generated rules are improved by applying the shortening ratio as 0.8. Some of the important results of the present investigation include maximum coverage of 100% is observed in the case of exhaustive and genetic algorithms. Keywords: Decomposition Tree, Rough Set, Rules, Rule Induction Algorithms

Extracting Decision Rules from Linguistic Data Describing Economic Phenomena. The Approach Based on Decision Systems over Ontological Graphs and PSO

The aim of the paper is to present a heuristic method for extracting the most general decision rules from linguistic data describing economic phenomena included in simple decision systems over ontological graphs. Such decision systems have been proposed to deal with linguistic attribute values, describing objects of interest, which are concepts placed in semantic spaces expressed by means of ontological graphs. Ontological graphs deliver some additional knowledge (the so-called background knowledge) about semantic relations between concepts which can be useful in classification processes. As heuristics, we propose to use Particle Swarm Optimization (PSO ), which is reported as a successful method in many applications.

Generalized decision rule approximations for stochastic programming via liftings

Stochastic programming provides a versatile framework for decision-making under uncertainty, but the resulting optimization problems can be computationally demanding. It has recently been shown that primal and dual linear decision rule approximations can yield tractable upper and lower bounds on the optimal value of a stochastic program. Unfortunately, linear decision rules often provide crude approximations that result in loose bounds. To address this problem, we propose a lifting technique that maps a given stochastic program to an equivalent problem on a higher-dimensional probability space. We prove that solving the lifted problem in primal and dual linear decision rules provides tighter bounds than those obtained from applying linear decision rules to the original problem. We also show that there is a one-to-one correspondence between linear decision rules in the lifted problem and families of nonlinear decision rules in the original problem. Finally, we identify structured liftings that give rise to highly flexible piecewise linear and nonlinear decision rules, and we assess their performance in the context of a dynamic production planning problem.

The Application of Dominance-based Rough Sets Theory to Evaluation of Transportation Systems

The paper presents an original procedure of evaluation of a transportation system, resulting in its assignment into a predefined class, representing the overall standard of the considered system and the level of transportation service. The method relies on the application of the dominance-based rough set theory (DRST), allows for thorough data exploration, evaluation of informational content of the considered characteristics and generation of certain decision rules that support t he evaluation process. In the analysis different characteristics (criteria and attributes) describing various aspects of a transportation system operations are taken into account. The assignment of a transportation system to a specific quality class is performed based on the values of characteristics which are compared with the evaluation pattern, i.e. the set of decision rules generated through the analysis of customers’ opinions and expectations concerning a transportation system. The method is composed of three major steps, including: 1) identification of the most important characteristics, 2) generation of the evaluation pattern, and 3) assignment of the transportation system to the appropriate class. In the evaluation process five key components of a transportation system, including: transportation means, human resources, informational resources, transportation infrastructure and technical equipment as well as organizational rules are considered.

Glacier mapping based on rough set theory in the Manas River watershed

Precise glacier information is important for assessing climate change in remote mountain areas. To obtain more accurate glacier mapping, rough set theory, which can deal with vague and uncertainty information, was introduced to obtain optimal knowledge rules for glacier mapping. Optical images, thermal infrared band data, texture information and morphometric parameters were combined to build a decision table used in our proposed rough set theory method. After discretizing the real value attributes, decision rules were calculated through the decision rule generation algorithm for glacier mapping. A decision classifier based on the generated rules classified the multispectral image into glacier and non-glacier areas. The result of maximum likelihood classification (MLC) was used to compare with the result of the classification based on the rough set theory. Confusion matrix and visual interpretation were used to evaluate the overall accuracy of the results of the two methods. The accuracies of the rough set method and maximum likelihood classification were compared, yielding overall accuracies of 94.15% and 93.88%, respectively. It showed the area difference based on rough set was smaller by comparing the glacier areas of the rough set method and MLC with visual interpreter, respectively. The high accuracy for glacier mapping and the small area difference for glacier based on rough set theory demonstrated that this method was effective and promising for glacier mapping.

A new architectural framework for rule-based healthcare system using semantic web technologies

One of the main focuses on healthcare systems is to increase treatment quality and the development of a fully automated healthcare system, without intervention from domain experts and IT people. In this paper, we propose a framework that performs rule-based healthcare system (RBHS). In the first step, with the use of classification algorithms the decision rules are generated. Then the rule engine executes the generated decision rules and provides the proper diagnostic message. In the next step, the system reason over the semantic web rule language (SWRL)-based rules and gives the treatment adaptation result based on the diagnosis. During execution the treatment adaptation scheme is personalised. We use a rule-based system for diabetes as an example to illustrate how the RBHS system works. In addition to that we construct diet ontology for diabetic and non-diabetic users.

A Novel Two-Stage Spectrum-Based Approach for Dimensionality Reduction: A Case Study on the Recognition of Handwritten Numerals

Dimensionality reduction (feature selection) is an important step in pattern recognition systems. Although there are different conventional approaches for feature selection, such as Principal Component Analysis, Random Projection, and Linear Discriminant Analysis, selecting optimal, effective, and robust features is usually a difficult task. In this paper, a new two-stage approach for dimensionality reduction is proposed. This method is based on one-dimensional and two-dimensional spectrum diagrams of standard deviation and minimum to maximum distributions for initial feature vector elements. The proposed algorithm is validated in an OCR application, by using two big standard benchmark handwritten OCR datasets, MNIST and Hoda. In the beginning, a 133-element feature vector was selected from the most used features, proposed in the literature. Finally, the size of initial feature vector was reduced from 100% to 59.40% (79 elements) for the MNIST dataset, and to 43.61% (58 elements) for the Hoda dataset, in order. Meanwhile, the accuracies of OCR systems are enhanced 2.95% for the MNIST dataset, and 4.71% for the Hoda dataset. The achieved results show an improvement in the precision of the system in comparison to the rival approaches, Principal Component Analysis and Random Projection. The proposed technique can also be useful for generating decision rules in a pattern recognition system using rule-based classifiers.

Improving Situational Awareness for Precursory Data Classification using Attribute Rough Set Reduction Approach

The task of modeling the distribution of a large number of earthquake events with frequent tremors detected prior to a main shock presents us unique challenges to model a robust classifier tool for rapid responses are needed in order to address victims. We have designed using a relational database for running a geophysical modeling application after connecting database record of all clusters of foreshock events from (1998-2010) for a complete catalog of seismicity analysis for the Himalayan basin. by Nath et al,2010. This paper develops a reduced rough set analysis method and implements this novel structure and reasoning process for foreshock cluster forecasting. In this study, we developed a reusable information technology infrastructure, called Efficient Machine Readable for Emergency Text Selection(EMRETS). The association and importance of precursory information in reference to earthquake rupture analysis is found out through attribute reduction based on rough set analysis. Secondly, find the importance of attributes through information entropy is a novel approach for high dimensional complex polynomial problems pre- dominant in geo-physical research and prospecting. Thirdly, we discuss the reducible indiscernible matrix and decision rule generation for a particular set of geographical co-ordinates leading to the spatial discovery of future earthquake having prior foreshock. This paper proposes a framework for extracting, classifying, analyzing, and presenting semi-structured catalog data sources through feature representation and selection.