Knowledge Discovery In Databases Research Articles

The Effect of The Prediction of The K-Nearest Neighbor Algorithm on Surviving COVID-19 Patients in Indonesia

This study aims to measure the prediction of survival of covid-19 patients with the best algorithm based on RMSE(Root Mean Square Error). The Covid-19 pandemic has lasted from December 2019 until now and is full of uncertainty about when this pandemic will end, so this research was carried out. In this study, the knowledge discovery database method was used by extracting data sets from Covid-19 patients from March 2020 to March 2021 for each province in Indonesia (Dataset from Kawal Covid-19 SintaRistekbrin) to predict survival during this pandemic as measured by the best algorithms include k-NN (k-Nearest Neighbor), SVM (Support Vector Machine), and/or Deep Learning. The measurement results using cross-validation and the optimal number of folds is 3 in the form of RSME, showing that the k-NN algorithm is an algorithm with RSME 0.101 +/-0.23 where the error rate is the lowest compared to the two algorithms above. Therefore, the k-NN algorithm was chosen as the algorithm for the predictive measurement of surviving Covid-19 patients.

A Knowledge Discovery Process Extended to Experimental Data for the Identification of Motor Misalignment Patterns

The diagnosis of misalignment plays a crucial role in the area of maintenance and repair since misalignment can lead to expensive downtime. To address this issue, several solutions have been developed, and both offline and online approaches are available. However, online strategies using a small number of sensors show a higher false positive rate than other approaches. The problem is a lack of knowledge regarding the interrelations of a fault, disturbances during the diagnosis process, and capable features and feature vectors. Knowledge discovery in database is a framework that allows extracting the missing knowledge. For technical systems, optimal results were achieved by aligning (partially) automated experiments with a data mining strategy, in this case classification. The results yield a greater understanding of the interrelations regarding parallel misalignment, i.e., feature vectors that show good results also with varying load and realistic fault levels. Moreover, the test data confirm a specificity (range 0 to 1) for classification between 0.87 and 1 with the found feature vectors. For angular misalignment, potential vectors were identified, but these need further validation with a modified experiment in future work. For the study, two induction motors with 1.1 kW and 7.5 kW were considered. Furthermore, the findings were compared with additional motors of the same rated power. The findings of this work can help to improve the implementation of sensorless diagnostics on machines and advance the research in this field.

Network intrusion detection and classification using machine learning predictions fusion

The primary objective of an intrusion detection system (IDS) is to monitor the network performance and to look into any indications of malformation over the network. While providing high-security network IDS played a vital role for the past couple of years. IDS will fail to identify all types of attacks, when it comes to anomaly detection, it is often connected with a high false alarm rate with accuracy and the detection rate is very average. Recently, IDS utilize machine learning methods, because of the way that machine learning algorithms demonstrated to have the capacity of learning and adjusting as well as permitting a proper reaction for real-time data. This work proposes a prediction-level fusion model for intrusion detection and classification using machine learning techniques. This work also proposes retraining of model for unknown attacks to increase the effectiveness of classification in IDS. The experiments are carried out on the network security layer knowledge discovery in database (NSL-KDD) dataset using the Konstanz information miner (KNIME) analytics platform. The experimental results showed a classification accuracy of 90.03% for a simple model to 96.31% for fusion and re-trained models. This result inspires the researchers to use machine learning techniques with a fusion model to build IDS.

Time Series Forecasting of Top 3 Ranking Cryptocurrencies

Cryptocurrency has become a phenomenon worldwide. Although not all countries have legalized it, it is considered a promising investment asset. Currently, there are three top-ranking cryptocurrencies: Bitcoin, Ethereum, and Tether. This research aims to compare the performance of five forecasting algorithms, namely Autoregressive Integrated Moving Average (ARIMA), Neural Network, Support Vector Machine, Linear Regression, and Generalized Linear Model, using the dataset of Bitcoin, Ethereum, and Tether cryptocurrencies. The research methodology employed is Knowledge Discovery In Databases (KDD). The technique involves assessing the performance based on the Root Mean Square Error (RMSE) and comparing the results to find the most optimal model performance. The research findings indicate that for Bitcoin cryptocurrency, the Neural Network algorithm produced the most optimal results with an RMSE of 9180.534. For Ethereum cryptocurrency, the Neural Network algorithm demonstrated the best performance with an RMSE value of 537.528. Furthermore, for Tether cryptocurrency, the ARIMA algorithm yielded the best performance with an RMSE value of 0.003. Keywords – bitcoin, cryptocurrency, ethereum, forecasting, tether

