Rapid Miner Tool Research Articles

Performance Comparasion of Adaboost and PSO Algorithms for Cervical Cancer Classification Using KNN Algorithm

Abstract. Cervical cancer impacts the female reproductive organs and stands as the second most common cancer among women worldwide. The World Health Organization (WHO) reports that annually, approximately 500,000 women are diagnosed with cervical cancer, and about 300,000 die from it. Many of these deaths result from insufficient early detection and preventive measures. There are four primary screening techniques for detecting cervical cancer cells: Hinselmann, Schiller, Cytology, and Biopsy. In this study, patient health history data is analyzed using the KNN algorithm, which is further optimized with Adaboost and PSO techniques. These optimization strategies are evaluated to identify the most precise model for detecting patterns in cervical cancer patients and predicting their screening outcomes. This study employs the RapidMiner tool. Findings reveal that the KNN algorithm effectively performs multilabel classification, and when optimized with PSO, there is a slight improvement in accuracy.Purpose: The aim of this research is to assess the performance of the K-Nearest Neighbor (KNN) algorithm in multilabel classification of cervical cancer and to optimize it using Adaboost and Particle Swarm Optimization (PSO) techniques. This research is significant as it offers a potentially more accurate diagnostic method for detecting cervical cancer using medical records.Methods/Study design/approach: The Cervical Cancer Risk Classification dataset from Kaggle was used in this study. Data preprocessing was conducted before applying the KNN algorithm. The KNN algorithm's performance was evaluated using a 10-fold cross-validation method, and results were measured using the Confusion Matrix. Additionally, the KNN algorithm was optimized using Adaboost and PSO to assess improvements in accuracy.Result/Findings: Experimental results indicated that the KNN algorithm achieved optimal accuracy with k=5, reaching 95.81%, 91.26%, 94.64%, and 93.01% for Hinselmann, Schiller, Cytology, and Biopsy targets, respectively. Adaboost did not significantly improve accuracy, while PSO slightly enhanced the Hinselmann target accuracy from 95.81% to 95.92%. The average training time for this experiment was around two minutes. These results demonstrate the effectiveness of the KNN algorithm in conducting multilabel classification for cervical cancer diagnosis.Novelty/Originality/Value: This research demonstrates that optimizing the KNN algorithm with PSO can enhance accuracy, though not significantly. This suggests potential for further development to improve cervical cancer diagnostic accuracy. Testing the model with the latest data and optimizing parameters may lead to better models and useful tools for early cervical cancer diagnosis.

Predicting corrosion current density in magnesium alloy battery anodes: machine learning approach using rapid miner

ABSTRACT Magnesium (Mg) alloys are increasingly recognised for their potential as anode materials in batteries due to their high specific capacity and cost-effectiveness. However, their susceptibility to corrosion poses a significant challenge to practical applications. This study explores the effectiveness of various machine learning models – local polynomial regression, Decision Trees, Random Forest, Gradient Boost, and Extreme Gradient Boost – in predicting the corrosion behaviour of Mg alloy battery anodes in a 3.5 wt.% NaCl electrolyte. Experimental data on the corrosion current density of Mg alloys are collected and analysed using the RapidMiner tool for model training and evaluation. The findings highlight that the Extreme Gradient Boost (XGBoost) model excels among other machine learning techniques in accurately forecasting the corrosion rates of Mg alloys. XGBoost achieves an exceptional R2 value of 0.9931 in Tafel plots, indicating a robust fit to the data as evidenced by scatter plots. Furthermore, the XGBoost model demonstrates strong positive monotonic and ordinal relationships between actual and predicted corrosion current densities, as indicated by a Spearman correlation coefficient of 1 and a Kendall tau coefficient of 0.99. These outcomes underscore XGBoost's potential as a powerful tool for enhancing the understanding and management of corrosion in Mg alloy battery applications.

