J48 Decision Tree Algorithm Research Articles

A comparative study of naive Bayes classifier and Bayes net classifier for fault diagnosis of roller bearing using sound signal

Bearing is an important and necessary part of any big or small machinery and for proper working of machinery the bearing condition should be good. Hence, there is a requirement for continuous bearing monitoring. For the condition monitoring of bearings sound signal can be used. This paper uses sound signal for condition monitoring of roller bearing by naive Bayes and Bayes net algorithms. The statistical features from the sound signals were extracted. Then features giving better results were selected using J48 decision tree algorithm. These selected features were classified using naive Bayes and Bayes net algorithm. The classification results for both naive Bayes and Bayes net algorithm for fault diagnosis of roller bearing using sound signals were compared and results were tabulated.

Fault diagnosis of helical gear box using variational mode decomposition and J48 algorithm

A faulty gear affects the functionality of the machine. Thus, it is necessary to diagnose the faults at an initial stage so as to reduce the losses incurred. Vibration signals contain dynamic information about the health condition of a rotating machine. Many researches are based on FFT and have its own drawback with non-stationary signals like the ones from gears. Hence, there is a need for development of new methodologies to infer diagnostic information from such signals. The vibrations produced by gears from good and simulated faulty conditions can be used to detect the faults in these gears. In the present study, variational mode decomposition (VMD) was used as a new signal pre-processing technique along with decision trees which have provided good classification performance. The technique decomposes the signals into various modes by identifying a compact frequency support around its central frequency, such that adding all the modes reconstructs the original signal. Descriptive statistical features were extracted from VMD processed signals. J48 decision tree algorithm was used to identify the useful features and the selected features were classified using the decision trees namely, J48, random tree and decision stump algorithm. The performance analyses of various algorithms are discussed in detail.

Application of Data Mining Techniques for Predicting CD4 Status of Patients on ART in Jimma and Bonga Hospitals, Ethiopia

Background: Many of the reports on HIV/AIDS shows that the number of ART registered patients are increasing from time to time. Despite those reports show increasing of patients’ number, they did not try to make prediction of attributes based on the given attributes more than statistical explanation. This study concerned to use data mining techniques on ART data base. The main objective of the study is to apply data mining techniques for predicting CD4 status of patients on ART in Jimma and Bonga Hospitals. Methodology: The study followed the CRISP-DM data mining methodology which has six phases: business understanding, data understanding, data preparation, model building, evaluation and deployment. For this study, data was taken from two hospitals of the south west of Ethiopia; Jimma and Bonga hospitals. Classification algorithm was used to predict CD4 status of the patients those who are following ART therapy. J48 is a technique used for building classification and PART is used to compare the result of J48 algorithm. Results: The best performance achieved by J48 decision tree algorithm is a generalized decision tree pruning with reduced attributes. The model classifies instances correctly (88.79%) and incorrectly (11.21%). The weighted average precision of the model is 0.88 with recall of 0.89 and ROC area of 0.85. The model has 760 numbers of leaves and 916 tree size. The time taken to build the model is 0.05 seconds. The analysis of this model shows that the model is quit efficient to predict CD4 status of patients those who are following ART. Conclusion: Classification done using J48 decision tree is the best model as compared to PART rule algorithm and that can be used for prediction. From the model built it is possible to conclude that attributes like: Eligible reason, ART status, ART start year, OA weight, OAWHO stage, Current regimen, Family planning, Functional status, Marital status, Past ARV are the most determining factors of CD4 status.

Application of Data Mining Techniques to Predict Adult Mortality: The Case of Butajira Rural Health Program, Butajira, Ethiopia

Background: Though adults are care providers and risk takers of a society, reports indicate that adult mortality conditions are not given much emphasis. This is due to a widespread perception that mortality among adults is low. Every year, more than 7•7 million children die before their fifth birthday; however, nearly 24 million of adults die under the age of 70 years. Identifying major determinants for adult death helps to alleviate the loss of the productive group. Therefore, this research is aimed to apply data mining techniques to build a model that can assist in predicting adult health status. Methods: The hybrid model that was developed for academic research was followed. Dataset was preprocessed for data transformation, missing values and outliers. WEKA 3.6.8 data mining tools and techniques such as J48 decision tree and Naive Bayes algorithms were employed to build the predictive model by using a sample dataset of 62,869 instances of both alive and died adults through three experiments and six scenarios. The area under the ROC curve for outcome class is used to evaluate performances of models from the predictive algorithms. Results: In this study as compared to Bayes, the performance of J48 pruned decision tree reveals that 97.2% of accurate results are possible for developing classification rules that can be used for prediction. If no education in family and the person is living in rural highland and lowland, the probability of experiencing adult death is 98.4% and 97.4% respectively with concomitant attributes in the rule generated. The likely chance of adult to survive in completed primary school, completed secondary school, and further education is (98.9%, 99%, 100%) respectively. Conclusion: Predictive model built with the use of data mining techniques suggests that education plays a considerable role as a root cause of adult death, followed by outmigration. The possibility of incorporating the findings of this study with knowledge based system should be explored so that experts can consult the system in their problem solving and decision making process. Further comprehensive and extensive experimentation is needed to substantially describe the loss experience of adult mortality in Ethiopia.