Analisis Sentimen Pengguna Twitter Terhadap Grup Musik BTS Menggunakan Algoritma Support Vector Machine

BTS is a music group from South Korea which has global popularity around the world. As a music group that has global popularity, BTS influence on social media is quite large, with this influence making many companies interested in making them Brand Ambassadors (BA). A company can do a sentiment analysis of the public figure they want to make BA, later this sentiment can be taken into consideration for them to be able to determine the BA of their product. Sentiment analysis can also be used to distinguish the attitude of customers, users or followers towards a brand, topic, or product with the help of their reviews. Based on this, this study will analyze the sentiments of Twitter users towards music group BTS, using the Knowledge Discovery Database (KDD) research methodology, with 5 stages namely Data Selection, Data Preprocessing, Data Transformation, Text Mining and Evaluation. By using the Support Vector Machine (SVM) algorithm with a linear kernel, this study will do 3 scenarios with the distribution of training data and testing data 90:10 in scenario 1, 80:20 in scenario 2, and 70:30 in scenario 3. Confusion Matrix is used to evaluate the performance of the algorithm used and the results show that the best performance of the model formed is in scenario 1 and scenario 2.

A Comprehensive Review on Association Rule Mining and its Algorithms: Apriori, FP Growth, and SS-FIM

Abstract: The technique of extracting usable information from databases is known as data mining. It is employed to manage massive amounts of data. In order to forecast future trends and behaviours, data mining is performed. Today, data mining is a technique that is growing helpful since it has many uses. Massive amounts of data are produced daily in many different fields. Data mining is utilised since it is challenging to analyse such a big volume of data. It is often referred to as database knowledge discovery (KDD). Data mining technologies are used to forecast the behaviour and pattern of the data, assisting organisations in decision-making [1]. Association Rule Mining is one of the data mining techniques discussed in the present article. It has three algorithms. 1. Apriori, 2. FP-Growth, and 3. SS-FIM. In the current work, all three algorithms are discussed. After examining each algorithm, it was determined that the SS-FIM algorithm had the best combination of features

Knowledge Discovery in Databases for a Football Match Result

The analysis of sports data and the possibility of using machine learning in the prediction of sports results is an increasingly popular topic of research and application. The main problem, apart from choosing the right algorithm, is to obtain data that allow for effective prediction. The article presents a comprehensive KDD (Knowledge Discovery in Databases) approach that allows for the appropriate preparation of data for sports prediction on sports data. The first part of the article covers the subject of KDD and sports data. The next section presents an approach to developing a dataset on top football leagues. The developed datasets are the main purpose of the article and have been made publicly available to the research community. In the latter part of the article, an experiment with the results based on heterogeneous groups of classifiers and the developed datasets is presented.

Penerapan Data Mining Untuk Menentukan Jenis Asuransi Yang Paling Diminati Di Pt Aia Financial Dengan Metode Clustering

Finding information that is concealed in databases is referred to as data mining. Data mining is a process used to extract and find relevant information and related knowledge from numerous very big datasets using statistical, mathematical, artificial intelligence, and machine learning approaches. The steps involved in the data mining process start with the selection of data from the source data to the target data, followed by preprocessing to enhance the quality of the data, transformation, data mining, and stages of interpretation and assessment that result in new information. Organizing data sets into many groups through the process of clustering ensures that objects in one group share a lot in common and differ greatly from those in other groups. Data mining technique K-Means Clustering divides data into one or more clusters. RapidMiner is the program used in this data mining application, which uses the Knowledge Discovery in Databases (KDD) stage. According to this report, AIA Family First Protection, AIA Prolink, and AIA Life Secure are the three most popular insurance products. With the use of this research, PT. AIA Financial will be able to identify the most popular insurance product categories and market these products to the general public.

Analisis Sentimen Ulasan Aplikasi Binar Pada Google Play Store Menggunakan Algoritma Naïve Bayes

Binar is an online learning platform that provides courses and certifications in the digital field. The Binar app has been downloaded 500,000 times and has a rating of 3.6 on the Google Play Store. However, user ratings sometimes do not match their reviews. In application development, not only the number of ratings but also user opinions need to be considered. Therefore, developers must be able to interpret every opinion given, and sentiment analysis was conducted using the Naïve Bayes Multinomial and Bernoulli algorithms along with Information Gain feature selection to interpret user opinions. This study used the Knowledge Discovery in Database (KDD) method. The data used consisted of 713 reviews of the Binar app, including 518 positive and 195 negative reviews. The best results were obtained in the 9:1 data split scenario with the Bernoulli Naïve Bayes model achieving an accuracy of 93.06%, precision of 87.04%, recall of 100%, f1-score of 93.07%, and AUC of 0.988.

