Dominance-based Rough Set Research Articles

An innovative decision rule approach to tyre pressure monitoring

Tyres are one of the most important safety components on a vehicle. Ignoring or failing to correctly set the tyre pressure may lead to accidents, and can affect the vehicle’s fuel efficiency and tyre lifespan. Hence, there is a need for a Tyre Pressure Monitoring System (TPMS) that can effectively monitor the tyre condition. The current threshold-based TPMSs are characterised by a high number of false alarms. This is mainly due to: (i) the non-static and dynamic relationship between tyre pressure and temperature; and, (ii) the measurement error of the pressure/temperature sensors that are used for data collection. In this paper, we propose an innovative decision rule-based approach to tyre monitoring. This approach relies on the Dominance-based Rough Set Approach (DRSA), which is a well-known multicriteria classification and preference learning method. The DRSA takes a decision table as an input and it generates a collection of if-then decision rules as an output. The issue caused by the dynamic pressure/temperature relationship is solved by fixing one of the parameters and then generating the decision rules based on the other parameter. The problem of false alarms is solved by a discretisation of the scale of the fixed parameter. Based on these solutions, we designed two types of analysis levels: pressure-oriented analysis and temperature-oriented analysis. The proposed approach has been validated and implemented within an important travelling company that operates in the South of England. The real-world tests showed that the proposed approach has improved the current system and has led to a substantial reduction of false alarms.

Defining Poverty Using Dominance-Based Rough Set Theory and Proposing Strategic Objectives for the United Nations Developing Countries

This article is the last of a series of three researches. The purpose of this research is to expose the results of using Dominance-based Rough Set Approach (DRSA) to help International organizations (both non-governmental organizations and governmental organizations) define poverty, identifying economical, sociological, political and technological strategic objectives for developing countries. More precisely, politicians, decision makers and international organizations will be able to study 23 various political, economical, sociological and technological indicators and classify all the countries according to the following three different categories: [A] Countries that are doing well according to the selected indicators; [B] Countries that need support to acquire category A status; [C] Countries ranked the lowest and meeting special support with regard to the criterion or criteria considered. The three categories are delimited by tertiles relative to the average ranking of the member states of the United Nations. The chosen criteria are measured in order to provide decision rules based on this classification. These decision rules thus focus on the strategic needs of countries with respect to improving their development and classification. We strongly believe that by targeting these identified needs, this research will help the sustainable development of countries in need to set realistic targets, prioritize International funding, evaluate economical growth and sociological improvements. Among the results of this article, priorities for countries ranked the lowest should focus on reducing adolescent fertility and increasing school life expectancy.

Dynamic Knowledge Update Using Three-Way Decisions in Dominance-Based Rough Sets Approach While the Object Set Varies

Dominance-based rough set approach is the extension of classical Pawlak rough set theories and methodologies, in which the information with preference-ordered relation on the domain of attribute value is fully considered. In the dominance-based information system, upper and lower approximations will be changed while the object set varies over time and the approximations need to be updated correspondingly for their variations result in changes of knowledge and rules. Considering that three-way decisions is a special class of general and effective human ways of problem solving and information processing, a new incremental maintenance mechanism using three-way decisions is proposed in this paper, namely, the universe is divided into three pair-wise disjoint subsets firstly, then appropriate strategies are developed and acted on each subsets. Furthermore, the corresponding methods for updating the approximations of upward unions and downward unions of decision classes are analyzed systematically under the variations of object set in the dominance-based information system from the perspective of three-way decisions. Moreover, considering vector representation and calculation is intuitive and concise, two incremental update algorithms of approximations are suggested and implemented in the MATLAB platform. Finally, some tests on data sets from UCI (UC Irvine Machine Learning Repository) are undertaken to verify the effectiveness of the proposed methods. Compared with the non-incremental updating methods, the proposed incremental updating method with three-way decisions generally exhibits a better performance.

Identifying strategic development objectives for European Union states using the Dominance-based Rough Set Approach: the case of Poland

Identifying strategic development objectives for European Union states using the Dominance-based Rough Set Approach: the case of Poland

Dominance-based rough sets in multi-scale intuitionistic fuzzy decision tables

To represent the structure of data measured at different granularity levels hierarchically, the concept of multi-scale information table or decision table has been presented from the perspective of granular computing. In this study, we mainly address the issue of optimal scale selection and rule acquisition in dominance-based multi-scale intuitionistic fuzzy (IF) decision tables. By considering a type of decision systems where condition attributes are taken as dominant IF values and those of decision attributes are symbolic dominance ones, we propose two dominance-based rough sets in multi-scale IF decision tables and discuss their optimal scale selection and rule acquisition algorithms, which are used to acquire decision rules for information system security auditing risk judgment for certified information systems auditors, candidate global supplier selection, and car classification. Unlike in the existing literature, the selection of the optimal scales and reduction are simultaneously performed in the viewpoint of local and global approximations in dominance-based multi-scale IF decision tables.

