Data Mining Classification Research Articles

ANALISIS SENTIMEN ULASAN APLIKASI LINKEDIN DALAM GOOGLE PLAY STORE DENGAN MODEL NAÏVE BAYES

In the era of digitalization, mobile applications have become a basic necessity in individual lives, especially in the work context. LinkedIn, as a mobile application that focuses on providing job and recruitment information. With the significant growth of LinkedIn users in 2023 with 930 registered users, many users will provide reviews on the Google Play Store regarding their experiences. The aim of this research is to collect information on reviews on the Google Play Store regarding the LinkedIn application service through sentiment analysis using the data mining classification method with the Naïve Bayes model. Through previous research, this method is considered better than other classification methods. After applying this model through testing on test data, test results were obtained which showed that the majority of LinkedIn application reviews had negative sentiment with an accuracy value of 84% which was presented in bar charts and wordcloud.

Development of Prediction Model for Heart Disease by Combining Clustering and Classification Techniques

The concerning trends in deaths related to heart disease some measures need to be in place to ensure early treatment and diagnosis of the disease. Therefore, one of the way can be done is by leveraging the abundance of medical data available. Advancement in technology today has improved the availability and accessibility huge amounts of valuable data and it only makes sense for us to explore the opportunities that lie in the data that could possibly save lives and reduce costs. Thu, this study aims to do that with the help of classification and clustering data mining techniques to predict heart disease based on some key indicators of the disease. Studies show that applying classifiers on clustered data can improve the performance of algorithms. Hence, this method will be explored in this study using the Naïve Bayes, Decision Tree and Random Forest classifiers together with both K-Means Clustering and Density-Based Clustering on the data analysis using tool WEKA. The performance of the each model will be measured and compared against each other using accuracy, precision, recall, specificity, AUC and model build time. Thus, this paper will focused on development of prediction model for heart disease by combining clustering and classification techniques in detail.

Anti-Money Laundering by Group-Aware Deep Graph Learning

Anti-money laundering (AML) is a classical data mining problem in finance applications. As well known, money laundering (ML) is critical to the effective operation of transnational and organized crime, which affects a country's economy, government, and social wellbeings. Financial services organizations facilitate the movement of money and have been enlisted by governments to assist with the detection and prevention of money laundering, which is a key tool in the fight to reduce crime and create sustainable economic development. In the application of AML, user identity and financial behavior data are widely used to detect laundering transactions. In recent years, an increasing number of money laundering activities have been conducted by organized criminal gangs while most existing works still treat the actions of each account as independent identity behavior without considering the group-level conspired interactions. Therefore, in this paper, we propose a group-aware deep graph learning-based approach for organized money-laundering detection. In particular, we design a community-centric encoder to represent the nodes and attributes in user transaction graphs and derive the adjacent gang behaviors. Then, we devise a scheme of local enhancement to accommodate nodes with similar transaction features, which are aggregated into gangs for downstream detection. Extensive experiments on the real-world dataset from one of the largest bank card alliances worldwide show that our proposed method outperforms state-of-the-art methods in both offline and online modes, showing the effectiveness of money laundering detection with group-aware deep graph learning.

Perbandingan Algoritma Klasifikasi Data Mining Pada Prediksi Penyakit Diabetes

Diabetes is a chronic disease that attacks humans. One of the causes of diabetes in humans is that sugar intake is too high which the body cannot balance due to absorption or activities carried out. Diabetes is often considered a common disease among people, but the impacts caused by this disease are very detrimental to humans. Based on this, it is necessary for everyone to know whether they suffer from diabetes or not. Therefore, this problem must be resolved appropriately, where it is necessary to predict whether someone will have diabetes or not. The prediction process is carried out to determine whether someone has diabetes or not by knowing the patterns or possible symptoms that cause someone to suffer from diabetes. In this research, the pattern formation process is based on data stored in the past collected in a dataset. A dataset is a collection of past data that occurred in fact and was then collected over a certain period of time on a large scale. Data mining is a method used to process data based on collections of past data, whether in datasets or others. In data mining, the data processing process is carried out using various techniques, one of which is the solution technique in data mining is classification. In this research, the Naïve Bayes algorithm, the K-Nearest Neighbor (K-NN) algorithm and the C4.5 algorithm will be used. In the data mining classification process, there are 3 (three) algorithms used, namely Naïve Bayes, K-Nearest Neighbor and C4.5. From the results of the tests that have been carried out, the accuracy performance results for the Naïve Bayes algorithm are 75%, accuracy for the K-Nearest Neighbor algorithm by 80.60% and the C4.5 algorithm by 91.80%. In this case, it indicates that the C4.5 algorithm has better performance compared to other algorithms. Therefore, the pattern results produced by the C4.5 algorithm are used to make predictions about diabetes.

