Data Mining Applications Research Articles

Max-Miner Method in Improving Food Sales Strategy

Competition in the corporate world is currently very intense, especially in the retail business sector, one of which is the retail food business. Building a minimarket retail business, provides staple foods that are unavoidable using information technology to support the smooth sales of these products. The use of this information technology has become a necessity in the retail minimarket business world whose aim is to provide maximum profits and minimal losses with promotions that must be done in terms of providing the best service in a retail minimarket that must use the best business strategy, but sometimes the retail minimarket manager constrained in determining the sales strategy. One of the factors is the difficulty of producing an analysis related to the products sold. Therefore we need an analysis of the application of data mining so that product sales at retail minimarkets are increasing and service to consumers is getting better. In this design data mining and algorithms are used, namely market basket analysis and Max-Miner. By applying the Max-Miner method in the data mining process for sales strategies at retail minimarkets, it will produce rule associations that will become product recommendations that will provide a decision in sales strategy for the mini market

Application of data mining to extract knowledge about the occurrence of fistulas after palatoplasty

Application of data mining to extract knowledge about the occurrence of fistulas after palatoplasty

Classification and prediction of student performance data using various machine learning algorithms

In today's competitive world, it is critical for an institute to forecast student performance, classify individuals based on their talents, and attempt to enhance their performance in future tests. Students should be advised well in advance to concentrate their efforts in a specific area in order to improve their academic achievement. This type of analysis assists an institute in lowering its failure rates. Based on their prior performance in comparable courses, this study predicts students' performance in a course. Data mining is a collection of techniques used to uncover hidden patterns in massive amounts of existing data. These patterns may be valuable for analysis and prediction. Education data mining refers to the collection of data mining applications in the field of education. These applications are concerned with the analysis of data from students and teachers. The analysis might be used for categorization or prediction. Machine learning such as Nave Bayes, ID3, C4.5, and SVM are investigated. UCI machinery student performance data set is used in experimental study. Algorithms are analysed on certain parameters like- accuracy, error rate.

Enhancing Customer Relationship Management Using Fuzzy Association Rules and the Evolutionary Genetic Algorithm

The importance of Customer Relationship Management (CRM) has never been higher. Thus, companies are forced to adopt new strategies to focus on customers, given the competitive climate in which they operate. Also, companies have been able to maintain customer data within large databases that contain all information related to customers, thanks to the tremendous technological development seen recently. Multilevel quantitative association mining is a significant field for achieving motivational associations between data components with multiple abstraction levels. This paper develops a methodology to support CRM to improve the relationship between retail companies and their customers in the retail sector to retain existing customers and attract more new customers, by applying data mining techniques using the genetic algorithm through which an integrated search is performed. The proposed model can be implemented because the proposed model does not need the minimum levels of support and trust required by the user, and it has been confirmed that the algorithm proposed in this research can powerfully create non-redundant fuzzy multi-level association rules, according to the results of these experiments.

Retracted] Multisensor Cross‐Media Data Mining Method Assisted by Expert System

The information generated by the multifaceted and multilevel processing of the data of multiple sensors is more meaningful than the information obtained by a single sensor and provides accurate information and decision‐making basis for various application systems. How to apply data mining theory to the multisensor cross‐media field has become a research hotspot. Through the analysis of multisensor cross‐media data, it is of great significance to dig out the important rules, information, or knowledge hidden in it and use it for cross‐media retrieval engine. The rule acquisition is the “bottleneck” problem of the expert system. This paper adopts the data mining method based on the rough set to acquire the rules and improves the basic algorithm of attribute reduction. Using the attribute reduction algorithm and the heuristic value reduction algorithm, the calculation is simplified and the reduction efficiency is improved. In the presentation, according to the characteristics of cross‐media and the application requirements of expert systems, this paper takes the case representation based on features as the basis and classifies cases according to feature attributes. In case retrieval, according to the hierarchical structure of case features, the entire case database is organized into a multilevel hierarchical index structure. In this paper, a cross‐media retrieval engine is constructed from the perspective of classifier design, and the Euclidean distance is used as the similarity matching model of image content. The mutual retrieval of images and audios preliminarily forms the design process of retrieval from one media type to another and establishes a corresponding cross‐media index. The experimental results show that the algorithm has better processing effect and higher accuracy than other algorithms. Different k‐nearest neighbor values were selected in the experiment, and it reached about 96% in the test environment of libsvm toolbox, which is better than the processing results of LE and LLE algorithms.