Distinguishing Vagueness from Ambiguity in Rough Set Approximations

In this paper we present a new approach to rough set approximations that permits to distinguish between two kinds of “imperfect” knowledge in a joint framework: on one hand, vagueness, due to imprecise knowledge and uncertainty typical of fuzzy sets, and on the other hand, ambiguity, due to granularity of knowledge originating from the coarseness typical of rough sets. The basic idea of our approach is that each concept is represented by an orthopair, that is, a pair of disjoint sets in the universe of knowledge. The first set in the pair contains all the objects that are considered as surely belonging to the concept, while the second set contains all the objects that surely do not belong to the concept. In this context, following some previous research conducted by us on the algebra of rough sets, we propose to define as rough approximation of the orthopair representing the considered concept another orthopair composed of lower approximations of the two sets in the first orthopair. We shall apply this idea to the classical rough set approach based on indiscernibility, as well as to the dominance-based rough set approach. We discuss also a variable precision rough approximation, and a fuzzy rough approximation of the orthopairs. Some didactic examples illustrate the proposed methodology.

Machine-learned models using hematological inflammation markers in the prediction of short-term acute coronary syndrome outcomes

BackgroundIncreased systemic and local inflammation play a vital role in the pathophysiology of acute coronary syndrome. This study aimed to assess the usefulness of selected machine learning methods and hematological markers of inflammation in predicting short-term outcomes of acute coronary syndrome (ACS).MethodsWe analyzed the predictive importance of laboratory and clinical features in 6769 hospitalizations of patients with ACS. Two binary classifications were considered: significant coronary lesion (SCL) or lack of SCL, and in-hospital death or survival. SCL was observed in 73% of patients. In-hospital mortality was observed in 1.4% of patients and it was higher in the case of patients with SCL. Ensembles of decision trees and decision rule models were trained to predict these classifications.ResultsThe best performing model for in-hospital mortality was based on the dominance-based rough set approach and the full set of laboratory as well as clinical features. This model achieved 81 ± 2.4% sensitivity and 81.1 ± 0.5% specificity in the detection of in-hospital mortality. The models trained for SCL performed considerably worse. The best performing model for detecting SCL achieved 56.9 ± 0.2% sensitivity and 66.9 ± 0.2% specificity. Dominance rough set approach classifier operating on the full set of clinical and laboratory features identifies presence or absence of diabetes, systolic and diastolic blood pressure and prothrombin time as having the highest confirmation measures (best predictive value) in the detection of in-hospital mortality. When we used the limited set of variables, neutrophil count, age, systolic and diastolic pressure and heart rate (taken at admission) achieved the high feature importance scores (provided by the gradient boosted trees classifier) as well as the positive confirmation measures (provided by the dominance-based rough set approach classifier).ConclusionsMachine learned models can rely on the association between the elevated inflammatory markers and the short-term ACS outcomes to provide accurate predictions. Moreover, such models can help assess the usefulness of laboratory and clinical features in predicting the in-hospital mortality of ACS patients.

Hybrid rough fuzzy soft classifier based multi-class classification model for agriculture crop selection

In this paper, rough, fuzzy and soft set approaches have been integrated to develop a multi-class classification model to assist the farmers in taking decision on crop cultivation for a given agriculture land. The model is divided into three major sections, namely weight calculation of variables, conversion of continuous data to fuzzified values and classification rule generation. Dominance-based rough set approach is used for the calculation of relative weights of variables. Fuzzy proximity relation is applied to convert the continuous data into fuzzified values. Bijective soft set approach is used to generate classification rules for five agriculture crops, namely paddy, groundnut, sugarcane, cumbu and ragi. The developed model has been tested with agriculture dataset which showed 92% accuracy for the validation dataset and proved to be confident and robust for agriculture development. Further, the performance of the proposed model is compared with three popular classifiers such as naive Bayes, support vector machine and J48. The obtained experimental results showed high predictive performance, and the potential of the proposed model is compared with the other classifiers.