APTM: Structurally Informative Network Representation Learning

Network representation learning algorithms provide a method to map complex network data into low-dimensional real vectors, aiming to capture and preserve structural information within the network. In recent years, these algorithms have found widespread applications in tasks such as link prediction and node classification in graph data mining. In this work, we propose a novel algorithm based on an adaptive transfer probability matrix. We use a deep neural network, comprising an autoencoder, to encode and reduce the dimensionality of the generated matrix, thereby encoding the intricate structural information of the network into low-dimensional real vectors. We evaluate the algorithm's performance through node classification, and in comparison with mainstream network representation learning algorithms, our proposed algorithm demonstrates favorable results. It outperforms baseline models in terms of micro-F1 scores on three datasets: PPI, Citeseer, and Wiki.

A density-based oversampling approach for class imbalance and data overlap

In data mining classification, class imbalance is characterized that different classes have an obvious difference in the number of samples. Most classifiers typically assume a balanced class distribution or assign equal classification error costs to different classes. Therefore, directly using imbalanced class will worsen the classification performance. The oversampling algorithms can achieve the balance by synthesizing new samples, but the uncontrollable positions of the synthetic samples may aggravate the data overlap and further deteriorate the classification performance. To tackle this challenge, an improved synthetic minority oversampling technique based on kernel density estimation and neighbor density selection (KDENDS_SMOTE) is proposed in this paper. First, each sample is mapped into a high-dimensional space to avoid the choice of the window width and to overcome the nonlinear separable limitation. Kernel density estimation is then used to derive the density ratio, which serves as a measure of the degree of data overlap. Subsequently, the stability degree of the density ratio is calculated using neighbor information, and a scoring mechanism combining the density ratio and its stability degree is proposed to assess the fitness of selected samples. Furthermore, the neighbor density selection based on the above scoring mechanism can guide SMOTE to generate new samples within a safe and stable region, away from areas with data overlap. Finally, compared with six advanced oversampling methods on fifteen real-world datasets, the KDENDS_SMOTE can effectively mitigate the data overlap and improve the classification performance.

Comparing Different Oversampling Methods in Predicting Multi-Class Educational Datasets Using Machine Learning Techniques

Abstract Predicting students’ academic performance is a critical research area, yet imbalanced educational datasets, characterized by unequal academic-level representation, present challenges for classifiers. While prior research has addressed the imbalance in binary-class datasets, this study focuses on multi-class datasets. A comparison of ten resampling methods (SMOTE, Adasyn, Distance SMOTE, BorderLineSMOTE, KmeansSMOTE, SVMSMOTE, LN SMOTE, MWSMOTE, Safe Level SMOTE, and SMOTETomek) is conducted alongside nine classification models: K-Nearest Neighbors (KNN), Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machine (SVM), Logistic Regression (LR), Extra Tree (ET), Random Forest (RT), Extreme Gradient Boosting (XGB), and Ada Boost (AdaB). Following a rigorous evaluation, including hyperparameter tuning and 10 fold cross-validations, KNN with SmoteTomek attains the highest accuracy of 83.7%, as demonstrated through an ablation study. These results emphasize SMOTETomek’s effectiveness in mitigating class imbalance in educational datasets and highlight KNN’s potential as an educational data mining classifier.

Research on a Decision Tree Classification Algorithm Based on Granular Matrices

The decision tree is one of the most important and representative classification algorithms in the field of machine learning, and it is an important technique for solving data mining classification tasks. In this paper, a decision tree classification algorithm based on granular matrices is proposed on the basis of granular computing theory. Firstly, the bit-multiplication and bit-sum operations of granular matrices are defined. The logical operations between granules are replaced by simple multiplication and addition operations, which reduces the operation time. Secondly, the similarity between granules is defined, the similarity metric matrix of the granular space is constructed, the classification actions are extracted from the similarity metric matrix, and the classification accuracy is defined by weighting the classification actions with the probability distribution of the granular space. Finally, the classification accuracy of the conditional attribute is used to select the splitting attributes of the decision tree as the nodes to form forks in the tree, and the similarity between granules is used to judge whether the data types in the sub-datasets are consistent to form the leaf nodes. The feasibility of the algorithm is demonstrated by means of case studies. The results of tests conducted on six UCI public datasets show that the algorithm has higher classification accuracy and better classification performance than the ID3 and C4.5.

