Search results

The purpose of this study is to alleviate the specified issues to a great extent. To promote patients’ health via early prediction of diseases, knowledge extraction using data mining approaches shows an integral part of e-health system. However, medical databases are highly imbalanced, voluminous, conflicting and complex in nature, and these can lead to erroneous diagnosis of diseases (i.e. detecting class-values of diseases). In literature, numerous standard disease decision support system (DDSS) have been proposed, but most of them are disease specific. Also, they usually suffer from several drawbacks like lack of understandability, incapability of operating rare cases, inefficiency in making quick and correct decision, etc.

Addressing the limitations of the existing systems, the present research introduces a two-step framework for designing a DDSS, in which the first step (data-level optimization) deals in identifying an optimal data-partition (Popt) for each disease data set and then the best training set for Popt in parallel manner. On the other hand, the second step explores a generic predictive model (integrating C4.5 and PRISM learners) over the discovered information for effective diagnosis of disease. The designed model is a generic one (i.e. not disease specific).

The empirical results (in terms of top three measures, namely, accuracy, true positive rate and false positive rate) obtained over 14 benchmark medical data sets (collected from https://archive.ics.uci.edu/ml) demonstrate that the hybrid model outperforms the base learners in almost all cases for initial diagnosis of the diseases. After all, the proposed DDSS may work as an e-doctor to detect diseases.

The model designed in this study is original, and the necessary parallelized methods are implemented in C on Cluster HPC machine (FUJITSU) with total 256 cores (under one Master node).

Spray curing has become the preferred curing method for most cement concrete members because of its lower cost and sound effect. However, the spray curing quality of members is vulnerable to random variation environment factors and anthropogenic interferences. This paper aims to introduce the machine learning algorithm into the spray curing system to optimize its control method to improve the spray curing quality of members.

The critical parameters affecting the spray curing quality of members were collected through experiments, such as the temperature and humidity of the member's surface, the temperature, humidity and wind speed of the environment. The C4.5 algorithm was used as a weak classifier algorithm, and the AdaBoost.M1 algorithm was used to cascade multiple weak classifiers to form a robust classifier according to the collected data.

The results showed that the model constructed by the AdaBoost.M1 algorithm had achieved higher accuracy and robustness among the two algorithms. Based on the classification model built by the AdaBoost.M1 algorithm, the spray curing system can cause automatic decision-making spray switching according to the member's real-time curing state and environment.

With the classification model constructed by the AdaBoost.M1 algorithm, the spray curing system can overcome the disadvantages that external factors greatly influence the current control method of the spray curing system, and the intelligent control of the spray curing system was realized to a certain extent. This paper provides a reference for applying machine learning algorithms in the intellectual transformation of bridge construction equipment.

The purpose of this paper is to build a new hierarchical intrusion detection system (IDS) based on a binary tree of different types of classifiers. The proposed IDS model must possess the following characteristics: combine a high detection rate and a low false alarm rate, and classify any connection in a specific category of network connection.

To build the binary tree, the authors cluster the different categories of network connections hierarchically based on the proportion of false-positives and false-negatives generated between each of the two categories. The built model is a binary tree with multi-levels. At first, the authors use the best classifier in the classification of the network connections in category A and category G2 that clusters the rest of the categories. Then, in the second level, they use the best classifier in the classification of G2 network connections in category B and category G3 that represents the different categories clustered in G2 without category B. This process is repeated until the last two categories of network connections. Note that one of these categories represents the normal connection, and the rest represent the different types of abnormal connections.

The experimentation on the labeled data set for flow-based intrusion detection, NSL-KDD and KDD’99 shows the high performance of the authors' model compared to the results obtained by some well-known classifiers and recent IDS models. The experiments’ results show that the authors' model gives a low false alarm rate and the highest detection rate. Moreover, the model is more accurate than some well-known classifiers like SVM, C4.5 decision tree, MLP neural network and naïve Bayes with accuracy equal to 83.26 per cent on NSL-KDD and equal to 99.92 per cent on the labeled data set for flow-based intrusion detection. As well, it is more accurate than the best of related works and recent IDS models with accuracy equal to 95.72 per cent on KDD’99.

This paper proposes a novel hierarchical IDS based on a binary tree of classifiers, where different types of classifiers are used to create a high-performance model. Therefore, it confirms the capacity of the hierarchical model to combine a high detection rate and a low false alarm rate.

