Classification Tree Technique Research Articles

The Implementation of Decision Tree Classification Techniques to Predict the Duration of Students Completing the Thesis at PTIK FKIP UNS

<span>This paper aims to determine the reasons why students take a long time in compiling their thesis. The slowness of students in compiling will have an impact on their graduation. This is a serious problem faced by educational institutions. Out of 328 students at PTIK FKIP UNS who took thesis credits, only 85 were able to graduate on time. Therefore, this study was conducted to identify the causes. The research data was taken from the alumnus class of 2012 to 2017. The data was processed using RapidMiner software. The technique used was the decision tree classification technique with the C4.5 algorithm, and to optimize the accuracy of the model, the Particle Swarm Optimization (PSO) algorithm was also added. This study got an accuracy rate of 76% and an AUC score of 0.733.</span>

Factors associated with the use of a public information system of dentist-prescribed antibiotics in Minas Gerais, Brazil.

This article aims to investigate the association between socioeconomic factors, health care organizations, and the use of a management and monitoring system for the dispensing of antibiotics prescribed by dentists in public health services in Minas Gerais, Brazil. This is an ecological-epidemiological study that analyzed secondary data from the Integrated Pharmaceutical Care Management System (SIGAF) of the Department of Health of the state of MG, Brazil, in 2017. Thirteen independent variables were analyzed to assess their influence on municipal adherence to SIGAF system considering dental prescriptions of antibiotics. Descriptive statistical analyses were performed, and the Classification and Regression Tree technique was used to identify the municipal variables associated with the outcome. A total of 57,279 antibiotic courses prescribed by dentists and recorded in SIGAF were examined. Socioeconomic factors were not associated with the use of SIGAF to record these prescriptions. Oral healthcare coverage was positively associated with the use of SIGAF for the dispensing of antibiotics prescribed by dentists. Dental Specialties Center were negatively associated with the outcome. Municipalities with high oral healthcare coverage and those without a Dental Specialties Center were more likely to use SIGAF.

Early assessment of the risk of child welfare involvement for preventive purposes

BackgroundTo prevent child welfare involvement by timely referring families to preventive interventions, it is essential to assess the risk of child welfare involvement in the general population as early as possible. ObjectiveTo examine whether child welfare involvement can be predicted by demographic, socioeconomic, and criminal history factors. Participants and settingData of 131,532 children and their parents were retrieved from Statistics Netherlands. Child welfare involvement comprised supervision orders, court-imposed guardianship, and/or out-of-home placements, that started in the year following the assessment of risk factors. MethodsAUC values were calculated to establish the predictive validity of individual factors for future child welfare involvement. A decision tree classification technique with a split-sample validation approach was used to develop and validate a predictive model. ResultsAn accumulation of risk factors proved to be highly predictive of child welfare involvement in the year following the assessment of the risk factors. The risk increased exponentially as the number of risk factors increases. The presence of four or more risk factors yielded a 10 times higher risk for child welfare involvement, and the presence of six or more risk factors yielded a 21 times higher risk relative to the absence of risk factors. The predictive model showed that above an accumulation of risk factors, the presence of specific risk factors further increased the risk. ConclusionsBoth the predictive model and the mere accumulation of risk factors can help professionals to estimate the risk of future child welfare involvement, which is important for timely referring families to preventive interventions or assistance.