Performance Improvement of K-Nearest Neighbor Algorithm in KIP Scholarship Recipient Selection

Abstract
 Law 12 of 2012 mandates that the government increase access to higher education for high achievers and underprivileged people. One of the efforts to realize this is by providing KIP Lectures. To ensure that beneficiaries are indeed eligible for KIP scholarships, it is necessary to classify scholarship recipients with data mining classification techniques correctly. The classification technique chosen is k-Nearest Neighbor (K-NN). K-NN is a classification method that relies heavily on the k parameter in carrying out classification. K-NN was applied to the KIP Scholarship applicant dataset at UIN Malang in 2022. The test scenario in this research is to compare the k-odd and k-even parameters to find the most optimal k value in K-NN. The highest accuracy value obtained by k-odd is 0.71 or 71% when k=9, and the highest for k-even is 0.67 or 67% when k=10. Using optimal k parameters is proven to improve k-NN performance. The K-NN algorithm with k-odd parameters, namely k=9, is the best method for classifying KIP scholarship recipients in this research. The results of this research can be considered in determining KIP scholarship recipients worthy of using K-NN.

Analysis of Employee Performance and factor Influencing It Using the C4.5 Algorithm Method

In an organization, individual performance greatly affects the performance of the organization, both government and private organizations in achieving predetermined goals. Good employee performance will have an impact on overall organizational performance, which in turn can be seen from organizational achievements. Some important factors that are taken into consideration by an employee performing well, among others; Employee salary, incentives/bonuses, work motivation, workload/quantity of work, reliability, and employability. This study uses a data mining classification approach using the c4.5 method in making a decision tree. The results of a trial of 25 datasets using the decision tree method were able to provide a conclusion that employees will be able to improve their performance if the organization provides a policy in the form of providing monthly incentives to employees as a bonus reward to those who are diligent in carrying out their work, such as having activeness and being able to work together.

Self-Organizing Computational System for Network Anomaly Exploration using Learning Algorithms

The forum in the nation for reporting information security flaws had 14,871 reports by the end of 2021, a 46.6% increase from 2020. The total of 5,567 high risk vulnerabilities, an increase of nearly 1,400 over the previous year. Evidently, both the total number of vulnerabilities found annually, and the total number of high-risk vulnerabilities are rising. In order for data mining technology to play a wider part in the predictive investigation of network security models, it is advised that its capability have to be improved. This paper combines the concepts of data mining (DM) with machine learning (ML), which introduces similar technologies from DM technology and security establishing collection channel, thereby finally introduces the computer network security maintenance process based on data mining in order to improve the application effect of DM in the predictive analysis of network security models. In this paper, a self-organizing neural network technique that detects denial of service (DOS) in complicated networks quickly, effectively, and precisely is introduced. It also analyses a number of frequently employed computer data mining methods, including association, clustering, classification, neural networks, regression, and web data mining. Finally, it introduces a computer data mining method based on the self-organizing (SO) algorithm. In comparison to conventional techniques, the SO algorithm-based computer data mining technology is also used in defensive detection tests against Dos attacks. A detection average accuracy rate of more than 98.56% and a detection average efficiency gain of more than 20% are demonstrated by experimental data to demonstrate that tests based on the Data Mining connected SO algorithm have superior defensive detection effects than standard algorithms.

Comparison Of K-Nearest Neighbor And CNN Classification Methods In Diabetic Data Sets

The number of diabetics worldwide is projected to increase by 204 million (48%), from 425 million in 2017 to 629 million in 2045. Indonesia ranks sixth out of ten countries with the most number of diabetics in the world or 10 million people. The majority of people with diabetes are between 20 and 64 years old, or 327 million people, compared to 123 million people between 65 and 99 years old. The incidence of diabetes increases by about 4.8% at the age of 55-64 years, and women (1.7%) suffer from diabetes more than men (1.4%). Therefore, the authors will create a program to determine the patient's diabetes. One approach is to use machine learning as a data mining classification technique. The author will do a classification comparison with the two methods, namely the KNN and CNN methods to provide the best results of the two methods for testing. So that the accuracy of the data from the diagnosis and photo images of the disease can be known to provide early treatment before the severity of the disease.