An Analysis of Data Mining Techniques in Software Engineering Database Design

With the advent of Internet information technology, data mining technology has been applied in various fields of China's social and industrial development, and has promoted the quality development of the industry [1]. Nowadays, people are widely influenced by Internet computer technology, and the application of computer technology has been indispensable in life, work and study. At the same time, data mining technology arises from Internet communication technology and is used as an important technical means of operation and development by various industries, especially the application of data mining technology in university software engineering teaching is becoming more and more widespread, but there are still many unavoidable problems that require scholars to pay more attention. The article analyzes and discusses the significance of applying data mining technology in software engineering [2], and on this basis, the application path of data mining technology in software engineering database design is proposed, including: mining information, mining loopholes, mining execution records and open source. It is hoped that relevant technology users can efficiently apply data mining technology to make software engineering It is more reasonable and efficient, so that software engineering can be further developed. This paper presents the implementation and performance analysis of the teacher management decision-making system based on data mining, and introduces the implementation process of the system in detail.

Применение мультимодального трансформера для прогнозирования выходных параметров насыщенных углеводородных соединений из состава тяжелой нефти в присутствии катализаторов

Heavy oil fields are a promising energy source in the future due to the depletion of natural sources of light oil. However, extraction, transportation and refining of heavy oil is significantly more complicated than light oil - difficulties arise at almost all technological stages. One of such stages is laboratory analytics of heavy oil and selection of the most optimal catalyst for extraction of required fractions from crude oil sample. Different catalysts are actively used in petrochemical laboratories, but special attention is paid to those of them, the basis of which is metal. In this study, catalysts based on six different metals namely zinc, nickel, copper, manganese, lead and sodium were analyzed. In order to analyze the yield ratios of the required components from the crude heavy oil composition, it is necessary to test different types of catalysts sequentially on a base sample. The yield of different hydrocarbons on a small volume of oil can be reliably estimated by chromatographic study, which takes about 68 minutes for both the base oil sample and the different catalysts. Since testing 6 different catalysts would require almost 7 hours of chromatographic analysis, a rational solution would be to apply data mining techniques to this task. A multimodal transformer model was proposed to solve this problem. It takes as input two modalities: a chromatogram of a sample of pure crude oil presented as graphical data and accompanying tabular data, which are also generated by the chromatograph and consist of text and numbers. At the output, the model produces predictive tabular data that formalize the redistributed group composition of the oil and describe both the names of the newly produced hydrocarbons and their two qualitative characteristics: time and yield area. Obtaining the prediction makes it possible to significantly reduce the time, hardware and human resources required to select the right type of catalyst in petrochemical laboratories. In the process of the study, it was found that training of the intellectual model on the data of one field allows to perform further similar forecast with acceptable accuracy for the data of another heavy oil field. The magnitude of the prediction error of the intelligent model satisfies the requirements set by the petrochemical laboratory for practical application of the multimodal transformer.

A Data-driven Causality Modeling Framework: An Empirical Study of Modeling the Effect of Indoor Air Quality Perception on Students' Cognitive Performance

In the information age, a paradigm revolution in applied data mining methods has emerged in response to the data explosion in management science research, but this artificial intelligence-based automated data-driven modeling process creates a new "big relationship" problem, where a large number of correlations obfuscate the identification of true causal relationships. In this paper, we use a causality modeling framework that combines correlation modeling and causality pruning processes. First, symbolic regression is used to model white-box correlations of human intelligence, and then spurious correlations that do not conform to causal graph theory are pruned so that ultimately causal relationships and explicit candidate models describing these relationships can be found automatically. In an empirical research problem, the framework is compared with a traditional hypothesis construction-validation process, and the results are consistent between the two. The proposed framework implements a data-driven "correlation + causation" automatic modeling capability, which will greatly improve modeling efficiency and reliability.