Klasifikasi Data Mining dalam Memprediksi Kinerja Karyawan dengan Metode Algoritma C4.5 pada Toko Merpati Simpang

It has been demonstrated that using computers greatly improves our ability to perform our duties. Information services are vital because, while employee performance may still be predicted manually, the process takes a long time. Data mining technologies, on the other hand, make it easier to anticipate employee success for loyal employees. Employee performance evaluation criteria are necessary in order to increase the accuracy of the assessment results, as Toko Merpati Simpang's employee performance assessments cannot be conducted carelessly. Employee performance has to be analyzed and categorized because up until now, manual employee performance evaluations have only used subjective criteria. The C4.5 Algorithm data mining approach is used in this evaluation of employee performance. The degree of accuracy will be ascertained by comparing these two approaches. Positive and negative emotions are the two categories of sentiment. The aim of this study is to ascertain the degree of accuracy of the comparison between the two tested techniques and to offer information on the quality of one of Toko Merpati Simpang's employee performance assessments using visitor sentiment. The test results will be evaluated using the Rapidminer tool to demonstrate the degree of accuracy for both testing approaches. Keywords: ,

Predicting Student Graduation Grades Using the C4.5 Algorithm: An Implementation Study

SMP KEMALA BHAYANGKARI as one of the private schools, students follow the learning process and are carried out mid-semester exams, semester final exams in completing junior high school level education. test scores as one of the requirements for grade promotion or for graduation mark the end of junior high school level, and the value of knowledge is one of the keys to a person's ability to complete education. predicting student graduation with the C4.5 algorithm method In the application of the C4.5 algorithm using Rapidminer tools, after manual calculation, the results will be tested using Rapidminer tools. the results of manual calculation of the C4.5 algorithm and the results of calculations using rapidminer tools will produce an Accuracy Level resulting from this calculation of 92.22% for graduation grades at SMP Kemala Bhayangkari Kotabumi.

A novel approach for the effective prediction of cardiovascular disease using applied artificial intelligence techniques.

The objective of this research is to develop an effective cardiovascular disease prediction framework using machine learning techniques and to achieve high accuracy for the prediction of cardiovascular disease. In this paper, we have utilized machine learning algorithms to predict cardiovascular disease on the basis of symptoms such as chest pain, age and blood pressure. This study incorporated five distinct datasets: Heart UCI, Stroke, Heart Statlog, Framingham and Coronary Heart dataset obtained from online sources. For the implementation of the framework, RapidMiner tool was used. The three-step approach includes pre-processing of the dataset, applying feature selection method on pre-processed dataset and then applying classification methods for prediction of results. We addressed missing values by replacing them with mean, and class imbalance was handled using sample bootstrapping. Various machine learning classifiers were applied out of which random forest with AdaBoost dataset using 10-fold cross-validation provided the high accuracy. The proposed model provides the highest accuracy of 99.48% on Heart Statlog, 93.90% on Heart UCI, 96.25% on Stroke dataset, 86% on Framingham dataset and 78.36% on Coronary heart disease dataset, respectively. In conclusion, the results of the study have shown remarkable potential of the proposed framework. By handling imbalance and missing values, a significantly accurate framework has been established that could effectively contribute to the prediction of cardiovascular disease at early stages.

Application of the K-Nearest Neighbor Method to Predict Demand for Goods from Customers at PT Sinergi Prima Enjineering

PT Sinergi Prima Enjineering, which is engaged in services, has been trusted as a contractor in several companies facing challenges in handling the large number of requests for goods and stock inventory management. This research aims to improve the prediction of demand for goods and inventory management using the calculation of the K-Nearest Neighbor (KNN) method and RapidMiner tools. With the comparison of calculations between KNN and RapidMiner using ten test data, the results are appropriate where the categorical grouping is often ordered totaling five data, moderately ordered totaling two data and rarely ordered totaling three data. The test results show that K = 3 produces a prediction accuracy of 91.98%. These results show that K-Nearest Neighbor can accurately anticipate future stock inventory and items that will be ordered by customers and it is hoped that the company can improve customer satisfaction and overall operational performance.

