Comparative analysis of different techniques in classification based on association rules

Credit decision-making is a vital process in the banking sector as it helps to reduce losses by identifying non-creditable individuals. Classification algorithms in data mining provide accurate results in the aforementioned area. But, such real-world lending environments require classification results to be easy to interpret. The lack of explicability of several existing classifiers makes banks reluctant in using them. An ideal classifier needs to be accurate with interpretability encapsulated within it. Decision trees are accurate, but for large datasets, the tree becomes very large and may not be comprehensible. Genetic programming (GP) is widely applied for solving classification problems since it can produce smaller trees by using tree-size, as fitness measure or by depth-limiting the trees. Hence, we propose an algorithm named GPeCT that merges decision tree and GP to produce a near-optimal decision tree classifier. We demonstrate the performance of GPeCT through experiments on large datasets from banking and other domains.

Researchers in higher education are beginning to explore the potential of data mining in analyzing data for the purpose of giving quality service and needs of their graduates. Thus, educational data mining emerges as one tools to study academic data to identify patterns and help for decision making affecting the education. This paper predicts the employability of IT graduates using nine variables. First, different classification algorithms in data mining were tested making logistic regression with accuracy of 78.4 is implemented. Based on logistic regression analysis, three academic variables directly affect; IT_Core, IT_Professional and Gender identified as significant predictors for employability. The data were collected based on the five year profiles of 515 students randomly selected at the placement office tracer study.

Credit decision-making is a vital process in the banking sector as it helps to reduce losses by identifying non-creditable individuals. Classification algorithms in data mining provide accurate results in the aforementioned area. But, such real-world lending environments require classification results to be easy to interpret. The lack of explicability of several existing classifiers makes banks reluctant in using them. An ideal classifier needs to be accurate with interpretability encapsulated within it. Decision trees are accurate, but for large datasets, the tree becomes very large and may not be comprehensible. Genetic programming (GP) is widely applied for solving classification problems since it can produce smaller trees by using tree-size, as fitness measure or by depth-limiting the trees. Hence, we propose an algorithm named GPeCT that merges decision tree and GP to produce a near-optimal decision tree classifier. We demonstrate the performance of GPeCT through experiments on large datasets from banking and other domains.

Pattern mining algorithms for data streams using itemset

A compact and understandable associative classifier based on overall coverage

Associative classification is a supervised classification method. Many experimental studies have shown that associative classification is a promising approach. However, the latter suffer from a major drawback: the huge number of the generated classification rules which takes efforts to select the best ones in order to construct the classifier. To overcome such drawback, we propose in this paper a new direct associative classification method called IGARC, an improvement of GARC approach that extracts directly generic associative classification rules from a training set in order to reduce the number of associative classification rules without jeopardising the classification accuracy. Moreover, we propose an algorithm called PN-GARC that deals with negative classification rules. Considering negated items in classification framework provides additional information describing the data and reduces the conflicts while classifying new objects. Nevertheless, there are a sheer number of rules when considering negated items. That is why, we will explore generic classification rules both negative and positive ones in order to study their behaviour and their usefulness on the studied datasets. A detailed description of IGARC method is presented, as well as the experimentation study on 12 benchmark datasets proving that it is highly competitive in terms of accuracy in comparison with popular classification approaches.

It is a recurrent finding that software development is often troubled by considerable delays as well as budget overruns and several solutions have been proposed in answer to this observation, software fault prediction being a prime example. Drawing upon machine learning techniques, software fault prediction tries to identify upfront software modules that are most likely to contain faults, thereby streamlining testing efforts and improving overall software quality. When deploying fault prediction models in a production environment, both prediction performance and model comprehensibility are typically taken into consideration, although the latter is commonly overlooked in the academic literature. Many classification methods have been suggested to conduct fault prediction; yet associative classification methods remain uninvestigated in this context. This paper proposes an associative classification (AC)-based fault prediction method, building upon the CBA2 algorithm. In an empirical comparison on 12 real-world datasets, the AC-based classifier is shown to achieve a predictive performance competitive to those of models induced by five other tree/rule-based classification techniques. In addition, our findings also highlight the comprehensibility of the AC-based models, while achieving similar prediction performance. Furthermore, the possibilities of cross project prediction are investigated, strengthening earlier findings on the feasibility of such approach when insufficient data on the target project is available.