Data Optimization using PSO and K-Means Algorithm

Tourism is one of the industries that contribute considerably to the country's economy. Tourism helps the country's economy expand by providing and increasing jobs, living standards, and triggering the rise of other tourist-related production. The tourism industry will become a multinational industry and the primary driver of the global economy in the twenty-first century. Tourism has generated significant foreign exchange for a number of countries. Indonesia, the world's biggest archipelagic country with 17,508 islands, often known as the archipelago or maritime country, has recognized the importance of the tourist sector to the Indonesian economy because tourism growth consistently outpaces economic growth. The research's goal is to map the number of tourist visits. The mapping is in the form of clusters based on countries. The technology utilized is classification data mining with the K-Means method and Particle Swarm Optimization (PSO). The dataset came from the Central Bureau of Statistics, a government organization (abbreviated as BPS). The research outcomes in cluster mapping, with the cluster results compared to standard K-Means and K-Means + PSO. RapidMiner software is used during the analytical process. The calculation results in the form of clusters will be evaluated using the Davies-Bouldin Index (DBI) parameter. The cluster value (k) used is k = 2, 3, 4, 5. The findings show that the K-Means + PSO optimization has the minimum DBI value for k = 5. Meanwhile, the DBI value for k = 5 is 0.134.

Naive Bayes and Decision Tree Algorithms for BRI Life Sharia Insurance Product Classification

Law 12 of 2012 mandates that the government increase access to higher education for high achievers and underprivileged people. One of the efforts to realize this is by providing KIP Lectures. To ensure that beneficiaries are eligible for KIP scholarships, it is necessary to classify scholarship recipients correctly using data mining classification techniques. The classification technique chosen is k-Nearest Neighbor (K-NN). K-NN is a classification method that relies heavily on the k parameter in carrying out classification. K-NN was applied to the KIP Scholarship applicant dataset at UIN Malang in 2022. The test scenario in this research is to compare the k-odd and k-even parameters to find the most optimal k value in K-NN. The highest accuracy value obtained by k-odd is 0.71 or 71% when k=9, and the highest for k-even is 0.67 or 67% when k=10. Using optimal k parameters is proven to improve k-NN performance. The K-NN algorithm with k-odd parameters, namely k=9, is the best method for classifying KIP scholarship recipients in this research. The results of this research can be considered in determining KIP scholarship recipients worthy of using K-NN.

Risk Prediction of Thalassemia Using Data Mining Classifiers

Medical data mining is concerned with prediction knowledge, which is a useful method for extracting hidden patterns from given data for specific purposes. Thalassemia is one of the most common inherited blood hematological disorders, and this paper adopted data mining classification techniques to generate results with high performance and accuracy for risk prediction of thalassemia. The dataset for this purpose was collected from NIBD (National Institute of Blood Diseases), a well-known institute and hospital for blood diseases in Karachi, Pakistan. They provided 301 records of CBC test reports containing positive and negative statuses of diagnosis of thalassemia traits. There were many instances in the report, of which 6 were used for our research purpose, i.e. Gender, MCV, HGB, HCT, MCHC, and RDW. The dataset was divided into training and test data using the WEKA tool. Four algorithms of data mining classification, namely J48 Decision Tree, Naïve Bayesian Network, SMO algorithm, and Multilayer Perceptron Neural Network were adopted to train the model and classify the patient having traits of thalassemia from normal persons with the use of the WEKA tool. Results revealed that out of all four algorithms, Naïve Bayes provided results with the highest accuracy of 99%.

Improving The Performance of the K-Nearest Neighbor Algorithm in the Selection of KIP Scholarship Recipients

Law 12 of 2012 mandates that the government increase access to higher education for high achievers and underprivileged people. One of the efforts to realize this is by providing KIP Lectures. To ensure that beneficiaries are indeed eligible for KIP scholarships, it is necessary to classify scholarship recipients with data mining classification techniques correctly. The classification technique chosen is k-Nearest Neighbor (K-NN). K-NN is a classification method that relies heavily on the k parameter in carrying out classification. K-NN was applied to the KIP Scholarship applicant dataset at UIN Malang in 2022. The test scenario in this research is to compare the k-odd and k-even parameters to find the most optimal k value in K-NN. The highest accuracy value obtained by k-odd is 0.71 or 71% when k=9, and the highest for k-even is 0.67 or 67% when k=10. Using optimal k parameters is proven to improve k-NN performance. The K-NN algorithm with k-odd parameters, namely k=9, is the best method for classifying KIP scholarship recipients in this research. The results of this research can be considered in determining KIP scholarship recipients worthy of using K-NN.