EXTRAÇÃO DE REGRAS DE CLASSIFICAÇÃO DE DADOS DE ESTAÇÃO METEOROLÓGICA PARA AUXÍLIO NA PREVISÃO DE ÍNDICE DE TEMPERATURA E UMIDADE PARA BOVINOS LEITEIROS

Forecasts for comfort index to dairy cattle are unavailable in Brazil and the extraction rules on weather behavior can assist in predicting the animal's comfort, especially for those who are in unprotected places. This study aims to develop a methodology for extracting predictive rules from heat stress conditions in dairy cattle. The analysis was performed using the database of the National Institute of Meteorology (INMET), referring to the hourly averages for the period between September 15th to November 13th of 2013 in Santa Maria - RS, Brazil. The input variables were time of day, air temperature, dew point temperature, relative humidity and the temperature and humidity index. The extraction of the rules was done by the technique of Data Mining and the classification task by building the J48 decision tree algorithm. The classification of the Temperature and Humidity Index (THI) was based on two classes, being NORMAL for THI values ​​less than or equal to 74, and ALERT to values ​​above 74, considered as a promoter of stress. Data mining has resulted in the description of 11 rules of the relationship between temperature, relative humidity and time of day with the THI. Data mining has enabled the understanding of the variables analyzed and the generated rules can help in forecasts based on meteorological forecasts and environmental temperature controllers and relative humidity schedule.

Entropy Variation and J48 Algorithm based Intrusion Detection System for Cloud Computing

Cloud Computing has achieved formidable impetus where IT infrastructures and applications are provided as service to end users. It provides shared pool of resources in addition with Data storage, computer processing power and specialized corporate and user applications. Users can access Cloud services any time anywhere and store large amount of data from anywhere, due to increases the popularity of Cloud computing there is risk of Security. Data which is stored on cloud may be vulnerable which is easy for attackers to compromise the virtual machines as zombies and explore these vulnerabilities in cloud system. Because when we move data or information in cloud we do not have any control on that data which can be handling by third party. Hence, there is the vital requirement of more security measures to protect cloud. In this paper we propose an Intrusion detection system which is based on Entropy variation and J48 Decision tree algorithm through which we can detect or prevent vulnerable virtual machines, Data center and Host from being compromised in the cloud also we can protect data and applications in Cloud like wide area network traffic. This proposed solution results gives more accuracy for attack detection and low false alarm rate. For simulation we use Cloud Sim (version 3.1) and used KDDCUP '99 Dataset to evaluate rules and testing datasets to detect intrusion.

Rapid and automated determination of rusted surface areas of a steel bridge for robotic maintenance systems

There has been an increased interest in the use of robotic systems to automate the blasting tasks in steel bridge maintenance. To utilize such robotic systems effectively, an automated process for determining the rusted areas on a steel bridge to be blasted is a prerequisite. This study proposes a method to rapidly and accurately determine rusted surface areas on a steel bridge that are to be blasted, within current standards. The proposed method consists of three steps: color space conversion, classification of rusted area via the J48 decision tree algorithm, and determination of blasting area. The method was validated using 119 test images showing both normal states and various degrees of rusting and rust distribution types. The experimental results showed that the success rate for determining the rusted areas that needed blasting was 97.48%. The average processing time was 0.57 s/image. The results demonstrate that the proposed method rapidly and accurately indicated whether blasting was necessary, and if so, where blasting should be performed, based on current standards of practice. • This study proposed a method to determine which rusted surface areas to be blasted. • Performance of the method was evaluated on 119 images acquired from steel bridges. • The success rate was 97.48% and the average processing time was only 0.57 s/image. • The method is fast, accurate, and in compliance with current standards of practice. • The method could be employed in robotic systems used for steel bridge maintenance.

Misfire detection in an IC engine using vibration signal and decision tree algorithms

Misfire in an IC engine continues to be a problem producing consequences like reduced fuel efficiency, increased power loss and emissions containing heavy concentration of uncombusted hydrocarbons. Misfiring creates a unique vibration pattern attributed to a particular cylinder. Useful features can be extracted from these patterns and can be analyzed to detect misfire. Statistical features of these vibration signals are extracted. Out of these, useful features were identified using the J48 decision tree algorithm and selected features are used with various decision trees. Classification accuracies from J48 algorithm, Best first tree algorithm, random forest tree algorithm, functional tree algorithm and linear model tree algorithm are compared and the best algorithm for such a system is suggested.