Airline Attitudes: Mining Data for Service Sentiment Insights

Published user reviews and opinions underlie sentiment analysis, thereby enabling extraction of essential data about service providers and indicating their market status. This study investigates smell sentimental grouping precision of standard information mining systems against six Indian carriers related social media microblog datasets. The sentiment analysis with classifiers, namely Bayes Net and SVM, achieved good results in their accuracies. The analysis centers on how well passenger sentiments can be separated out by the six airlines. On WEKA tool, Bayes Net gained the highest accuracy whereas on Rapid Miner tool, SVM yields better classification than all others. A look at passengers experience with the airlines According to reviews, Go Air has been one of the top-knotch in terms of service received by customers and turns out to be highest rated airline based on passenger feedback.

DETERMINATION OF COFFEE FRUIT MATURITY LEVEL USING IMAGE HISTOGRAM AND K-NEAREST NEIGHBOR

Coffee has a very important role in the Indonesian economy, as one of the country's foreign exchange contributors in the plantation sector. Therefore, coffee processing is very important in determining the quality of coffee. The procedure for choosing and evaluating the coffee fruit's physical quality is one of the most crucial steps. The step of determining the maturity level of coffee fruit is carried out using the image histogram and K-Nearest Neighbor (KNN) method. This research uses the KNN algorithm with classification stages that will show the level of accuracy value according to the value of k = 5 used when processing the classification of coffee fruit image data. In order to complete this step, the features of the coffee fruit are identified using its color. The qualities of quality coffee fruit, which is flawlessly red in color. Twenty images total—ten of which are of ripe coffee fruit and ten of which are of raw coffee fruit—were used in this study. The test results were carried out using rapidminer tools using 40% training data and 60% testing data from the total data set. Based on the test results, it gives an accuracy value of 100%, meaning that the data set can be used in the next stage as valid data to be used.

Classification of Diabetes Mellitus Sufferers Eating Patterns Using K-Nearest Neighbors, Naïve Bayes and Decission Tree

The study investigates three classification algorithms, namely K-Nearest Neighbor (K-NN), Naïve Bayes, and Decision Tree, for the classification of Diabetes Mellitus using a dataset from Kaggle. K-NN relies on distance calculations between test and training data, using the Euclidean distance formula. The choice of k, representing the nearest neighbor, significantly influences K-NN's effectiveness. Naïve Bayes, a probabilistic method, predicts class probabilities based on past events, and it employs the Gaussian distribution method for continuous data. Decision Trees, form prediction models with easily implementable rules. Data collection involves obtaining a Diabetes Mellitus dataset with eight attributes. Data preprocessing includes cleaning and normalization to minimize inconsistencies and incomplete data. The classification algorithms are applied using the Rapidminer tool, and the results are compared for accuracy. Naïve Bayes yields 77.34% accuracy, K-NN performance depends on the chosen k value, and Decision Trees generate rules for classification. The study provides insights into the strengths and weaknesses of each algorithm for diabetes classification

Implementation of Data Mining to Predict Student Study Period with Decision Tree Algorithm (C4.5)

Graduating on time is what every student wants to accomplish in college. Students of Prof. Dr. Hamka Muhammadiyah University are one of those who have this dream. Based on 2020 graduates data from the Tracer Study, 60% said the university had a high enough impact on improving competence. This data indicates that university needs to evaluate improvement of academic quality. Often, students have difficulty finding information about important factors that support achieving timely graduation. A prediction analysis is needed to provide information about the student's graduation study period. For this analysis, data mining is implemented using the classification function of the decision tree (C4.5) algorithm with RapidMiner tools. The methodology for implementing data mining follows the stages of Knowledge Discovery In Database (KDD), beginning with data collection, preprocessing, transformation, data mining, and evaluation. The research findings consist of visualization and decision tree rules that reveal GPA as the most influential factor in determining a student's study period.There is other information, namely, students graduated on time (less than equal to 4 years) amounted to 170 or 54.5% and students did not graduate on time (more than 4 years) amounted to 142 or 45.6%. Testing the performance of decision tree (C4.5) utilizing confusion matrix through RapidMiner tools, resulted in accuracy reaching 83.87%, with precision of 87.50% and recall of 91.18%. Provides evidence that the decision tree algorithm (C4.5) has optimal performance to provide valuable information about predicting student graduation in order to increase student enrollment with the right study period.