Analysis of Student Performance Applying Data Mining Techniques in a Virtual Learning Environment

Students' academic performance is a key factor for educational institutions and society, which is an important indicator of the quality of the teaching-learning process and the appropriation of knowledge. Its analysis allows an understanding of the behavior of students and teachers, generating valuable knowledge for making timely academic decisions. In this study, the following phases were carried out: (i) identification of factors that influence the academic performance of engineering university students, (ii) early prediction of academic success, and (iii) identification of use patterns in a virtual learning environment (VLE). The Knowledge Discovery in Databases (KDD) methodology was applied based on predictive and descriptive data mining techniques, using academic and socioeconomic data and interactions (resources and activities) with the VLE. The tools and programming languages ​​used were Pentaho Data Integration for data integration and processing; Jupyter Notebook, Python, and Scikit-Learn for correlation analysis and prediction modeling; and R Studio for the clustering task. The results show that VLE resources such as files, links, and activities such as participation in forums are factors that lead to good academic performance. On the other hand, it was possible to make predictions of academic success (pass or not) with an accuracy greater than 95% and to identify the main patterns of use of the VLE. The group with excellent academic performance (grades 9 to 10) is recognized for using file-type resources and high participation in class and forum activities.

Comparison of distance metric in k-mean algorithm for clustering wheat grain datasheet

One of the data mining models is clustering, clustering models can be used to create groupings of data. Clustering is done by creating groups of data that are close to each other. The research was conducted by clustering wheat seed datasheets. The wheat grain datasheet contains various types of wheat data. The purpose of this research is to create a clustering model. The algorithm used is the K-means algorithm and a comparison is made with several distance Metric algorithms. The datasheet used was tested with the K-means algorithm and tested the clustering value (k) ranging from k = 2 to k = 6. Comparison of clustering results with K-means is also done by comparing with distance metric algorithms, namely Euclidean distance, Manhattan distance, and Chebychev distance. All testing processes are evaluated, and the evaluation is done to select many good groupings. The evaluation process is carried out using the Davis-Bouldin method. The results of the grouping that has been done, each seen Davis Bouldin evaluation. The evaluation value of Davis Bouldin is sought from the smallest value and if the evaluation result is negative, the value is solved. The research method used is Knowledge Discovery in Database (KDD). The results showed that the same datasheet and using the K-means algorithm and the same evaluation resulted in different evaluation values. The Euclidian, Manhattan, and Chebychev algorithms produce the best k value of 2, The conclusion of the wheat seed datasheet clustering research produces a value of k = 2

Komparasi Metode Apriori dan FP-Growth Data Mining Untuk Mengetahui Pola Penjualan

Sales data is generally still rarely used, as well as the Perfume Corner shop just piling up in the database, even though there are problems experienced by the store regarding sales data for the best-selling products and to increase the number of sales of subsequent perfume products, so that the store can survive and develop even better. The algorithm that can be used to manage sales data to overcome this problem is Apriori. The research method used in this research is the KDD (Knowledge Discovery in Database) process. This research produces a high frequency pattern for itemsets with a minimum support value of 20% resulting in products that become The Most Tree Items namely Jo Malone 82.49%, Zarra 28.25%, and Zwitsal 20.34%. While the association rules formed from the value of Min. Supp 20% and Min. Conf 80%, get a combination of 2 itemsets, namely Jo Malone and Zarra. Whereas for the combination of 3 itemsets, namely Jo Malone, Zarra and Baccarte with valid and strong status, it is proven by a lift value greater than 1, therefore the association rules are very appropriate to be used.