Domain-wise approaches for updating approximations with multi-dimensional variation of ordered information systems

Dominance-based Rough Set Approach (DRSA) is widely applied in processing the information and data with a preference-order relation. In many real-life Ordered Information Systems (OIS), the attribute values and objects may vary simultaneously with time. Existing incremental updating algorithms in OIS are all single-dimensional, i.e., they only consider the individual variation of attribute values or objects. In addition, all the objects need to be compared with each other during updating. In this paper, we focus on designing specific multi-dimensional algorithms to efficiently update approximations on the simultaneous variation of attribute values and objects in OIS. The properties of the P-generalized decision in DRSA are firstly presented. Then two novel notions e.g., the P-generalized decision lower and upper domains, are defined, respectively. These two notions indicate the practical dominance dependence between different objects in OIS, and provide a simplified definition of the P-generalized decision. The approach based on these two notions can greatly reduce the comparisons between different objects, as well as provide novel strategies which can efficiently obtain the updated P-generalized decision to obtain the updated approximations. Furthermore, we develop two domain-wise algorithms which correspond to two cases of the multi-dimensional variation of attribute values and objects in OIS, respectively. Moreover, in implementing the algorithms, the sorting strategies are integrated to solve the problem of time-consuming traversals. A series of experimental results illustrate that our proposed domain-wise algorithms are not only more efficient than both of the traditional static algorithm and the integrated single-dimensional algorithms for dealing with the multi-dimensional variation of attribute values and objects, but also more efficient than the single-dimensional algorithms for dealing with the individual variation of objects in OIS.

Exploring Users’ Self-Disclosure Intention on Social Networking Applying Novel Soft Computing Theories

In recent years, users have increasingly focused on the privacy of social networking sites (SNS); users have reduced their self-disclosure intention. To attract users, SNS rely on active platforms that collect accurate user information, even though that information is supposed to be private. SNS marketers must understand the key elements for sustainable operation. This study aims to understand the influence of motivation (extrinsic and intrinsic) and self-disclosure on SNS through soft computing theories. First, based on a survey of 1108 users of SNS, this study used a dominance-based rough set approach to determine decision rules for self-disclosure intention on SNS. In addition, based on 11 social networking industry experts’ perspectives, this study validated the influence between the motivation attributes by using Decision-Making Trial and Evaluation Laboratory (DEMATEL). In this paper, the decision rules of users’ self-disclosure preference are presented, and the influences between motivation attributes are graphically depicted as a flow network graph. These findings can assist in addressing real-world decision problems, and can aid SNS marketers in anticipating, evaluating, and acting in accord with the self-disclosure motivations of SNS users. In this paper, practical and research implications are offered.

Attribute reduction in incomplete ordered information systems with fuzzy decision

Rough set theory has been applied extensively to attribute reduction. Classical rough sets provide a theoretical framework for attribute reduction based on complete data with regular attributes where the domains are not ordered by preference. However, their scope does not include incomplete data with fuzzy decisions under a preference-ordered domain, which are common in real real-world applications. Therefore, in this study, a general framework is proposed for attribute reduction from incomplete ordered information systems with fuzzy decisions by combining dominance-based rough sets with α-cut sets, where α is the fuzzy decision attribute value. First, the judgement theorems and discernibility functions are established by applying Boolean reasoning techniques to attribute reduction in consistent and inconsistent incomplete ordered information systems with fuzzy decision. In addition, forward and backward attribute reduction algorithms are designed for consistent and inconsistent systems, respectively, to find near-optimal attribute reducts. Finally, the experimental results based on different datasets, demonstrate that the proposed algorithms are more effective for attribute reduction in most cases than other reduction algorithms.

Optimization of pellets manufacturing process using rough set theory

Pharmaceutical pellets are spherical agglomerates manufactured in extrusion/spheronization process. The composition of the pellets, the amount of active pharmaceutical ingredient (API) and the type of used excipients have an influence on the shape and quality of dosage form. A proper quality of the pellets can also be achieved by identifying the most important technological process parameters. In this paper, a knowledge discovery method, called dominance-based rough set approach (DRSA) has been applied to evaluate critical process parameters in pellets manufacturing. For this purpose, a set of condition attributes (amount of API; type and amount of excipient used; process parameters such as screw and rotation speed, time and temperature of spheronization) and a decision attribute (quality of the pellets defined by the aspect ratio) were used to set up an information system. The DRSA analysis allowed to induce decision rules containing information about process parameters which have a significant impact on the quality of manufactured pellets. Those rules can be used to optimize the process of pellets manufacturing.