Analysis of Epileptic EEG Signals with Simple Random Sampling J48 Algorithm

This paper describes the application of a Simple Random Sampling J48 (SRS-J48) model for classification of electroencephalogram (EEG) signals. Decision making is performed in two stages: feature extraction and classification. Eight statistical features are extracted from a two-level sample set model based on SRS technique and then classified by the J48 decision tree algorithm in Weka. The classification accuracy of the SRS-J48 is 16.35% higher than that of J48 according to the five groups of experiment with only 13% execution time on average. Besides, the proposed SRS-J48 algorithm has competitive or even better results on some of the experimental groups than Siuly’s Simple Random Sampling-Least Square-Support Vector Machine (SRS-LS-SVM).

Understanding Vocalization Might Help to Assess Stressful Conditions in Piglets.

Simple SummaryThis research aimed to analyze the possibility of assessing piglets’ welfare using the records of their vocalization. The trial was done in a pig commercial farm, and we recorded the vocal signals from piglets in several stressful exposure situations. Data mining techniques were applied to the processed signals in order to obtain a stress classification using the recorded data. We found that, using the piglets’ vocalization, it was possible to identify the most frequent stressful conditions at the farrowing phase, namely: pain, cold and hunger.Assessing pigs’ welfare is one of the most challenging subjects in intensive pig farming. Animal vocalization analysis is a noninvasive procedure and may be used as a tool for assessing animal welfare status. The objective of this research was to identify stress conditions in piglets reared in farrowing pens through their vocalization. Vocal signals were collected from 40 animals under the following situations: normal (baseline), feeling cold, in pain, and feeling hunger. A unidirectional microphone positioned about 15 cm from the animals’ mouth was used for recording the acoustic signals. The microphone was connected to a digital recorder, where the signals were digitized at the 44,100 Hz frequency. The collected sounds were edited and analyzed. The J48 decision tree algorithm available at the Weka® data mining software was used for stress classification. It was possible to categorize diverse conditions from the piglets’ vocalization during the farrowing phase (pain, cold and hunger), with an accuracy rate of 81.12%. Results indicated that vocalization might be an effective welfare indicator, and it could be applied for assessing distress from pain, cold and hunger in farrowing piglets.

A Comparative Study of Hidden Markov Model and Support Vector Machine in Anomaly Intrusion Detection

This paper aims to analyse the performance of Hidden Markov Model (HMM) and Support Vector Machine (SVM) for anomaly intrusion detection. These techniques discriminate between normal and abnormal behaviour of network traffic. The specific focus of this study is to investigate and identify distinguishable TCP services that comprise of both normal and abnormal types of TCP packets, using J48 decision tree algorithm. The publicly available KDD Cup 1999 dataset has been used in training and evaluation of such techniques. Experimental results demonstrate that the HMM is able to classify network traffic with approximately 76% to 99% accuracy while SVM classifies it with approximately 80% to 99% accuracy.

A Data Mining Approach for the Detection of Denial of Service Attack

Denial of Service (DoS) attacks constitutes one of the major threats and among the hardest security problems currently facing computer networks and particularly the Internet. A DoS attack can easily exhausts the computing and communication resources of its victim within a short period of time. Because of the seriousness of the problem many defense mechanisms have been proposed to fight these attacks. In this paper, we propose an approach that detects DoS attacks using data mining classification techniques. The approach is based on classifying “normal” traffic against “abnormal” traffic in the sense of DoS attacks. The paper investigates and evaluates the performance of J48 decision tree algorithm for the detection of DoS attacks and compares it with two rule based algorithms, namely OneR and Decision table. The selected algorithms were tested with benchmark 1998 DARPA Intrusion Detection data. Our research results show that both Decision tree and rule based classifiers deliver highly accurate results – greater than 99% accuracy – and exhibit high level of overall performance. DOI: http://dx.doi.org/10.11591/ij-ai.v2i2.1937

Detecting flooded areas with machine learning techniques: case study of the Selška Sora river flash flood in September 2007

Floods seem to appear with increased frequency from one year to another. They create great damage to property and in some cases even result in lost lives. However, a quick and effective response by rescue services can greatly reduce the consequences. Machine learning techniques can reduce the time necessary for flood mapping. We test various machine learning methods to find the one with the highest classification accuracy. We also present the most important points for quick and effective machine learning procedures on remote sensing data. First, the data must be prepared correctly. We use satellite images, digital terrain models (DTMs), and the river network. The data in its primary form (e.g., bands of multispectral satellite images or DTMs) is insufficient. We also need certain derived attributes, such as the vegetation index or the slope derived from the DTM. Second, we must select suitable training samples and a suitable machine learning method. This approach to determining floods is presented in a case study of flash floods in the Selška Sora river valley. Machine learning techniques have proven successful in quickly determining flooded areas. The best results are produced by the J48 decision tree algorithm. The success of the ensemble machine learning methods is comparable to the J48 algorithm, while the JRip classification is not as good.