Penerapan Data Mining Menggunakan Metode Algoritma Naive Bayes Classifier untuk Mendukung Strategi Promosi

The purpose of this study is to make it easier to find out the prediction results with the classification method from PPDB data for 2021-2022 to support promotion strategies at the Skye Foundation with the Naive Bayes Classifier Algorithm and to find out the level of accuracy in classification in determining promotion strategies. In this study the authors used 210 data as training data and 53 data as testing data for manual testing in excel and for testing using RapidMiner tools. Based on the results of the analysis in the research conducted, the results of predicting registration and non-registration quickly and accurately from tests carried out by comparing training data with data testing using RapidMiner tools obtained an accuracy rate of 92.45%.

Imbalanced Multiclass Data Classification Using Combined Data Sampling and Deep Learning Method

Multiclass Classification for finding pattern refers to classifying each data to part of the classes or labels that are generally more than two. The foremost challenge in classifying is with imbalanced data that have large portion data known to be majority class, and small portion known as minority class that leads to poor understanding of samples and less accurate results. The existing works discussed Random Upsampling, Random Downsampling, SMOTE methods individually with FeedForward Neural Network and found Random Oversampling gave better results .However, it generates more duplicate data and has less accuracy. Hence , this research work put forward Combined Random Over-Under Sampling approach that was preprocessed prior with Replacing Missing value with mean, Feature selection, Noise Filtering. Meanwhile this work extends the existing FeedForward Neural Network to Deep Learning . The proposed work is implemented in Rapidminer tool, assessed with appropriate evaluation measures for training and testing data individually.

Analisis Tingkat Kepuasan Karyawan Terhadap Fasilitas Hunian Perkebunan Ptpn IV Unit Marjandi Dengan Menggunakan Algoritma C4.5

Residential is a place to live or everything related to a place to live and in a narrow sense can be interpreted as a residential area or residential building. In connection with the quality of residential facilities, employee satisfaction is an important aspect in improving the quality of the provided residential facilities. If the satisfaction of the residential facilities provided to employees is not met, the employees will feel uncomfortable and will sue the plantation of PTPN IV Marjandi Unit. Analyze the level of employee satisfaction with the quality of the residential provided by the plantation to its employees is The purpose of this study. One of the efforts to find out the satisfaction of employees of residential facilities is to use the C4.5 algorithm classification, where Dataset is obtained through the Questionnaire technique that will be given to employees at PTPN IV Marjandi Unit. As for the satisfaction of employees of residential facilities in terms of aspects (1) water facilities, (2) electricity facilities, (3) kitchen conditions, (4) bedroom conditions, (5) toilet conditions in dwellings inhabited by PTPN IV Unit plantation employees Marjandi. Data processing is assisted using Microsoft Excel software and to validate study data using a rapidminer tool. It is hoped that this research can provide solutions or become a reference for plantations to improve the quality of residential facilities provided to PTPN IV Marjandi Unit employees..