IMPLEMENTASI DATA MINING DENGAN ALGORITMA NAIVE BAYES UNTUK KLASIFIKASI KELAYAKAN PENERIMA BANTUAN SEMBAKO

The percentage of the population receiving assistance was 5.19%, this figure increased compared to March 2022, while the line of aid recipients was recorded in 2022 of 783,730 people per capita per month. The number of poor households in Batu Aji District is Bukit Tempayang Village with 15,857 inhabitants, 300 poor households, 37,531 inhabitants in Buliang Village, 289 poor households, 28,693 The beneficiary of assistance is an inability to meet basic needs including food, clothing, education and housing. Thus it is necessary to carry out a strategy in dealing with the level of beneficiaries, namely by providing accurate and targeted data on beneficiaries. Naive Bayes is an algorithm that exists in data mining and is part of the data mining classification technique by using probability and statistical techniques to estimate or predict opportunities that will occur based on previous opportunities, namely there are two feasible and inappropriate classes. The results of the classification that will be carried out later will help in the processing of assistance to help make decisions regarding the classification of determining basic food recipients. And by testing calculations manually and using rapid minner software, you get an accuracy value of 80%.

Breast Cancer Recurrence Prediction Using Data Mining

Today, the Healthcare sector is generating a vast amount of meticulously detailed information on individuals and their medical issues. Medical data may be investigated for hidden patterns or connections using data mining techniques. The spread and return of the illness are the main causes of breast cancer fatalities. One of the most important areas of study in the field of medicine is the early diagnosis of breast cancer. Healthcare businesses can benefit from data mining by forecasting patient illnesses and behaviour trends. This is performed by analyzing data from many angles and by using data mining to find hidden patterns and associations between seemingly unrelated bits of information. Months after the original diagnosis and breast cancer therapy, a metastatic recurrence of the disease may happen. Only a few number of techniques have been researched in the state of the art due to the challenges associated with early breast cancer recurrence prediction during its medication. The aim of the research is to review the work done by different researchers in this field and to compare the various traditional data mining classifiers (Logistic Regression, Naïve Bayes, K-Nearest Neighbours, Decision Tree, Support Vector Machines, and Random Forest) applied to Breast Cancer Data Set from UCI Machine Learning Repository. Out of these, the best results were given by Random Forest Algorithm with 98.4% accuracy.

KLASIFIKASI KETEPATAN KELULUSAN SISWA PADA SMK YADIKA 9 BINTARA JAYA KOTA BEKASI MENGGUNAKAN ALGORITMA C4.5

Education is the learning of knowledge, skills, and habits of a group of people passed from one generation to the next through teaching, training, or research. SMK Yadika 9 Bintara Jaya Bekasi City is one of the private vocational schools, However, along the way, there are still problems that occur every year, namely the problem of the right number of student graduations and late. The number of students who graduate on time is an indicator of the effectiveness of a vocational school both public and private. SMK needs to detect the behavior of active students so that factors that cause students not to graduate on time can be seen. Data collection using techniques in the form of interviews and observations, in this study is related to systematic literature review to find predictive modeling using data mining classification methods. Data mining classification in determining the accuracy of student graduation with the CRISP-DM testing method. The results of this study resulted that the C4.5 algorithm is the most optimal method of performance in determining the graduation accuracy classification of students of SMK Yadika 9 Bintara Jaya Bekasi City.

Enhancing Software Project Monitoring with Multidimensional Data Repository Mining

Software project development and maintenance activities have been reported in various repositories. The data contained in these repositories have been widely used in various studies on specific problems, e.g., predicting bug appearance, allocating issues to developers, and identifying duplicated issues. Developed analysis schemes are usually based on simplified data models while issue report details are neglected. Confronting this problem requires a deep and wide-ranging exploration of software repository contents adapted to their specificities, which differs significantly from classical data mining. This paper is targeted at three aspects: the structural and semantic exploration of repositories, deriving characteristic features in value and time perspectives, and defining the space of project monitoring goals. The considerations presented demonstrate a holistic image of the project development process, which is useful in the assessment of its efficiency and identification of imperfections. The original analysis introduced in this work was verified using open source and some commercial software project repositories.