The optimal level of factors for high daily milk yield in automatic milking system

Optimising daily milk yield in automatic milking systems (AMS) became one of the basic issues raised by many farmers. The aim of this study was to determine the best combination of factors and their levels suggested by the decision tree for high daily milk yield per AMS in dairy cattle herds. The study involved 4854 Polish Holstein-Friesian cows milked in 20 barns with Lely AMS. Statistical analysis was performed using two methods: multifactorial analysis of variance and classification tree technique. Milk yield, number of cows, free robot time, milking speed, and cow treatment time (i.e. cleaning of teats before and after milking) were the significant determinants of total milk yield for all cows per milking robot and were most frequently used to construct the decision tree. The tree shows that the highest total milk yield per robot was observed for the group of cows that produced more than 30 kg of milk per day, with a milking speed of over 2.40 kg/min. and during robot days in which free time did not exceed 10% of the day. In AMS dairy herds, efforts should be made to select animals that transmit high milk yield and high milking speed while shortening free robot time and increasing daily milk yield per robot. Also, attention should be given to the percentage of box time spent on preparing the cow for milking and after milking because prolongation of this variable has a negative effect on daily milk yield per AMS.

Secure and Sustainable Framework for Cattle Recognition Using Wireless Multimedia Networks and Machine Learning Techniques

Towards the successful operation of any farm, effective livestock management is necessary. The efficiency, affordability, and scalability of management solutions play a crucial role in modern farming. In this paper, a novel low-cost framework is proposed to recognize for health monitoring of individual cattle based on their sensed behavioral activity data and muzzle (nose) pattern image database using incremental decision tree classification techniques and accelerometer-based activity monitoring method. The proposed system performs image prepossessing on the captured muzzle point images of cattle to mitigate the noise, enhance the contrast, and increase quality. We extract the minutiae features to improve the system’s accuracy for recognizing the cattle for accurate and early classification of behavioral activity for health monitoring. The proposed system uses a support vector machine and incremental decision tree classifiers to classify the extracted feature of cattle’s muzzle images and health database. The server has encrypted databases that consist of captured muzzle images, owner database, health sensor database provided by the owners and systems. We use a similarity score measurement using incremental decision tree classification for matching and classify the muzzle point featutes with the database for cattle identification and health monitoring. We also developed a prototype for evaluating the accuracy of the proposed system with 97.99% accuarcy for unique identification of individual cattle.

Detection of Cardiovascular Disease Using Ensemble Feature Engineering With Decision Tree

Cardiovascular diseases are a cluster of heart-related issues, including many comorbidities, which are becoming a leading cause of human death across the globe. Hence, an essential framework is demanded for the early detection of CVDs which can help to prevent premature death. The application of Artificial Intelligence (AI) in healthcare has opted for this challenge and makes it easier to detect CVDs using a computational model. In this study, the authors built a reduced dataset using ensemble feature selection methods and got five features as per their weight values. Support Vector Machine, Logistic Regression, and Decision Tree classification techniques are utilized to check the effectiveness of newly designed datasets through different validation approaches. The authors also worked on data processing and visualization techniques, including Principal Component Analysis (PCA), and T-sne for understanding the data structure. From the findings, it was possible to conclude that DT has achieved an optimal accuracy and AUC of 98.9% and 0.99 ROC with leave one out Cross Validation (CV).

Retraction Note to: Automatic classification of emotions in news articles through ensemble decision tree classification techniques

Retraction Note to: Automatic classification of emotions in news articles through ensemble decision tree classification techniques

Trastorno específico del lenguaje en Andalucía, España: prevalencia en función del subtipo y del género

IntroductionSpecific Language Impairment (SLI) is a significant impairment in language acquisition (−1.25 SD), resistant to intervention and not justified by physical, neurological, intellectual, sensory or social causes, requiring specialized intervention. There is little information on the prevalence of this group in Spain. ObjectivesThree objectives, to calculate the prevalence of SLI in general and subtypes globally and according to gender, to determine the ages of diagnosis of SLI and to know its distribution by provinces in Andalusia. Materials and methodsPrevalence: Reference population: 537,879 Primary School students. The prevalence is calculated by types, provinces and sex and homogeneity with hierarchical cluster is analyzed. Detection ages and SLI allocation in Andalusia: Reference population: 1,185,061 students. Frequencies and percentages of SLI are calculated by typology, province, age group and sex. Classification tree technique (CHAID method) is used. ResultsPrevalence: 8.27 per thousand (‰) in Andalusia (Spain). Screening ages. At age 8, 86.9%, mainly expressive SLI (E-SLI). Semantic-pragmatic disorder (SPD) is detected mainly from the age of 8 years. DistributionSex: 74% are boys. Age groups: 2.9% are between 3-6 years old, 32% between 6 and 8, 42% between 8 and 12 and 23% over 12. Typologies: expressive SLI (E-SLI): 56%, expressive-receptive SLI (ER-SLI): 33% and semantic-pragmatic disorder (SPD): 11%. ConclusionsSLI students, due to their specific characteristics and dimensions, should be considered in educational policies.

