Data Mining Models Research Articles

Comparative forensic discrimination of pink lipsticks using fourier transform infra-red and Raman spectroscopy

As a routine, lipstick is daily used by females and can easily be transferred to clothes, cups, tissue papers, and other objects. The analysis of lipsticks is a relatively new and exciting field in forensics, helping to identify suspects in criminal cases where lipstick evidence has been left at crime scenes. By matching a specific brand of lipstick to a sample, investigators can positively connect certain individuals to locations or people, helping to aid in their investigation and subsequent proceedings. In this present study, 20 different pink shade lipsticks of the same manufacturer were analyzed using Vacuum FT-IR, and Raman spectroscopy to show a differentiation percentage of 95.8% between the samples. Data analysis using data mining techniques was performed on FT-IR spectra. Principle Component Analysis (PCA) was used as a data mining model for classification purposes, and it was able to distinguish between lipsticks samples based on their FT-IR spectra.

Model based on machine learning to predict the approval of an engineering course

In the context of virtual and distance education, the course "Introduction to Telecommunications Engineering" at the National Open and Distance University (UNAD) has faced the challenge of student attrition at first enrolment. To address this problem, a research study was carried out to develop a Data Mining model based on the CRISP-DM methodology. The study used a dataset with 808 student records collected in different academic periods of the year 2021. Several Machine Learning techniques were applied, such as Random Forest, Tree Decision, Knn, SVM and Neural Network, evaluating their performance through k-10 cross-validation. The results revealed that the Random Forest model obtained the best performance, predicting with 80% accuracy the passing rate of the course. This model is relevant for identifying students at risk of dropping out of the course early, allowing the implementation of proactive strategies to address dropout and improve course indicators. Although some results were slightly lower than previous research, the model remains a valuable tool to support student retention strategies in the "Introduction to Telecommunications Engineering" course at UNAD.

A comparative study on data mining models for weather forecasting: A case study on Chittagong, Bangladesh

The primary focus of this study is to analyze and predict the patterns of this essential feature of the natural world. This study analyses and predicts the daily weather patterns of a specific urban area. This article utilizes weather data over 20 years to analyze the climate patterns of Chittagong city. A total of 12 distinct Data Mining models were employed to predict daily weather patterns. The algorithms can be categorized into three distinct types, namely rules-based, tree-based, and function-based. To evaluate the effectiveness of the models, various performance metrics were computed, including precision, recall, accuracy, F-measure, and the area under the receiver operating characteristic curve (ROC area). Based on the results obtained, it can be concluded that among the 12 algorithms evaluated, J48 exhibits the highest level of performance and accuracy. The J48 classifier demonstrated an accuracy of 82.30%, precision of 82.40%, recall of 82.20%, f-measure of 84.20%, and a ROC area of 97.8%. Furthermore, a comprehensive analysis of the confusion matrix for all twelve algorithms was conducted to facilitate further evaluation.

An ensemble based data mining model for contingency analysis of power system under STLO

<span lang="EN-US">In a large, interconnected power system, contingency analysis is a useful tool for pinpointing the potential consequences of post-event scenarios on the system's safety. In this work, the Newton-Raphson technique is applied to every single outage of a transmission line to compute the load flows. For the static security classification of the power system, the line voltage stability performance index (LVSI) is used. There are three levels of static security of power system namely: non-critical (the least severe), semi-critically insecure (the next lowest severe), and critical (the next highest severe). The various data mining techniques such as decision trees, bagging-based ensemble methods, and boosting-based ensemble methods were applied to assess the severity of the line under various loading and contingency conditions. Test systems based on the IEEE 30 bus system were used with the proposed machine learning classifiers. The experimental results proved that bagging based ensemble method provided better accuracy compared to the decision tree and the AdaBoost ensemble method for predicting the power system security assessment. The bagging-based ensemble method has a predictive accuracy of 85% and an AUC of 0.94.</span>