Improved model for traffic management system using KDD and big data: case study Jordan traffic

This paper addresses the challenge of the increasing number of traffic accidents resulting from population growth and increased vehicle usage, leading to significant economic and environmental impacts. To address this issue, this paper proposes an improved model for traffic accident management that employs Knowledge Discovery in Databases (KDD) and extensive data analysis techniques to extract the main factors contributing to car accidents. A case study of traffic accidents in Jordan is conducted, utilizing actual data from the Department of Statistics and the Public Security Directorate Records between 2016 to 2020 to identify the primary factors contributing to road accidents and their effects on various types of injuries. The study identifies 11 factors that contribute to car accidents, with driver error being the primary factor for increasing the number of accidents and injuries. Additionally, minor car accidents are found to cause more injuries based on the analysis of the proposed approach. The findings provide a solid foundation for developing an accurate and precise scheme to reduce or eliminate the number of road accidents, which is essential for transitioning to smart cities. Using big data and KDD techniques could significantly improve current traffic accident management practices, informing the development of new policies and regulations to improve road safety and reduce economic and environmental impacts. This paper presents a promising solution for addressing traffic accident management, contributing to safer and sustainable urban environments.

Techniques for Data Mining Prediction in the Health Care Sector

Data mining is another term for knowledge discovery in databases (KDD). It's an interdisciplinary field that focuses on rooting meaningful knowledge from data in all sectors similar as health, education, and business. Currently, with the covid epidemic affecting everyone and rising coronavirus cases causing nursing home beds, oxygen, vaccines and individuals to be denied by hospitals, the health structure of the elderly is in the spotlight. There's a wealth of information accessible in the medical world about these diseases. Data booby-trapping concepts may be used to prize meaningful styles from this type of material in order to prognosticate unborn followings. This study emphasizes on several mining approaches that will be applied in the therapy assiduity to achieve the stylish results.

Analisis Trend Berita Media Online di Masa Pandemi Covid-19 dengan Metode Monte Carlo

The News reports for online media always follows the trending issues that are actually predictable. In general, online media news apart from presenting various events that have occurred, also focuses on reporting issues which is of concern to the public. During the Covid-19 pandemic, almost all mass media, including online media, continued to report various news about the Covid-19 case and its impact on society. The problem is how to understand the news accumulation in a certain period of time to find out the trend in reporting on Covid-19. Because the news has resulted in a huge accumulation of data related to Covid-19 news which can be used to build data mining. The purpose of this study is to analyze online media coverage during the Covid-19 pandemic using the Monte Carlo method to find out news trends. The process uses a data mining approach by collecting a lot of data from existing reports and processing the data to find important information from the data set. The data used to analyze news trends were taken from the main news on the online media Suarapembaruan.com for seven (7) months and processed using knowledge discovery in database (KDD) techniques. The Monte Carlo method is applied to data mining to calculate probability values or possible values that will occur in the future. The results of this study indicate that the trend of news related to Covid-19 in the future will be dominated by News on the Handling of Covid-19 with a trend percentage rate of 23.12 percent. Meanwhile, the news about the Covid-19 Case and the Impact of Covid-19 was getting lower with a trend rate of 21.82 percent and 19.22 percent. Meanwhile, The Other News with related to Covid-19 has a higher trend percentage of 35.84 percent

CLASSYFYING STUDENTS DECISION MAKING ABILITY USING K-NEAREST NEIGHBOR FOR DETERMINING STUDENTS SUPPLEMENTARY LEARNING

Mathematical decision-making ability is a complex cognitive process of finding problems solutions which is continuously explored and optimized for undergraduate students. The current research only focus on categorization on class score average into high, medium and low abilities. As a result, the lecturers do not have any standard categories to classify students’ abilities as a reference in planning supplementary learning that could optimize undergraduate students’ abilities. Therefore, the purpose of this study is to determine a classification model for undergraduate students' mathematical decision-making abilities that require supplementary learning based on the identification of the shortest distance of a new data from an existing data directory. The research method involved data mining techniques with the KNN classification model through the Knowledge Discovery in Database (KDD) process starting from data selection, pre-processing, transformation, data mining and interpretation/evaluation [1]. A total of 100 data were used as research samples which were divided into training data and testing data. Based on the test results, it is obtained that the accuracy of the classification model is 95% for the parameter value k = 15, meaning that each predicted testing data for the classification class is close to the actual condition with the number of neighbors 15 data from the training data.