An ordinal classification of brownfield remediation projects in China for the allocation of government funding

Given the increasing number of brownfields in China, the Chinese government is now facing a huge financing gap in brownfield remediation projects. Accordingly, the authors present a framework that can assist a provincial government in classifying its brownfields so that the funding can be appropriately allocated. More specifically, the framework starts from constructing a criteria system based on the existing literature and expert opinions. Then, in consonance with the criteria system, a sample set of brownfields is measured by experts and then trained using a Dominance-Based Rough Set Approach to derive judging rules. After the validation, those rules are employed to generate an ordinal classification of the complete set of brownfields. The resulting ordinal classification provides the government with a sequence of brownfields from the highest priority to the lowest, which can support the government in budgeting and planning. Furthermore, as the number and conditions of the brownfields change over time, the criteria system and corresponding categories can be updated and reapplied at the beginning of each fiscal year (or each planning horizon) to facilitate the government’s decision making. The proposed decision process can also be applied to assist the investment decisions of any brownfield stakeholders in other counties with limited funding based on their specific preference.

A prediction method for the cellphone replacement: Rough set-based analysis

Owing to the communication customers’ data are large scale, inconsistent and incomplete, it is hard to predict who will replace their cellphones in the near future. In order to solve this problem, a prediction method based on dominance-based rough set approach (DRSA) is proposed in this paper. Firstly, the customers’ features are extracted based on experts’ knowledge. So, we get the customer decision table. After that, the maximum certainty of each class union is used as the corresponding variable threshold value to control the decision rule extraction program. Therefore, the extracted decision rules can reflect the data characteristics effectively. In this process, it can learn knowledge automatically without setting the threshold values of consistency level depending on prior domain knowledge or by a complex tail-and-error procedure. Additionally, it strengthens the performance for dealing with inconsistent information systems. The efficiency of this method is illustrated by simulation experiments.

Identifying Strategic Development Objectives for African Countries Using Dominance-Based Rough Set Approach: The Poverty String Theory

The objective of this article is to expose the results of a research using Dominance-based Rough Set Approach (DRSA) to help African countries and international organizations (both non-governmental organizations and governmental organizations), to identify economical, sociological, political and technological strategic objectives for international development. We hope that the results of this research will aid politicians and leaders to prioritize African countries strategic development objectives according to political, economical, sociological and technological (PEST) needs. In this study we use 23 various indicators to classify all the African countries according to the following three different categories: [A] African countries that are doing well according to the selected indicators; [B] African countries that need support to acquire category A status; [C] African countries ranked the lowest and needing special support with regard to the criterion or criteria considered. The three categories are delimited by tertiles obtained from the average ranking of countries. The chosen criteria are measured in order to provide decision rules based on this classification. These decision rules thus focus on the political, economic, sociological and technological needs of countries with respect to improve their development and classification. We strongly believe that by targeting these identified needs, this research will help the development of African countries, target and prioritize International funding, evaluate economic growth and sociological improvements. Our results, from both the correlation matrix and DRSA, clearly demonstrate that top priority should be given to analphabetism, school life and reducing the amount of adolescents pregnancies in order to improve both economically and sociologically. Also, our analysis of the African map belonging to the overall classification results, puts the light over the fact that most countries in category C are, geographically connected to one another, what we named the “Poverty String’’. This is the first research of a series of three articles using DRSA in identifying strategic objectives for international development. The second research will discuss the use of DRSA to identify strategic objectives for Bosnia Herzegovina as a potential candidate to the European Union. The third research will use DRSA to help define poverty for all the United Nations countries and propose decision rules for international development.