Use of Statistical Analysis Tools for ECG Processing

The purpose of the article is to study the algorithms of statistical and intellectual data analysis and their use for processing and analysing electrocardiograms (ECG). The methods and algorithms that form the basis of statistical data processing and analysis are considered. The research methods are based on the application of statistical methods and algorithms for the analysis and pre-processing of medical data. Pre-processing is a necessary step in data processing, which makes it possible to analyze more efficiently, build more accurate models and reduce their dimensionality. Scientific novelty. The article analyses classical statistical methods used at the first stage of data processing. Their effectiveness and necessity in medical data analysis are proved. The results of the work are demonstrated on real data processing, namely, on electrocardiogram processing. Conclusions. The application of statistical analysis methods for pre-processing medical data is considered on the example of ECG processing and analysis. The main statistical characteristics were calculated: mean, variance, standard deviation, mode, median, skewness coefficient, kurtosis coefficient, and coefficient of variation. The obtained data is used to estimate the law of data distribution, test hypotheses about the laws of distribution, and normalize data. A correlation matrix is constructed for the ECG observation matrix, eigenvalues and eigenvectors are calculated, and principal components are determined on their basis. The use of principal components makes it possible to reduce the dimensionality of the data for a deeper analysis. In this study, the amount of data was reduced by a factor of four. A discrete Fourier transform was performed. The analysis of the Fourier’s transform results made it possible to isolate high-frequency electromagnetic interference transmitted through the cable from the power supply network to the device, and the frequency of the interference was determined. Motion artefacts associated with the patient’s breathing were detected. The frequency of such interference is in the range from 1 to 4 Hz. The ECG points were classified and an ECG module was built for further analysis. The classification results made it possible to identify a set of points in the vicinity of R-peaks. This makes it possible to localize QRS complexes without using a complex mathematical apparatus. The results of medical data pre-processing allow to reduce the data dimensionality, identify the presence or absence of linear relationships, and evaluate the frequency characteristics of the data. Based on the pre-processing, it is possible to plan further research and build better models for data mining.

Real-time operation of municipal anaerobic digestion using an ensemble data mining framework

This study presents a novel approach for real-time operation of anaerobic digestion using an ensemble decision-making framework composed of weak learner data mining models. The framework utilises simple but practical features such as waste composition, added water and feeding volume to predict biogas yield and to generate an optimised weekly operation pattern to maximise biogas production and minimise operational costs. The effectiveness of this framework is validated through a real-world case study conducted in the UK. Comparative analysis with benchmark models demonstrates a significant improvement in prediction accuracy, increasing from the range of 50–80% with benchmark models to 91% with the proposed framework. The results also show the efficacy of the weekly operation pattern, which leads to a substantial 78% increase in biogas generation during the testing period. Moreover, the pattern contributes to a reduction of 71% in total days required for feeding and 30% in total days required for pre-feeding.

Passenger Flow Path Prediction Based on Urban Rail Transit AFC Data: An Example of Chengdu, China

The development of the automatic fare collection (AFC) systems provides significant support for predicting passenger flow on urban rail transit. This paper extracts passenger travel patterns using AFC data on urban rail transit in Chengdu, China, over a one-month period. Passengers are divided into two categories based on their travel habits and data mining models, and multinomial logit (MNL) models are separately used to predict their destinations. Furthermore, a two-way search algorithm is developed to search the optimal paths between origin-destination (OD) pairs by considering interchange constraints. Start a path search through the origin point and destination point, respectively, until the shortest path is found. The maximum effectiveness of a path is measured by travel time, interchange time, and the number of interchanges between the OD pairs. Finally, the validity of the proposed passenger flow path prediction method is verified by using the AFC data of Chengdu metropolitan rail transit from April 2018.

Research on the ideologic and politic education of college students based on big data analysis

Abstract This paper firstly researches the ideological and political education in higher education institutions in the background of the big data era and puts forward the innovation strategy of ideological and political education mode in higher education institutions in the big data era. Secondly, it studies the application of data mining technology in colleges and universities for ideological and political education. Based on data mining technology, the data on ideological and political education is determined by the clustering algorithm, followed by a quantitative examination of the policy text of ideological and political education of colleges and universities by calculating the quantitative statistical analysis of policy text, word frequency, and word cloud analysis. Then, the research subjects were selected, and the initial data were obtained using questionnaires, further pre-processing, and further empirical analysis of ideological and political education based on data mining technology. The consequences show that the proportions contained in the three types of score bands, namely 17.99%, 59.38% and 9.96%, basically match with the sample proportions accounted for by the three types of grades assessed by clustering, namely 30.51%, 62.02% and 8.47%, which confirms that such a model of data mining adopted according to the quantitative table of ideologic and politic education assessment in higher education institution is a very successful model. This study contributes to the innovation of online ideological and political education work in higher education institutions in the context of big data.

Vehicle Trajectory Data Mining for Artificial Intelligence and Real-Time Traffic Information Extraction

It aims to improve the efficiency of information collection and extraction in the current intelligent transportation system, and accurately mine the vehicle trajectory data By using Artificial Intelligence (AI) and Deep Learning methods, the trajectory data generated during vehicle driving are deeply mined and analyzed, and the characteristics of driving behavior of vehicle drivers are modeled and analyzed in detail. Then, a method of mining driving behavior characteristics based on Convolutional Neural Network (CNN) and vehicle trajectory is proposed. Based on the mathematical principle of wavelet packet and Least Square Support Vector Machine (LSSVM), a combined model of trajectory mining is constructed and applied to the short-term prediction of traffic flow. The traffic flow of Binjiang Road and Renmin Road in Guangzhou, Guangdong Province from August 19 to August 21, 2021 is predicted to verify the accuracy of the trajectory mining combined model. The results show that the combination model of data mining has good fitting effect, and the average accuracy is above 0.8. Besides, the effectiveness of the Deep Learning model in driver behavior classification is verified. The accuracy of the classification model is 75.2% for trajectory, and that is 76.8% for driver behavior characteristics. It is of great significance to effectively utilize the knowledge data in Intelligent Transportation System (ITS) and extract valuable information from it, which has certain reference value for the subsequent refined prediction of vehicle behavior.