Associative classification (AC) is a branch of data mining that utilizes association rules (ARs) for classification. ARs are extracted from databases that satisfy some statistical criteria such as minimal support. However, in some practical applications, useful ARs may be found among infrequent, but closely related, itemsets that are filtered out by high minimal support. In this study, a new measure, named “condenseness,” is presented for evaluating whether infrequent ruleitems that are filtered out by minimal support can form strong ARs for classification. For an infrequent ruleitem, the condenseness is the average of lift of all ARs that can be generated from the ruleitem. A ruleitem with a high condenseness means that its elements are closely related and can serve for AC even if it does not have high support. Based on the concept of condenseness, a new associative classifier is developed and presented – condensed association rules for classification (CARC). CARC generates ARs using a modified Apriori algorithm and develops new strategies of rule inference. With the condenseness measure and strategies for rule inference, more useful ARs can be produced and improve the effectiveness of association classification. Empirical evidences show that CARC mitigates the problems caused by setting too high/low minimal support and has a better performance on classification.

Video classification is an important task for processing, analysis and retrieval of videos. The traditional method of Video classification generally use HMM theoretic. However, the HMM method has limitation to analysis video data. To solve this problem, we proposed a novel approach for video classification which uses the association rules. Firstly, we mined the actual dependence relationship between video states when the state model is constructed. Secondly, the reliability of dependence relationship is predicted with the restriction of association distance. Furthermore, the association rules are obtained by exploring state transition patterns. The experiment results demonstrate that the proposed method is efficient and suitable for various types of video data.

SummaryOne of the most important challenges of African countries is the effective use of information and communication technologies and its generalization in the different sectors (education, economic, political, and so on). This will have a great impact on different aspects of society and economic activities by making everyday procedures easier and more efficient. In this same context, we are interested in this article by the proposition of a decision support framework based on the use of data mining (DM) techniques. Using DM in e‐government is the process of translating data to appropriate knowledge which can be useful for decision‐making. We propose a framework whose goal is the generation of association rules (ARs) for better decision‐making. This framework includes two approaches: (1) the first approach applies different DM algorithms and (2) the second approach optimizes the first one by considering two different metaheuristics: the Genetic algorithm and the Cuckoo search algorithm. A new relevance measure called “Weighted Dominance” has been considered to evaluate the quality of the generated ARs. Extensive experiments have been conducted using different datasets. The results obtained demonstrated the effectiveness of combining DM and optimization algorithms compared to the first approach. Finally, a case study related to an Algerian mobile phone company has been presented illustrating the use of our framework in the decision‐making process.

Purpose Cement as one of the major components of construction activities, releases a tremendous amount of carbon dioxide (CO2) into the atmosphere, resulting in adverse environmental impacts and high energy consumption. Increasing demand for CO2 consumption has urged construction companies and decision-makers to consider ecological efficiency affected by CO2 consumption. Therefore, this paper aims to develop a method capable of analyzing and assessing the eco-efficiency determining factor in Iran’s 22 local cement companies over 2015–2019. Design/methodology/approach This research uses two well-known artificial intelligence approaches, namely, optimization data envelopment analysis (DEA) and machine learning algorithms at the first and second steps, respectively, to fulfill the research aim. Meanwhile, to find the superior model, the CCR model, BBC model and additive DEA models to measure the efficiency of decision processes are used. A proportional decreasing or increasing of inputs/outputs is the main concern in measuring efficiency which neglect slacks, and hence, is a critical limitation of radial models. Thus, the additive model by considering desirable and undesirable outputs, as a well-known DEA non-proportional and non-radial model, is used to solve the problem. Additive models measure efficiency via slack variables. Considering both input-oriented and output-oriented is one of the main advantages of the additive model. Findings After applying the proposed model, the Malmquist productivity index is computed to evaluate the productivity of companies over 2015–2019. Although DEA is an appreciated method for evaluating, it fails to extract unknown information. Thus, machine learning algorithms play an important role in this step. Association rules are used to extract hidden rules and to introduce the three strongest rules. Finally, three data mining classification algorithms in three different tools have been applied to introduce the superior algorithm and tool. A new converting two-stage to single-stage model is proposed to obtain the eco-efficiency of the whole system. This model is proposed to fix the efficiency of a two-stage process and prevent the dependency on various weights. Converting undesirable outputs and desirable inputs to final desirable inputs in a single-stage model to minimize inputs, as well as turning desirable outputs to final desirable outputs in the single-stage model to maximize outputs to have a positive effect on the efficiency of the whole process. Originality/value The performance of the proposed approach provides us with a chance to recognize pattern recognition of the whole, combining DEA and data mining techniques during the selected period (five years from 2015 to 2019). Meanwhile, the cement industry is one of the foremost manufacturers of naturally harmful material using an undesirable by-product; specific stress is given to that pollution control investment or undesirable output while evaluating energy use efficiency. The significant concentration of the study is to respond to five preliminary questions.