The growth of the web and the increasing number of documents electronically available has been paralleled by the emergence of harmful web pages content such as pornography, violence, racism, etc. This emergence involved the necessity of providing filtering systems designed to secure the internet access. Most of them process mainly the adult content and focus on blocking pornography, marginalizing violence. The purpose of this paper is to propose a violent web content detection and filtering system, which uses textual and structural content‐based analysis.

The violent web content detection and filtering system uses textual and structural content‐based analysis based on a violent keyword dictionary. The paper focuses on the keyword dictionary preparation, and presents a comparative study of different data mining techniques to block violent content web pages.

The solution presented in this paper showed its effectiveness by scoring a 89 per cent classification accuracy rate on its test data set.

Many future work directions can be considered. This paper analyzed only the web page, and an additional analysis of the visual content can be one of the directions of future work. Future research is underway to develop effective filtering tools for other types of harmful web pages, such as racist, etc.

The paper's major contributions are first, the study and comparison of several decision tree building algorithms to build a violent web classifier based on a textual and structural content‐based analysis for improving web filtering. Second, showing laborious dictionary building by finding automatically discriminative indicative keywords.

The purpose of this paper is to explore the effective taxonomies of airline green operations strategy.

To this end, a sample of 23 airlines from five regions (North America, South America, Europe, Asia and the Middle East) was surveyed. The annual sustainability reports of the surveyed airlines for the period 2013‒2016 were retrieved from the Global Reporting Initiatives website. K-means clustering analysis was used to generate taxonomic clusters of airline green operations strategy. A special data analysis technique, called rank analysis, was also adopted to identify the significant green actions and develop indicative models.

This study revealed that three effective taxonomies were adopted by airlines: a low-effect strategic pattern, a low-to-moderate effect strategic pattern and a high-effect strategic pattern. A different combination of green operation actions characterized each strategic pattern.

The research contribution of taxonomies of green operations strategy has so far been limited, country focused and concentrated on the manufacturing sector. This study reported the taxonomies and performed an in-depth analysis of the categories of effective actions taken to promote green performance. Moreover, this study developed indicative models for the relationship between categories of action and green performance for each strategic pattern, an action that has seldom been reported by previous studies of green operations strategies for airlines.

Data mining is widely considered necessary in many business applications for effective decision-making. The importance of business data mining is reflected by the existence of numerous surveys in the literature focusing on the investigation of related works using data mining techniques for solving specific business problems. The purpose of this paper is to answer the following question: What are the widely used data mining techniques in business applications?

The aim of this paper is to examine related surveys in the literature and thus to identify the frequently applied data mining techniques. To ensure the recent relevance and quality of the conclusions, the criterion for selecting related studies are that the works be published in reputed journals within the past 10 years.

There are 33 different data mining techniques employed in eight different application areas. Most of them are supervised learning techniques and the application area where such techniques are most often seen is bankruptcy prediction, followed by the areas of customer relationship management, fraud detection, intrusion detection and recommender systems. Furthermore, the widely used ten data mining techniques for business applications are the decision tree (including C4.5 decision tree and classification and regression tree), genetic algorithm, k-nearest neighbor, multilayer perceptron neural network, naïve Bayes and support vector machine as the supervised learning techniques and association rule, expectation maximization and k-means as the unsupervised learning techniques.

The originality of this paper is to survey the recent 10 years of related survey and review articles about data mining in business applications to identify the most popular techniques.

The purpose of this research is to explore the critical success factors of mobile animation games, by exploring the game itself, information systems, game motivation and promotional activities, as well as conducting research and analysis on mobile animation game players.

This research used the Analysis Hierarchy Process (AHP) method and the consistent fuzzy preference relationship for data analysis. In this study, collect 1,286 valid questionnaires through online questionnaire surveys. And comparing the two games “Legend Showdown” and “Tower of Gods and Demons”, players believe that the more successful mobile animation game is “Legend Showdown”.

Through experimental design, and the consistent fuzzy preference relationship for data analysis. The results found that the critical factors considered by the player in relation to the mobile animation game were firstly the information system, followed by promotional activities, game motivation and finally the game itself.

Because this research does not involve the concept of fuzzy theory at all, it is easy to produce subjective, uncertain and ambiguity issues when comparing pairwise. We recommended that follow-up researchers can use fuzzy semantic preference relations to solve this problem.