KOMPARASI ALGORITMA DECISION TREE, NAIVE BAYES DAN K-NEAREST NEIGHBOR UNTUK MENENTUKAN KUALITAS UDARA DI PROVINSI DKI JAKARTA

Air quality in DKI Jakarta Province refers to the state and cleanliness of the air in the area at a given time. The type and concentration of air pollutants are among the indicators used to assess air quality. DKI Jakarta's air pollution is severe, causing respiratory irritation, respiratory illnesses, and long-term health issues such as cardiovascular disease and lung cancer. Air pollution can also harm the environment by limiting visibility and harming ecosystems. The problem with the research is that no appropriate and relevant features for predicting air quality were used. The goal of this study is to identify and compare algorithms with the highest accuracy between decision trees. In determining air quality, naive Bayes and k-nearest neighbor are used. According to the findings of the K-5fold evaluation process performed with the RapidMiner tool, the accuracy of the Decision Tree algorithm was 95.89%, the accuracy of the Nave Bayes algorithm was 93.15%, and the accuracy of the K-NN algorithm was 91.78%. Based on these findings, the decision tree method has the greatest or best accuracy when compared to the Nave Bayes and K-NN algorithms.

SENTIMEN ANALISIS PENGGUNA TWITTER TERHADAP SEA GAMES 2023 DENGAN METODE NAIVE BAYES

The development of computers and telecommunications has made people often express their opinions through social media Twitter. As recently, after the 2023 SEA Games were held in Cambodia, various kinds of tweets appeared from various sports. From this opinion, it can be seen that the country's success in holding the Sea Games by conducting a sentiment analysis. So this research was conducted to find out how the public opinion regarding the holding of the 2023 SEA Games using the Naive Bayes Classifier algorithm. The data used was 1888 data obtained by crawling Twitter data using the keywords SEA Games and RapidMiner tools. The division of training data and testing data is carried out into two comparisons, namely 50:50 and 60:40. The best accuracy results are obtained with a data comparison of 50:50, which is 70.15%, while a data comparison of 60:40 has a smaller result, namely 70.07%.

Review of Machine Learning Techniques for Power Quality Performance Evaluation in Grid-Connected Systems

In the current energy usage scenario, the demands on energy load and the tariffs on the usage of electricity are two main areas that require a lot of attention. Energy forecasting is an ideal solution that would help us to better understand future needs and formulate solutions accordingly. Some important factors to investigate are the quantity and quality of smart grids as they are significantly influenced by the transportation, storage, and load management of energy. This research work is a review of various machine learning algorithms for energy grid applications like energy consumption, production, energy management, design, vehicle-to-grid transfers, and demand response. Ranking is performed with the help of key parameters and is evaluated using the Rapid Miner tool. The proposed manuscript uses various machine learning techniques for the evaluation of power quality performance to validate an efficient algorithm ranking in a grid-connected system for energy management applications. The use of renewable energy resources in grid-connected systems is more common in modern power systems. Universally, the energy usage sector (commercial and non-commercial) is undergoing an increase in demand for energy utilization that has substantial economic and ecological consequences. To overcome these issues, an integrated, ecofriendly, and smart system that meets the high energy demands is implemented in various buildings and other grid-connected applications. Among various machine learning techniques, an evaluation of seven algorithms—Naïve Bayes, artificial neural networks, linear regression, support vector machine, Q-learning, Gaussian mixture model, and principle component analysis—was conducted to determine which algorithm is the most effective in predicting energy balance. Among these algorithms, the decision tree, linear regression, and neural networks had more accurate results than the other algorithms used. As a result of this research, a proposal for energy forecast, energy balance, and management was compiled. A comparative statement of various algorithms concludes with results which suit energy management applications with high accuracy and low error rates.

Sentiment Analysis and Classification on Amazon Products using Improved Support Vector Machine for Multiclass Classification

There is a huge increase in number of peoples who have been accessing many social networking sites especially user post or reviews for a specific product, company, brand, individual, forums and movies etc. These reviews are helpful in judging customer perception on certain thing. The development of algorithms that could automate the categorization of distinct comments based on feedback from consumers became an analyst project, and this automated classification process is known as sentiment analysis. This research main goal is to analyze Amazon product reviews using an approach to Machine Learning (ML) built around TF-IDF and then employ the Support Vector Machine (SVM) algorithm to categorize the sentiment scores and sentences. SVM can handle binomial classification but the customer reviews is mostly classified into positive, negative and neutral and in some applications, it is fine grained into star ratings such 1-5 or sometimes 1-10. Also, in some applications features or attributes are high in number in which some are irrelevant. Hence, this work applies feature subset algorithm and improves the existing SVM to handle multiclass classification. The Sentiment analysis, Rapidminer tool is considered for classification and the results are visualized, assessed with suitable classification metrics.