Security Risk Management System for the Construction and Operation of Smart Port Area Based on BP Neural Network Algorithm

With the construction of smart cities, smart ports are developing rapidly, providing favorable conditions for enhancing port competitiveness, promoting industrial upgrading, and economic transformation. With the continuous progress of society and science and technology, the safety risks of intelligent ports are also increasing. In order to better address these risks, it is necessary to establish a stable and efficient management system. This paper applies data mining and fuzzy control theory technologies for large-volume, high-precision real-time information collection, based on the research of the BP neural network algorithm. By establishing and analyzing a safety risk management system model based on the BP multilayer neuron network, it provides a rationality evaluation method and design concept. After designing the safety risk management system, this paper conducts a performance test on the system. The test results show that through optimization, the accuracy, generalization ability, and reliability of the model can be improved. Its accuracy is as high as 92% or more, fully demonstrating the system's reliability and superiority!

Stock market uncertainty determination with news headlines: A digital twin approach

<abstract><p>We present a novel digital twin model that implements advanced artificial intelligence techniques to robustly link news and stock market uncertainty. On the basis of central results in financial economics, our model efficiently identifies, quantifies, and forecasts the uncertainty encapsulated in the news by mirroring the human mind's information processing mechanisms. After obtaining full statistical descriptions of the timeline and contextual patterns of the appearances of specific words, the applied data mining techniques lead to the definition of regions of homogeneous knowledge. The absence of a clear assignment of informative elements to specific knowledge regions is regarded as uncertainty, which is then measured and quantified using Shannon Entropy. As compared with standard models, the empirical analyses demonstrate the effectiveness of this approach in anticipating stock market uncertainty, thus showcasing a meaningful integration of natural language processing, artificial intelligence, and information theory to comprehend the perception of uncertainty encapsulated in the news by market agents and its subsequent impact on stock markets.</p></abstract>

Process Mining in Manufacturing: A Literature Review

Process mining in manufacturing is a newly expanding field of research in the application of data mining and machine learning techniques and the focus of business processes. Although it is an exciting subject of the recent past and business processes, sufficient research has not been done. Decision support systems such as enterprise resource planning, customer relationship management, and management information systems store the most valuable resource data of process details and event logs. In the advanced information systems of tomorrow, the process management, analysis, and modelling functions of modern enterprises will take their place as a necessity. As a requirement, the fundamental purpose of process mining in production is to refine data from event logs, automatically create process models, compare models with event logs, and improve and make development continuous. Our work is to contribute to application and research studies by drawing attention to process mining in the context of production. It is based on the literature review and primary stages of business process mining publications in the last decade with a production focus. An overview is discussed as a roadmap for future research with meaningful results.

DATA MINING UNTUK KLASIFIKASI STATUS PANDEMI COVID 19

We all know that since the number of Covid-19 cases has increased in Indonesia, many problems have arisen in society. Covid-19 has crippled the socio-economic conditions of all Indonesian people. As a result of the Covid-19 problem, the Indonesian government must establish policies such as issuing rules for social distancing, and also calling for Work From Home for those who work as employees, imposing restrictions on each region, building hospitals to handle Covid-19, and others. With the existence of government policy decisions will have an impact on all people, both the community. Socio-economic problems arise and have a direct impact on society. The purpose of this study is to prediction the covid 19 pandemic. This study applies data mining techniques with the naïve Bayes algorithm with software implementation using Tanagra software. The results of this study can be used to see the clustering pattern of the Covid 19 pandemic, and it can be seen from the results of the probabilities of the Covid data so far for the spread of Covid 19 which has 90.7% correct predictions and 9.3% wrong predictions.