A Decision Tree Algorithm Pertaining to the Student Performance Analysis and Prediction

of an educational institute can be measured in terms of successful students of the institute. The analysis related to the prediction of students academic performance in higher education seems an essential requirement for the improvement in quality education. Data mining techniques play an important role in data analysis. For the construction of a classification model which could predict performance of students, particularly for engineering branches, a decision tree algorithm associated with the data mining techniques have been used in the research. A number of factors may affect the performance of students. Here some significant factors have been considered while constructing the decision tree for classifying students according to their attributes (grades). In this paper four different decision tree algorithms J48, NBtree, Reptree and Simple cart were compared and J48 decision tree algorithm is found to be the best suitable algorithm for model construction. Cross validation method and percentage split method were used to evaluate the efficiency of the different algorithms. The traditional KDD process has been used as a methodology. The WEKA (Waikato Environment for Knowledge Analysis) tool was used for analysis and prediction. . Results obtained in the present study may be helpful for identifying the weak students so that management could take appropriate actions, and success rate of students could be increased sufficiently.

Comparative analysis of J48 with statistical and machine learning methods in predicting fault-prone classes using object-oriented systems

There are available metrics for predicting fault prone classes, which May help software organizations for planning and performing testing activities. This May be possible due to proper allocation of resources on fault prone parts of the design and code of the software. Hence, importance and usefulness of such metrics is understandable, but empirical validation of these metrics is always a great challenge. J48 decision tree algorithm has been successfully applied for solving classification problems in many applications. This paper evaluates the capability of algorithm and compares its performance with nine statistical and machine learning methods in predicting fault prone software classes using publicly available NASA data set. The results indicate that the prediction performance of J48 is generally better than other statistical and machine learning models. However, similar types of studies are required to be carried out in order to establish the acceptability of the J48 model.

Wavelet decomposition and support vector machine for fault diagnosis of monoblock centrifugal pump

Monoblock centrifugal pumps play a very critical role in a variety of applications and condition monitoring of the various mechanical components of centrifugal pump becomes essential which in turn increases the productivity and reduces the breakdowns. Vibration-based continuous monitoring and analysis using machine learning approaches are gaining momentum. Particularly, artificial neural networks fuzzy logic was employed for continuous monitoring and fault diagnosis. This paper presents the use of support vector machine (SVM) algorithm for fault diagnosis through discrete wavelet features extracted from vibration signals of good and faulty conditions of the components of centrifugal pump. The classification accuracies were computed for different types of classifiers such as artificial neural network (ANN), support vector machine (SVM) and J48 decision tree algorithm.

Quantitative Assessment of Abdominal Aortic Aneurysm Geometry

Recent studies have shown that the maximum transverse diameter of an abdominal aortic aneurysm (AAA) and expansion rate are not entirely reliable indicators of rupture potential. We hypothesize that aneurysm morphology and wall thickness are more predictive of rupture risk and can be the deciding factors in the clinical management of the disease. A non-invasive, image-based evaluation of AAA shape was implemented on a retrospective study of 10 ruptured and 66 unruptured aneurysms. Three-dimensional models were generated from segmented, contrast-enhanced computed tomography images. Geometric indices and regional variations in wall thickness were estimated based on novel segmentation algorithms. A model was created using a J48 decision tree algorithm and its performance was assessed using ten-fold cross validation. Feature selection was performed using the χ2-test. The model correctly classified 65 datasets and had an average prediction accuracy of 86.6% (κ=0.37). The highest ranked features were sac length, sac height, volume, surface area, maximum diameter, bulge height, and intra-luminal thrombus volume. Given that individual AAAs have complex shapes with local changes in surface curvature and wall thickness, the assessment of AAA rupture risk should be based on the accurate quantification of aneurysmal sac shape and size.

Fault diagnosis of spur bevel gear box using discrete wavelet features and Decision Tree classification

The wavelet transform (WT) is used to represent all possible types of transients in vibration signals generated by faults in a gear box. It is shown that the transform provides a powerful tool for condition monitoring and fault diagnosis. The vibration signal of a spur bevel gear box in different conditions is used to demonstrate the application of various wavelets in feature extraction. In present work, a discrete wavelet, Daubechies wavelets (db1–db15) is used for feature extraction and their relative effectiveness in feature extraction is compared. The major steps in pattern classification are feature extraction and classification. This paper investigates the use of discrete wavelets for feature extraction and a Decision Tree for classification. J48 Decision Tree algorithm has been used for feature selection as well as for classification. This paper illustrates the powerfulness and flexibility of the discrete wavelet transform to decompose linear and non-linear processing of vibration signal.