ANALISIS KLASIFIKASI REKAPITULASI PENGADUAN PELANGGAN UP3 PT. PLN SEMARANG MENGGUNAKAN ALGORITMA QUEST (QUICK, UNBIASED, AND EFFICIENT STATISTICAL TREE)

Every company must have a way to solve the problems faced by its customers, PT. PLN Persero, the Indonesian national energy utility, must have a method to handle consumer complaints. PT. PLN Persero has a recovery time strategy for resolving consumer concerns, but it is not always effective in doing so. The QUEST algorithm (Quick, Unbiased, and Efficient Statistical Tree) approach is used to classify the problem of the recovery time policy failing on specific complaints. Classification of complaint data in order to obtain characteristics and factors as the main influence on the complaints and be able to provide new opinions for PT. PLN to address customer complaints. The QUEST method is a classification tree technique with two nodes per split that yields an unbiased variable. The QUEST method may be used with both category and numerical data. QUEST uses three stages to create a classification tree: picking the splitting variable, identifying the split point, and pausing the split. The classification tree generated has a tree depth of four layers and obtained three essential factors in the classification, namely weather, the number of customers experiencing the same event, and distance from the site. The classification tree accuracy level is 0.851 (or 85.1%), with a prediction error rate of 0.149 (or 14.9%).Keywords: binary classification tree, recovery time, QUEST algorithm.

Comparison of Different Machine Learning Techniques to Predict Diabetic Kidney Disease.

Background Diabetic kidney disease (DKD), one of the complications of diabetes in patients, leads to progressive loss of kidney function. Timely intervention is known to improve outcomes. Therefore, screening patients to identify high-risk populations is important. Machine learning classification techniques can be applied to patient datasets to identify high-risk patients by building a predictive model. Objective This study aims to identify a suitable classification technique for predicting DKD by applying different classification techniques to a DKD dataset and comparing their performance using WEKA machine learning software. Methods The performance of nine different classification techniques was analyzed on a DKD dataset with 410 instances and 18 attributes. Data preprocessing was carried out using the PartitionMembershipFilter. A 10-fold cross validation was performed on the dataset. The performance was assessed on the basis of the execution time, accuracy, correctly and incorrectly classified instances, kappa statistics (K), mean absolute error, root mean squared error, and true values of the confusion matrix. Results With an accuracy of 93.6585% and a higher K value (0.8731), IBK and random tree classification techniques were found to be the best performing techniques. Moreover, they also exhibited the lowest root mean squared error rate (0.2496). There were 15 false-positive instances and 11 false-negative instances with these prediction models. Conclusions This study identified IBK and random tree classification techniques as the best performing classifiers and accurate prediction methods for DKD.

University graduates’ segmentation: determinant factors

ABSTRACT The opinions of students constitute valuable information to be taken into account in university management. In this study, university satisfaction is analyzed, by means of a proxy variable (intention to repeat the same university course) following a holistic approach that analyses its relation with socio-demographic characteristics of the university, university experience, and information on postgraduates and their employment situation. Data from the Instituto de Estadistica Nacional de España (National Institute of Statistics of Spain) are employed, with a sample of over 30,000 graduates. Graduate satisfaction profiles are identified using the classification tree technique, as are the variables with greater discriminatory power that explain the intention of a student to repeat the same university course (employment situation, geographical zone, scientific discipline, and scholarships, particularly Erasmus grants); as well as the segments of university graduates. The information leads us to present recommendations for taking decisions in the field of education, general policy, territorial policy, and the management of each university.