Identifying Strategic Development Objectives for European Union’s Potential Candidate States Using Dominance-Based Rough Set Approach: Case Study of Bosnia and Herzegovina

This research is the second of a series of three researches to cope with strategic development objectives using Dominance-based Rough Set Approach (DRSA). The objective of this article is to expose the results of a research using DRSA to help the European Union (EU) identifying political, economical, sociological and technological strategic objectives for potential candidate countries planning to join the EU. Using the proposed methodology, politicians and leaders will be able to prioritize strategic development objectives according to political, economical, sociological and technological (PEST) needs of a specific candidate country to the EU. More precisely, the proposed methodology classifies all the European Union’s countries according to the following three different categories: [A] EU countries that are doing well according to the selected indicators; [B] EU countries that need support to acquire category A status; [C] EU countries ranked the lowest and needing special support with regard to the criterion or criteria considered. The three categories are delimited by tertiles relative to the average ranking of all EU countries including a potential candidate country, Bosnia and Herzegovina. Afterwards, DRSA provides decision rules based on this classification. These decision rules thus focus on the PEST needs of countries with respect to improve their development and classification by pointing out what was needed to be part of the different categories. We strongly believe that by targeting these identified needs, this research will help the development of the European Union’s economy, target and prioritize economical and sociological improvements with the use of strategic objectives for any candidate country. One of the results concerning our case study with Bosnia and Herzegovina is about the fact that this potential country has a weakness in the percentage of women in politics. Indeed, our research as shown that this criterion has an impact for the overall classification of the EU countries.

Incremental rough set approach for hierarchical multicriteria classification

Multicriteria classification refers to classify objects evaluated by a set of criteria to preference-ordered decision classes. Dominance-based rough set approach has been successfully introduced to express and reason inconsistencies with a dominance principle in multicriteria classification problems. Hierarchical attribute values exist extensively within many real-world applications, which provide a hierarchical form to organize, view and analyze data from different perspectives for accommodating the preference variability. In this study, we consider an extension of dominance-based rough set approach by applying an incremental learning technique for hierarchical multicriteria classification while attribute values dynamically vary across different levels of granulations. We formalize the dynamic characteristics of knowledge granules with the cut refinement and coarsening through attribute value taxonomies in the hierarchical multicriteria decision systems. In consequence, incremental algorithms for computing dominance-based rough approximations of preference-ordered decision classes are developed by applying the resulted prior-knowledge as the input, and only recomputing those outputs which depend on the changed attribute values. This paper presents the theoretical foundation of the proposed approach. Example analysis and experimental evaluation are also provided for illustration of the feasibility and efficiency.

Tolerance Dominance Relation in Incomplete Ordered Decision Systems

To deal with inconsistencies coming from consideration of criteria, that is, attributes with preference-ordered domains (scales), such as product quality, market share, and debt ratio, the dominance-based rough set approach has been proposed. However, it is not much useful for analyzing incomplete information. In this paper, we attempt to present research focusing on incomplete ordered decision systems where all attributes are considered as criteria and some objects’ criterion values are unknown. Under the assumption that all unknown values are the “do not care” semantics, the concept of tolerance dominance relation is introduced. Then, the rough approximations induced by it are obtained. And their properties are investigated. Especially, the important properties of rough approximations, such as rough inclusion, complementarity, identity, and monotonicity, are preserved well-known in the presented methodology. Moreover, an example is employed to illustrate these results. We expand the potential applications of dominance-based rough set approach.

Evidential probability of signals on a price herd predictions: Case study on solar energy companies

Many investors fail due to following prices. An imitating herd in the stock market influences not only people's wealth but economic stability. This research proposes a notion of price herd which simulates price behavior and its practice. Based on this concept, a model evidentially solving the signal probabilities of the price herd to predict its behavior is proposed. Empirically, the model is applied in the financial database, available from Taiwan Economic Journal, to analyze the solar energy industry during 2009–2014. In the results, it successfully identifies the herding signal, predicts the price downward 43% (0.99 close to the reality) in 2011, discloses a debtor herd beyond investors, and reveals the rational behavior of the price herd. Its technique centers in the prediction of information cascade, the induction of dominance-based rough set approach, and the approximations and granules of rough set theory.

Learning ensemble classifiers for diabetic retinopathy assessment.

Diabetic retinopathy is one of the most common comorbidities of diabetes. Unfortunately, the recommended annual screening of the eye fundus of diabetic patients is too resource-consuming. Therefore, it is necessary to develop tools that may help doctors to determine the risk of each patient to attain this condition, so that patients with a low risk may be screened less frequently and the use of resources can be improved. This paper explores the use of two kinds of ensemble classifiers learned from data: fuzzy random forest and dominance-based rough set balanced rule ensemble. These classifiers use a small set of attributes which represent main risk factors to determine whether a patient is in risk of developing diabetic retinopathy. The levels of specificity and sensitivity obtained in the presented study are over 80%. This study is thus a first successful step towards the construction of a personalized decision support system that could help physicians in daily clinical practice.