Intelligent ship inspection analytics: Ship deficiency data mining for port state control

To provide safe, secure, and environment-friendly maritime transport, it is an essential point to define and take appropriate countermeasures against the substandard ships in maritime transport. Port State Control (PSC) is one of the critical inspection regimes recognized worldwide to define substandard ships. However, one of the biggest challenge faced in ship inspection is the necessity of checking many items through limited time and qualified human resources. Therefore, in this study, it is aimed to develop a decision support system to increase the effectiveness of ship inspections. In accordance with this, we proposed an intelligent ship inspection analytics (I-SIA) model based on the Knowledge Discovery in Database (KDD) process by utilizing fuzzy c-means clustering and apriori algorithms. The I-SIA model provides the determination of ship deficiency patterns based on specific ship attributes, and subsequently predicts main/sub inspection items to be focused on through former inspection records of the ship to be inspected. In the case study performed for a selected ship, I-SIA predicts critical 5 main and 23 sub-inspection items that need to be checked with high priority, among 17 main inspection items and out of more than 500 sub-inspection items. Herewith, I-SIA provides a ship-specific and dynamic ship inspection based on deficiencies recorded in previous inspections and detected during inspection. Thus, it becomes possible to enhance the effectiveness of the inspection via defined deficiency patterns.

A New Approach for Optimizing the Extraction of Association Rules

Association rule methods are among the most used approaches for Knowledge Discovery in Databases (KDD), as they allow discovering and extracting hidden meaningful relationships between attributes or items in large datasets in the form of rules. Algorithms to extract these rules require considerable time and large memory spaces. This paper presents an algorithm that decomposes this complex problem into subproblems and processes items by category according to their support. Very frequent items and fairly frequent items are studied together. To evaluate the performance of the proposed algorithm, it was compared with Eclat and LCMFreq on two actual transactional databases. The experimental results showed that the proposed algorithm was faster in execution time and demonstrated its efficiency in memory consumption.

An intrusion detection system using Exponential Henry Gas Solubility Optimization based Deep Neuro Fuzzy Network in MANET

The smoothness and dynamic nature of the exploitation are significantly influenced by Mobile Ad hoc Networks (MANETs). The wireless and active environments, on the other hand, are vulnerable to a range of threats. Enhancing the reliable intrusion detection model is crucial to safeguarding the MANET from various malicious attacks. In this paper, an effective intrusion detection method for MANET is presented based on a Deep Neuro Fuzzy Network based on Exponential-Henry Gas Solubility Optimization (EHGSO). The newly designed EHGSO algorithm is used to select the optimal routes in the early phases of safe routing. The fitness metrics that define this approach include energy, distance, neighbourhood quality, and link quality. The proposed EHGSO combines the Henry Gas Solubility Optimization (HGSO) with Exponential Weighted Moving Average (EWMA). The second phase, in which the conveyed data packets are altered and the Knowledge discovery in databases (KDD) features are extracted, starts the intrusion detection phase at the base station. Following the extraction of the KDD features, data augmentation is performed. The Deep Neuro Fuzzy Network is trained using the suggested EHGSO method before performing intrusion detection. The proposed method demonstrates higher performance when compared to all other existing technologies. In without attacks scenario, the values achieved by the proposed method considering the metrics, such as energy, throughput, packet drop, jitter, Performance and Development Review (PDR), precision, and recall is 0.342 J, 134975 kbps, 4.123, 0.086, 95.877%, 0.950, and 0.924, respectively.

A Robust-Reliable Decision-Making Methodology Based on a Combination of Stakeholders’ Preferences Simulation and KDD Techniques for Selecting Automotive Platform Benchmark

The automotive family design is known as one of the most complex engineering design problems with multiple groups of stakeholders involved from different domains of interest and contradictory attributes. Taking into account all stakeholders’ preferences, which are generally symmetrical, non-deterministic distributions around a mean value, and determining the right value of attributes for each alternative are two basic challenges for these types of decision-making problems. In this research, the possibility to achieve a robust-reliable decision by focusing on the two aforementioned challenges is explored. In the proposed methodology, a random simulation technique is used to elicit stakeholders’ preferences and determine the relative importance of attributes. The decision space and values of attributes are determined using the Knowledge Discovery in Databases (KDD) technique, and to achieve a robust-reliable decision, statistical and sensitivity analyses are performed. By implementing this methodology, the decision-maker is assured that the preferences of all stakeholders are taken into account and the determined values for attributes are reliable with the least degree of uncertainty. The proposed methodology aims to select benchmark platforms for the development of an automotive family. The decision space includes 546 automobiles in 11 different segments based on 34 platforms. There are 6223 unique possible states of stakeholders’ preferences. As a result, five platforms with the highest degree of desirability and robustness to diversity and uncertainty in the stakeholders’ preferences are selected. The presented methodology can be implemented in complex decision-making problems, including a large and diverse number of stakeholders and multiple attributes. In addition, this methodology is compatible with many Multi-Attribute Decision-Making (MADM) techniques, including SAW, AHP, SWARA, and TOPSIS.