Helping Others Helps Me: Prosocial Behavior and Satisfaction With Life During the COVID-19 Pandemic.

Prosocial behavior (PsB) and its effects have been analyzed in times of crisis and natural disasters, although never before in the face of such exceptional circumstances as those created by the COVID-19 pandemic. This research analyzes the role of PsB on satisfaction with life (SWL) in Colombia, considering the negative emotional impact of events that began in February 2020. We conduct an exploratory analysis using a sample of Colombia’s general population (N = 2,574; 53.2% women) with an average age of 44.66 years (SD = 15.36). Using the Classification Tree technique, we find that engaging in one or more PsBs (e.g., donating money or sharing food) enhances SWL and decreases the impact of negative emotions such as pessimism, indecisiveness, and irritability that have emerged during the COVID-19 pandemic. These findings are significant because they confirm the importance of life satisfaction as a personal resource for coping with complex situations and provide evidence of the benefits of PsB on one’s wellbeing.

분류나무 기법에 기초한 공급사슬 회복성, 강건성, 민첩성과 공급사슬 성과의 평가

Purpose This study empirically analyzes the impact of three types of supply chain strategies(resilience, robustness, agility) for supply chain risk on supply chain performances. Methods We statistically measure the main types of supply chain strategies into resilience, robustness and agility using data collected by survey methods. SC strategies'' performance is evaluated by adopting general supply chain performances. In order to unveil more precise causality relationship between strategies and performance, we use classification tree technique in the field of data mining. Results The results reveal that all types of supply chain strategies are important for overall performances. In particular, regulating supply chain process for reducing cycle time is crucial for supply chain quality. Besides, meeting customer needs is essential for order fulfillment, recovering material logistics is for delivery speed, meeting customer needs is for partner satisfaction, and recovering material logistics is for independent delivery. Conclusion The findings of this study provide important empirical evidences on supply chain performances by each supply chain strategies with results supporting existing conceptual literatures.

A new ML-based approach to enhance student engagement in online environment.

The educational research is increasingly emphasizing the potential of student engagement and its impact on performance, retention and persistence. This construct has emerged as an important paradigm in the higher education field for many decades. However, evaluating and predicting the student’s engagement level in an online environment remains a challenge. The purpose of this study is to suggest an intelligent predictive system that predicts the student’s engagement level and then provides the students with feedback to enhance their motivation and dedication. Three categories of students are defined depending on their engagement level (Not Engaged, Passively Engaged, and Actively Engaged). We applied three different machine-learning algorithms, namely Decision Tree, Support Vector Machine and Artificial Neural Network, to students’ activities recorded in Learning Management System reports. The results demonstrate that machine learning algorithms could predict the student’s engagement level. In addition, according to the performance metrics of the different algorithms, the Artificial Neural Network has a greater accuracy rate (85%) compared to the Support Vector Machine (80%) and Decision Tree (75%) classification techniques. Based on these results, the intelligent predictive system sends feedback to the students and alerts the instructor once a student’s engagement level decreases. The instructor can identify the students’ difficulties during the course and motivate them through e-mail reminders, course messages, or scheduling an online meeting.

Activity Tracking Using Ear-Level Accelerometers.