Analysis of the use of pop singing in musical theater singing based on data analysis

Abstract In this paper, we first systematically sorted out the characteristics of pop singing and musical theater singing and explored the generation of emotion and melody application of pop singing in musical theater singing. Then, the gray GM(1,1) model is analyzed, and a gray data mining model based on data analysis is constructed to predict the application of pop singing in combination with the data mining model under data analysis. Finally, the prediction analysis was conducted to use characterization, timbre and style in pop singing, respectively. The results showed that the score of character building was 90, and the errors between the predicted and actual values of timbre and style were between ±0.1 and ±1, respectively, which were within an acceptable range.

Analysis of news communication strategies in the era of full media based on data mining

Abstract In this paper, we first study the representation of news text, build the content tree using the CTWE text method, and combine the word vector with the interrelationship between words in the content tree. Then, a feature word extraction technique is applied to filter key information, and Single-pass hierarchical clustering algorithm is used to classify the news content, and a data mining algorithm is applied to realize news delivery. Finally, the impact of information’s initial value and credibility on the transmission process is examined, and the model’s feasibility and practicality are assessed. The results show that the RMSE value of the data mining model is 0.0408, the Pearson correlation coefficient is 0.9334, and the cosine similarity is 0.9596, and the model in this paper has the smallest deviation from the real data and the greatest similarity compared with other models. This study confirms the unique advantages of the data mining model in news dissemination.

Sales Product Clustering Using RFM Calculation Model And K-Means Algorithm on Primskystore

The optimization of data processing procedures can yield high-quality information that is then utilized by business owners in the decision-making process. An effective company plan, particularly in the realm of promotion, holds significant importance for store proprietors in order to foster sustained business expansion. Primskystore is an e-commerce platform operating inside the retail industry, where the efficacy of product sales promotion is perceived to be suboptimal. Sales product advertising continues to revolve around a single sort of product. The objective of this work is to develop a data mining model through the implementation of a web-based application that utilizes the K-means clustering approach and the RFM model. The utilization of the K-Means clustering approach and the RFM model can facilitate the clustering of sales products at retailers. The data mining application employed on this website utilizes the Python programming language, MySQL as the database, and the Unified Modeling Language for system model creation. The findings of this research encompass the development of a web-based data mining tool designed to present the outcomes of product sales clustering within retail establishments. The system has the capability to present the outcomes of the RFM calculation model and KMeans clustering. Specifically, it reveals the presence of three distinct clusters, with cluster 0 accounting for 30% of the data, cluster 1 representing 7.5% of the data, and cluster 2 encompassing 62.5% of the data. This web-based data mining program has the potential to assist retailers in determining an optimal promotional business strategy.

Factors affecting pedestrian injury severity in pedestrian-vehicle crashes: Insights from a data mining and mixed logit model approach

There has been a concerning rise in pedestrian fatalities resulting from traffic accidents. This study conducts an in-depth analysis of the factors that contribute to the injury severity of pedestrians considering human, vehicle, roadway, and environmental characteristics of the crashes. A novel approach is proposed by combing a data mining technique and mixed logit models. First, the Apriori algorithm is employed to uncover patterns between fatal and incapacitating injury outcomes and their influencing factors. Then, mixed logit models are developed to investigate heterogeneity across all observations under two different lighting conditions. It is found that eight variables show heterogeneity affecting the injury severity of crash outcomes. Results also indicate that drivers older than 65 years old will increase the probability of pedestrian injury severity at dark-unlighted roads. Additionally, the present of signal and double yellow line has could increase the injury severity. Creating optimal lighting conditions at pedestrian crossings during nighttime hours and enhancing safety education initiatives for pedestrians are critical factors to improve pedestrian safety. The finding of this study will benefit policymakers and road safety professionals develop more effective strategies for preventing pedestrian injuries in vehicle crashes.