This study proposed a new approach that takes the critical factors for the mobile animation game. According to the research results, the critical success factor of mobile animation games is the information system, as it could provide a reference direction for game manufacturers when designing or formulating marketing strategies in the future.

The purpose of this study is to analyse various risks associated with third-party logistics (3PL) in the coffee supply chain and to present a framework that computes the influence of these risks on the critical success factors of the coffee supply chain.

The risks have been identified through a comprehensive literature review and validation by industry experts. The paper utilises an interpretive structural modelling (ISM) methodology for developing a hierarchical relationship among the CSFs. Furthermore, fuzzy MICMAC analysis is carried out to categorise these CSFs based on their driving power and dependence value. The fuzzy technique for order preferences by the similarity of an ideal solution (fuzzy-TOPSIS) approach has been applied to prioritise the risks associated with 3PL based on their ability to influence the CSFs of the coffee SC. Furthermore, we performed a sensitivity analysis to analyse the stability of the results obtained in this study.

This study illustrates ten risks associated with 3PL and five CSFs in the coffee supply chain. The analysis revealed that coffee enterprises need to develop a balanced pricing strategy to ensure a sustainable competitive advantage, whereas the lack of direct customer communication is the most dominant 3PL risk affecting the CSFs.

This research provides coffee enterprises with a generalised framework with set parameters that can be used to attain a successful coffee supply chain in any developing nation.

The study contributes to the literature by being the first kind of study, which has used fuzzy ISM-MICMAC to analyse the CSFs of the coffee supply chain and fuzzy-TOPSIS for analysing the impact of various risks associated with the 3PL in the coffee supply chain. Thus, this work can be considered a benchmark for future research and advancement in the coffee business field.

Worker activity identification and classification is the most crucial and difficult stage in work sampling studies. Manual methods of recording are tedious and prone to error and, hence automating the task of observing and classifying worker activities is an important step towards improving the current practice. Very recently, accelerometer-based systems have been explored to automate activity recognition in construction, but it had been carried out in controlled environment. The purpose of this paper is to cover the evaluation of the system in field situations.

Experimental investigation was carried out on crews of iron workers and carpenters with accelerometer data loggers worn at selected locations on the human body. The accelerometer data collection was spread over a time period of two weeks, and video recording of the worker activities was concurrently carried out to serve as ground truth, the reference used for comparison. The activity recognition analysis was carried out on accelerometer data features using a decision tree algorithm.

It was found that the classification using the individual training scheme performed better when compared with the collective training scheme for both the trades. The field studies results showed that the classification accuracies for iron work and carpentry are 90.07 and 77.74 per cent, respectively, using decision tree classifier. It was found that similarities of movements were a major cause for lower accuracy of recognition.

The work being preliminary in nature has used the basic classifier and pre-processing methods and, standard settings of algorithms.

The paper has investigated accelerometer-based method for construction labour activity classification in field situations.

Along with the ever growingWeb is the proliferation of objectionable content, such as sex, violence, racism, etc. We need efficient tools for classifying and filtering undesirable web content. In this paper, we investigate this problem through WebGuard, our automatic machine learning based pornographic website classification and filtering system. Facing the Internet more and more visual and multimedia as exemplified by pornographic websites, we focus here our attention on the use of skin color related visual content based analysis along with textual and structural content based analysis for improving pornographic website filtering. While the most commercial filtering products on the marketplace are mainly based on textual content‐based analysis such as indicative keywords detection or manually collected black list checking, the originality of our work resides on the addition of structural and visual content‐based analysis to the classical textual content‐based analysis along with several major‐data mining techniques for learning and classifying. Experimented on a testbed of 400 websites including 200 adult sites and 200 non pornographic ones, WebGuard, our Web filtering engine scored a 96.1% classification accuracy rate when only textual and structural content based analysis are used, and 97.4% classification accuracy rate when skin color related visual content based analysis is driven in addition. Further experiments on a black list of 12 311 adult websites manually collected and classified by the French Ministry of Education showed that WebGuard scored 87.82% classification accuracy rate when using only textual and structural content‐based analysis, and 95.62% classification accuracy rate when the visual content‐based analysis is driven in addition. The basic framework of WebGuard can apply to other categorization problems of websites which combine, as most of them do today, textual and visual content.