Introduction: By means of adding more sensor technology, modern hearing aids (HAs) strive to become better, more personalized, and self-adaptive devices that can handle environmental changes and cope with the day-to-day fitness of the users. The latest HA technology available in the market already combines sound analysis with motion activity classification based on accelerometers to adjust settings. While there is a lot of research in activity tracking using accelerometers in sports applications and consumer electronics, there is not yet much in hearing research.Objective: This study investigates the feasibility of activity tracking with ear-level accelerometers and how it compares to waist-mounted accelerometers, which is a more common measurement location.Method: The activity classification methods in this study are based on supervised learning. The experimental set up consisted of 21 subjects, equipped with two XSens MTw Awinda at ear-level and one at waist-level, performing nine different activities.Results: The highest accuracy on our experimental data as obtained with the combination of Bagging and Classification tree techniques. The total accuracy over all activities and users was 84% (ear-level), 90% (waist-level), and 91% (ear-level + waist-level). Most prominently, the classes, namely, standing, jogging, laying (on one side), laying (face-down), and walking all have an accuracy of above 90%. Furthermore, estimated ear-level step-detection accuracy was 95% in walking and 90% in jogging.Conclusion: It is demonstrated that several activities can be classified, using ear-level accelerometers, with an accuracy that is on par with waist-level. It is indicated that step-detection accuracy is comparable to a high-performance wrist device. These findings are encouraging for the development of activity applications in hearing healthcare.

How Does the Facilitation Effort of Clinical Educators Interact With Aspects of Organizational Context to Affect Research Use in Long-term Care? Evidence From CHAID Analysis.

Organizational context influences the effect of facilitation efforts on research use in care settings. The interactions of these factors are complex. Therefore, the use of traditional statistical methods to examine their interrelationships is often impractical. Big Data analytics can automatically detect patterns within the data. We applied the chi-squared automatic interaction detection (CHAID) algorithm and classification tree technique to explore the dynamic and interdependent relationships between the implementation science concepts-context, facilitation, and research use. Observational, cross-sectional study based on survey data collected from a representative sample of nursing homes in western Canada. We assessed three major constructs: (a) Conceptual research utilization (CRU) using the CRU scale; (b) facilitation of research use measured by the frequency of contacts between the frontline staff and a clinical educator, or person who brings new ideas to the care unit; and (c) organizational context at the unit level using the Alberta Context Tool (ACT). CHAID analysis was performed to detect the interactions between facilitation and context variables. Results were illustrated in a classification tree to provide a straightforward visualization. Data from 312 care units in three provinces were included in the final analysis. Results indicate significant multiway interactions between facilitation and various aspects of the organizational context, including leadership, culture, evaluation, structural resources, and organizational slack (staffing). Findings suggested the preconditions of the care settings where research use can be maximized. CHAID analysis helped transform data into usable knowledge. Our findings provide insight into the dynamic relationships of facilitators' efforts and organizational context, and how these factors' interplay and their interdependence together may influence research use. Knowledge of the combined effects of facilitators' efforts and various aspects of organizational context on research use can contribute to effective strategies to narrow the evidence-practice gap in care settings.

Predicting Children’s Nutritional Status Using QUEST Method

Concerning WHO’s standard of nutritional status, the Indonesian Ministry of Health office’s result of monitoring nutritional status in 2017 discovered a nutritional problem among children in Padang, West Sumatra. This study was aimed at determining factors that affect the nutritional status and their malnutrition level for the children under five years in Padang, West Sumatera.Data analysis was conducted using a classification tree technique, namely the QUEST method. By the research, the QUEST method analysis showed some underweight and severely underweight cases were found to be common among children under three years of age. On the one hand, severely underweight and underweight cases were commonly identified among female children. On the other, the prevalence of overweight case is common among the male. At last, the analysis of under-three male children indicated that overweight cases were found in children who did not receive exclusive breastfeeding. Keywords: nutritional status; classifications; QUEST method.