Proposed framework for applying data mining techniques to detect key performance indicators for food deterioration

One of the most prosperous domains that Data mining accomplished a great progress is Food Security and safety. Some of Data mining techniques studies applied several machine learning algorithms to enhance and traceability of food supply chain safety procedures and some of them applying machine learning methodologies with several feature selection methods for detecting and predicting the most significant key performance indicators affect food safety. In this research we proposed an adaptive data mining model applying nine machine learning algorithms (Naive Bayes, Bayes Net Key -Nearest Neighbor (KNN), Multilayer Perceptron (MLP), Random Forest (RF), Support Vector Machine (SVM), J48, Hoeffding tree, Logistic Model Tree) with feature selection wrapper methods (forward and backward techniques) for detecting food deterioration’s key performance indicators. Therefore, results before and after applying wrapper feature selection methods have been compared, analyzed, and interpreted. In conclusion the proposed model applied effectively and successfully detected the most significant indicators for meat safety and quality with the aim of helping farmers and suppliers for being sure of delivering safety meat for consumer and diminishing the cost of monitoring meat safety.

Mixture correntropy based robust multi-view K-means clustering

Multi-view clustering has been a significant research problem in unsupervised clustering in recent years and has important applications in computer vision, data mining and other fields. To explore more clustering information from different views and improve performance, researchers have developed a variety of multi-view clustering algorithms by applying the same L1 or L2,1 norm constraints for different views, which essentially assumes that the errors satisfy the characteristics described by the corresponding norms. It is clear that such assumption is not always satisfied in the real-world datasets. Therefore, the performance of existing algorithms can decrease under the complex noise that may be contained in the data. To solve the above problem, we introduce an information theoretic learning-based metric into a multi-view model to address noise. First, we propose a new mixture correntropy (MC) metric. Then, we propose a MC based robust multi-view K-means clustering (MC-RMVKM) method by applying the MC to an existing model. Furthermore, an efficient iterative algorithm is derived to solve the joint MC-RMVKM learning problem. Extensive results on real-world multi-view data show that the MC-based algorithm outperforms other state-of-the-art algorithms.

PENERAPAN DATA MINING DENGAN METODE SINGLE MOVING AVERAGE UNTUK MENENTUKAN JUMLAH PENJUALAN OBAT

Anita Farma 2 Pharmacy is one of the private pharmacies in the city of Baubau which still implements a manual system for drug sales management and has not implemented a method to predict future drug sales. So that there are various kinds of drugs that experience excess stock and some drugs experience stock shortages. The purpose of this study is to make it easier to predict future drug stock requirements at Anita Farma 2 Pharmacy using the application of data mining with the web-based Single Moving Average method. The method used in this study is the Single Moving Average (SMA) method which is a technique for calculating the average of a number from the latest actual value, updated as new values become available, carried out to forecast the following periods using the method Mean Absolute Percentage Error (MAPE) to calculate the accuracy of predictions on the Single Moving Average method. The results of the research system used the Single Moving Average method, namely period 2 (M2) of 23.25%, period 3 (M3) of 26.23% and period 4 (M4) of 28.22% using 5 drug samples with sales for 5 years from January 2017 to December 2021 Among the 3 periods of drug sales forecasting tested using manual calculations with MAPE it can be concluded that period (M2) is better than periods M3 and M4 with a percentage of MAPE "Good Forecasting" of 16.67% and the smallest error rate of 23.25 %.

Data mining application with machine learning algorithms to manage interest rate risk

In trade, the risks taken may increase the expected income; however, they may also cause large amounts of losses as well. Banks transfer the capital and the deposits they collect from their clients to the individuals or institutions in need of profit, taking certain risks into account. One of the important risks taken in this process of capital transfer is the market's change in interest or profit share rates. If the bank transfers the deposit collected with a certain commitment to the market at a lower rate, it will make a loss. Models for predicting future interest or profit share rates gain importance for preventing this situation. The aim of this study is to determine which variables will be taken into account for the loan interest rate that banks will offer to their customers during the lending process, and to create a machine learning model that can predict the loan interest rate that the bank will offer to its customers by using these variables. Multiple Linear Regression analysis was performed to demonstrate the relationship between the variables selected based on the literature review, expert opinions, and the interest rate. In order to facilitate decision-makers in practice, Random Forests, Decision Trees, K-Nearest Neighbours (KNN), Artificial Neural Networks (ANN), and Support Vector Machine (SVM) algorithms from machine learning algorithms were compared by using the prediction model. Accuracy Rate, Cohen's Kappa, Precision, Sensitivity, and F-Measure measurements were used to compare the algorithms used in the study. According to the analysis results, it was observed that the Random Forest algorithm was more successful on the first model consisting of weekly data. The Decision Tree algorithm succeeded more on the second model consisting of monthly data prediction performance. In the model consisting of weekly data, USD Selling Price, Stock Index (BIST100), and Central Bank Gold Reserve from the Multiple Linear Regression variables were found significant in affecting the interest rate.