Pengelompokan Untuk Penjualan Obat Dengan Menggunakan Algoritma K-Means

Drug grouping is an arrangement that adjusts to the flow of placement or drug layout is more suitable for standard processes. Utilization of existing data through the clustering method approach can be applied to analyze in grouping drug data on data availability and inventory in warehouses so as to provide knowledge and information. The clustering method is processed using the K-Means algorithm where the results also show a new knowledge, namely the grouping of drug data based on 2 clusters. Cluster 1 is a high need category with availability of 71 out of 100 availability categories based on the amount of drug data tested, then cluster 2 is a drug category with moderate or low availability, namely 29 out of 100 availability categories based on the number of drug data tested. Tests using Rapid Miner tools can also produce similar insights, namely each cluster has cluster group members according to manual calculations such as Cluster_0 in Rapid Miner has 72 cluster members representing the Medium cluster, Cluster_1 has 72 cluster group members as high cluster representations, and Cluster_2 has 3 cluster members corresponding to low representation.

Klasterisasi Stok Produk Retail Untuk Menetukan Pergerakan Kebutuhan Konsumen Dengan Algoritma K-Means

− Retail product clustering is a product arrangement that is adjusted to the flow of placement or this layout is more suitable for product placement according to standards. Utilization of existing data through the clustering method approach can be applied in analyzing product grouping of data on availability and inventory of goods in warehouses so that it can provide knowledge and information. The clustering method is processed using the K-Means algorithm, where the results also show a new insight, namely grouping products based on 3 clusters. Cluster 1 is a product category with low availability or Low, namely 939 out of 1000 availability categories based on the number of products tested, then cluster 2 is a product category with medium or Medium availability, namely 51 out of 1000 availability categories based on the number of products tested, and finally cluster 3 is a product category with fairly high availability or High, namely 10 out of 100 availability categories based on the number of products tested. Tests using Rapid Miner tools can also produce similar insights, namely that each cluster has cluster group members according to manual calculations such as Cluster_0 in Rapid Miner has 51 cluster members representing the Medium cluster, Cluster_1 has 939 cluster group members representing the Low cluster, and Cluster_2 has 10 cluster members corresponding to the cluster representation High.

ANALISIS DATA PENJUALAN PADA SUPERMARKET XYZ MENGGUNAKAN METODE MARKET BASKET

The economic growth in Indonesia has increased significantly, positioning Indonesia as the 4th largest economy after China, India, and Malaysia. This is also accompanied by national retail sales reaching a total of Rp6,044 trillion. One example of a retail business in Indonesia is Supermarket XYZ, which operates in Indonesia and sells essential items such as sugar, eggs, butter, and others. This research discusses an issue that exists in Indonesia, for example, Supermarket XYZ. The supermarket faces a problem with its employees struggling to determine the appropriate product layout based on consumer purchasing behavior. The market basket analysis method is used in this research to assist Supermarket XYZ's employees in determining the product layout. Transaction data is collected through internet searches, specifically on the Kaggle website. The data is then processed using the market basket analysis method, both manually and with the aid of the RapidMiner tool, to ensure the validity of the calculations. Once obtained, the results are implemented into a program to provide predictions for product placement layout. The research findings show that out of the 10 most purchased products by consumers, the top four products are chocolate, apples, bread, and ice cream. Furthermore, the suitable product combinations for layout placement yield 3 combinations: chocolate placed near apples with a support rate of 23%, chocolate placed near bread with a support rate of 22%, and chocolate placed near ice cream with a support rate of 21%.