In order to solve the problems of high misevaluation rate and low work efficiency in the current mental health evaluation process of college students, a mental health evaluation system based on data mining algorithms is proposed. First, analyze the research status of college students’ mental health evaluation and data mining algorithms and build a mental health evaluation system framework; then, collect college students’ mental health questionnaire data, use the Apriori algorithm based on a three-dimensional matrix to analyze and classify, traverse each attribute of each transaction in the two-dimensional matrix, and directly obtain the frequent item set, frequent binomial set, and frequent three-item set by reading the three-dimensional attribute matrix and the mental health evaluation data to obtain mental health intelligent evaluation results. Finally, specific simulation experiments are used to analyze the feasibility and superiority of the mental health intelligent evaluation system. The results show that the system in the article overcomes the shortcomings of the current mental health intelligent evaluation system, improves the accuracy of mental health intelligent evaluation, improves the efficiency of mental health intelligent evaluation, and the system is more stable, which can meet the actual requirements of current college students’ mental health evaluation.

Discovering how Mesenchymal Stem Cells (MSCs) can be differentiated is an important topic in stem cell therapy and tissue engineering. In a general context, such differentiation analysis can be modeled as a classification problem in data mining. Specifically, this is concerned with the single-label multi-class classification task. Previous studies on this topic suggests the Associative Classification (AC) rather than other alternative (Classification) techniques, and presented classification results based on the CMAR (Classification based on Multiple Association Rules) associative classifier. Other AC algorithms include: CBA (Classification Based on Associations), PRM (Predictive Rule Mining), CPAR (Classification based on Predictive Association Rules) and TFPC (Total From Partial Classification). The main aim of this chapter is to compare the performance of different associative classifiers, in terms of classification accuracy, efficiency, number of rules to be generated, quality of such rules, and the maximum number of attributes in rule-antecedents, with respect to MSC differentiation analysis.

Although the healthcare sector is generally “information-rich,” sadly, not all data are mined, which is necessary for finding hidden patterns and making wise decisions. In medical research and especially in the prediction of heart disease, sophisticated data mining methods are used. These methods are also used to predict the data that is contained in databases. In this article, prediction systems for heart failure have been examined using a larger number of input attributes. The method analyses 13 variables, including medical words like sex, blood pressure, and cholesterol, to forecast the risk that a patient will develop heart disease. 13 attributes have been used for prediction. In this study, smoking and obesity were introduced as two more characteristics. In the existing systems, we have used Naïve Bayes and neural networking systems, which have less accuracy compared to the proposed system algorithms. On a database of heart diseases, the proposed data mining classification algorithms J48 Decision Trees, Bagging, and Adaboost are examined. Based on the accuracy, these techniques' performances are contrasted. According to our findings, the Adaboost, J48 decision tree, and bagging accuracy are 92.1 percent, 91.62 percent, and 90.52 percent, respectively. Our investigation reveals that Adaboost, out of these three classification methods, accurately predicts heart failure.

With the popularity of smart wearable systems, sensor signal processing poses more challenges to machine learning in embedded scenarios. For example, traditional machine-learning methods for data classification, especially in real time, are computationally intensive. The deployment of Artificial Intelligence algorithms on embedded hardware for fast data classification and accurate fall detection poses a huge challenge in achieving power-efficient embedded systems. Therefore, by exploiting the associative memory feature of Hopfield Neural Network, a hardware module has been designed to simulate the Neural Network algorithm which uses sensor data integration and data classification for recognizing the fall. By adopting the Hebbian learning method for training neural networks, weights of human activity features are obtained and implemented/embedded into the hardware design. Here, the neural network weight of fall activity is achieved through data preprocessing, and then the weight is mapped to the amplification factor setting in the hardware. The designs are checked with validation scenarios, and the experiment is completed with a Hopfield neural network in the analog module. Through simulations, the classification accuracy of the fall data reached 88.9% which compares well with some other results achieved by the software-based machine-learning algorithms, which verify the feasibility of our hardware design. The designed system performs the complex signal calculations of the hardware’s feedback signal, replacing the software-based method. A straightforward circuit design is used to meet the weight setting from the Hopfield neural network, which is maximizing the reusability and flexibility of the circuit design.