Spread of COVID-19 in Zambia: An assessment of environmental and socioeconomic factors using a classification tree approach

The global pandemic emergent from SARS-COV-2 (COVID-19) has continued to cause both health and socio-economic challenges worldwide. However, there is limited information on the factors affecting the dynamics of COVID-19, especially in developing countries, including African countries. In this study, we have focused on understanding the association of COVID-19 cases with environmental and socioeconomic factors in Zambia - a sub-Saharan African country. We used Zambia's district-level COVID-19 data, covering 18 March 2020 (i.e., from first reported cases) to 17 July 2020. Geospatial approaches were used to organize, extract and establish the dataset, while a classification tree (CT) technique was employed to analyze the factors associated with the COVID-19 cases. The analyses were conducted in two stages: (1) the binary analysis of occurrences of COVID-19 (i.e., COVID-19 or No COVID-19), and (2) a risk level analysis which grouped the number of cases into four risk levels (high, moderate, low and very low). The results showed that the distribution of COVID-19 cases in Zambia was significantly influenced by the socioeconomic factors compared to environmental factors. More specifically, the binary model showed that distance to the airport, population density and distance to the town centres were the most combination influential factors, while the risk level analysis indicated that areas with high rates of human immuno-deficient virus (HIV) infection had relatively high chances of having many COVID-19 cases compared to areas with low HIV rates. The districts that are far from major urban establishments and that experience higher temperatures have lower chances of having COVID-19 cases. This study makes two major contributions towards the understanding of COVID-19 dynamics: (1) the methodology presented here can be effectively applied in other areas to understand the association of environmental and socioeconomic factors with COVID-19 cases, and (2), the findings from this study present the empirical evidence of the relationship between COVID-19 cases and their associated environmental and socioeconomic factors. Further studies are needed to understand the relationship of this disease and the associated factors in different cultural settings, seasons and age groups, especially as the COVID-19 cases increase and spread in many countries.

MULTIPLE REGRESSION MODELING THE CHEMICAL COMPOSITION EFFECTS ON THE TENSILE STRENGTH OF Ni-BASED SUPERALLOYS

Based on the previously obtained results of the Data Mining - based Classification and Regression Trees (C&RT) technique application to industrial Ni-based superalloys, multiple regression models were developed to describe quantitatively the chemical composition effects on the alloys tensile strength. The regression models obtained were verified for adequacy in three ways, contrary to the conventional approach requiring only several inadequacy variance values calculations. Namely, the following verification methods were applied: traditional R2 coefficient calculations; comparisons of computer and experimental frequency distributions of the quality indexes; Monte-Carlo simulated and experimental data based scatter plots comparing. It was emphasized that, according to the general statistic principle, close coincidence of the Monte-Carlo simulated and real frequency distributions of any measured characteristic values means full formal (statistical) identity of the real and simulated objects. Hence, in addition, coincidence of the corresponding scatter plots statistical parameters should be considered as a sign of the physical identity of real and modeled regression dependencies. All the applied verification methods show the excellent results that leads to conclusion about high workability of the obtained models in practical and theoretical applications. An additional general advantage of the models was noted: their physically interpretable structure, that provides the opportunity to determine and analyze the effects of individual chemical elements and their interactions under real industrial conditions. Most of the obtained results and the formulated conclusions, especially related to the chemical elements individual effects are in good accordance with the known experimental and theoretic data for polycrystalline Ni-based superalloys. Meantime, some unexpected features, particularly, for the interactions of Nb, Ti, Cu with each other and grain boundaries were revealed, showing necessity of further in dept investigations.

Embedding Quality into Software Product Line Variability Artifacts

The success of any software product line development project is closely tied to its domain variability management. Whereas a lot of effort has been put into functional variability management by the SPL community, non-functional variability is considered implicit. The result has been dissatisfaction among clients due to resultant poor quality systems. This work presents an integrated requirement specification template for quality and functional requirements at software product line variation points. The implementation of this approach at the analytical description phase increases the visibility of quality requirements obliging developers to implement them. The approach proposes the use of decision tree classification techniques to support the weaving of functional quality attributes at respective variation points. This work, therefore, promotes software product line variability management objectives by proposing new functional quality artifacts during requirements specification phase. The approach is illustrated with an exemplar mobile phone family data storage requirements case study.