Функциональная модель системы автоматизированного проектирования железных дорог в информационно-цифровой среде

Purpose: Development of new technology for automated solution for railway design tasks in informational-digital environment. Methods: Structural system analysis methodology, system theory, the principle of decomposition are applied. Results: Based on analysis results for structure and functional possibilities of widely used modern systems of railway and highway automated design, the procedures sequence for existing technology of railway automated design was defined. Functional model of railway automated design system in informational-digital environment, based on the application of informational modeling technologies, data mining and mathematical optimization methods, is proposed. Practical significance: The proposed functional model represents the basis for the development of automated design system that will allow implementing new technology for railway automated design, reducing labor costs and time for the development of design solutions, improving their quality and efficiency.

Towards Better Detection of Fraud in Health Insurance Claims in Kenya: Use of Naïve Bayes Classification Algorithm

The extent, possibility, and complexity of the healthcare industry have attracted widespread fraud that has contributed to rising healthcare costs hence affecting patients’ health and negatively impacting the economy of many countries. Despite putting up various technologies and strategies to fight fraud such as planned, targeted audits, random audits, whistle-blowing, and biometric systems, fraud in claims has continued to be a challenge in most of the health insurance providers in Kenya. This paper explored the application of data mining in detecting fraud in health insurance claims in Kenya. Classification algorithms (Naïve Bayes, Decision Tree and K-Nearest Neighbour) were used to build predictive models for the knowledge discovery process. After conducting several experiments, the resulting models showed that the Naïve Bayes works well in detecting fraud in claims with 91.790% classification accuracy and 74.12% testing hit rate. A prototype was developed based on the rules extracted from the Naïve Bayes model, which, if adopted, will save costs by detecting fraud as it is committed. Fraud detection in health insurance claims is much needed in many countries so as to help reduce loss of money and in return improve service delivery to patients.

The Prediction Model of Air Quality in Changsha City Based on Artificial Neural Network

The air quality is a hot topic noticed by all the present society. Although modern technology has helped to collect data of the air position, the huge amount of irregular data also makes it hard to detect the pattern of air pollution. Therefore, this research is to apply data mining to the analysis of the air quality. Data mining is a comprehensive mining of information including trends, features and correlations hidden in data. Data mining can extract the value of data which was not known from numerous random data. This research is based on the daily average values of six major pollutants in Changsha from January 1, 2015, to December 31, 2017, and analyzes the relationship between the air quality and socioeconomic factors of Changsha to establish a prediction model using B-P back propagation neural network. The final model of the research fits the actual data well and predicts the future value of API precisely. However, considering the slow convergence rate and the addition of new samples which may affect the previously learnt samples, further exploration on the B-P neural network model can be launched with other mathematical theories.

Comparative Analysis of Data Mining Techniques to Predict Cardiovascular Disease

Cardiovascular disease is the leading cause of death. In recent days, most people are living with cardiovascular disease because of their unhealthy lifestyle and the most alarming issue is the majority of them do not get any symptoms in the early stage. This is why this disease is becoming more deadly. However, medical science has a large amount of data regarding cardiovascular disease, so this data can be used to apply data mining techniques to predict cardiovascular disease at the early stage to reduce its deadly effect. Here, five data mining classification techniques, such as: Naïve Bayes, K-Nearest Neighbors, Support Vector Machine, Random Forest and Decision Tree were implemented in the WEKA tool to get the best accuracy rate and a dataset of 12 attributes with more than 300 instances was used to apply all the data mining techniques to get the best accuracy rate. After doing this research people who are at the early stage of cardiovascular disease or probably going to be a victim can be identified more accurately.