An approach for optimization of controllable drilling parameters for motorized bottom hole assembly in a specific formation

This study focuses on optimizing drilling parameters when using Positive Displacement Motors (PDMs). In drilling operations involving mud motors, weight-on-bit (WOB) alterations lead to variations in the system's parasitic pressure drop. Consequently, this affects the optimum flow rate and the hydraulic power of the bit. Also, if the flow rate changes, the bit's rotations per minute (RPM) also change. In other words, using PDMs creates a link between the hydraulic system and the drilling speed, such that changing drilling parameters such as the WOB causes changes in the hydraulic system's performance. Therefore, one possible way to optimize the drilling parameters is to consider the drilling rate and hydraulic system simultaneously using a multi-objective approach. This study used an integrated approach encompassing data mining and mathematical modeling, employing a multi-objective framework to identify optimal parameters. The approach was applied to Dariyan Formation drilling data. The data mining approach revealed a well-distributed data set covering optimal and suboptimal zones suitable for optimization. In data mining, the identification of optimal conditions included a WOB of 11500 lb, a rotation speed of 105.8 rev/min, and a flow rate of 843 gpm, leading to an ROP of 44.23 ft/h. In multi-objective optimization, the optimal parameters consisted of a WOB of 14480 lb, a rotation speed of 115 rev/min, and a flow rate of 920.8 gpm, resulting in an ROP of 40.49 ft/h. Comparing optimal results with the drilling data shows a substantial MSE reduction of over 35 %. The results show the good performance of this approach in detecting the optimal and non-optimal drilling variables.

Enhancing urban flood forecasting in drainage systems using dynamic ensemble-based data mining

This study presents a novel approach for urban flood forecasting in drainage systems using a dynamic ensemble-based data mining model which has yet to be utilised properly in this context. The proposed method incorporates an event identification technique and rainfall feature extraction to develop weak learner data mining models. These models are then stacked to create a time-series ensemble model using a decision tree algorithm and confusion matrix-based blending method. The proposed model was compared to other commonly used ensemble models in a real-world urban drainage system in the UK. The results show that the proposed model achieves a higher hit rate compared to other benchmark models, with a hit rate of around 85% vs 70 % for the next 3 h of forecasting. Additionally, the proposed smart model can accurately classify various timesteps of flood or non-flood events without significant lag times, resulting in fewer false alarms, reduced unnecessary risk management actions, and lower costs in real-time early warning applications. The findings also demonstrate that two features, “antecedent precipitation history” and “seasonal time occurrence of rainfall,” significantly enhance the accuracy of flood forecasting with a hit rate accuracy ranging from 60 % to 10 % for a lead time of 15 min to 3 h.

Detection and prediction of diabetes using effective biomarkers

ABSTRACT Diabetes is a prevalent and costly condition, with early diagnosis pivotal in mitigating ‎its progression and complications. The diagnostic process often contends with data ‎ambiguity and decision uncertainty, adding complexity to achieving definitive ‎outcomes. This study addresses the diabetes diagnostic challenge through data mining ‎and machine learning techniques. It involves training various machine learning ‎algorithms and conducting statistical analysis on a dataset comprising 520 patients, ‎encompassing both normal and diabetic cases, to discern influential features.‎ Incorporating 17 features as classifier inputs, this research evaluates the diagnostic ‎performance using four reputable techniques: support vector machine (SVM), random ‎forest (RF), multi-layer perceptron (MLP), and k-nearest neighbor (kNN). The outcomes ‎underscore the SVM model's superior performance, boasting accuracy, specificity, and ‎sensitivity values of 98.78±1.96%, 99.28±1.63%, and 97.32±2.45%, ‎respectively, across 50 iterations. The findings establish SVM as the preferred method ‎for diabetes diagnosis.‎ This study highlights the efficacy of data mining and machine learning models in ‎diabetes diagnosis. While these methods exhibit respectable predictive accuracy, their ‎integration with a physician's assessment promises even better patient outcomes.‎

Data Mining and Fusion Framework for In-Home Monitoring Applications.

Sensor Data Fusion (SDT) algorithms and models have been widely used in diverse applications. One of the main challenges of SDT includes how to deal with heterogeneous and complex datasets with different formats. The present work utilised both homogenous and heterogeneous datasets to propose a novel SDT framework. It compares data mining-based fusion software packages such as RapidMiner Studio, Anaconda, Weka, and Orange, and proposes a data fusion framework suitable for in-home applications. A total of 574 privacy-friendly (binary) images and 1722 datasets gleaned from thermal and Radar sensing solutions, respectively, were fused using the software packages on instances of homogeneous and heterogeneous data aggregation. Experimental results indicated that the proposed fusion framework achieved an average Classification Accuracy of 84.7% and 95.7% on homogeneous and heterogeneous datasets, respectively, with the help of data mining and machine learning models such as Naïve Bayes, Decision Tree, Neural Network, Random Forest, Stochastic Gradient Descent, Support Vector Machine, and CN2 Induction. Further evaluation of the Sensor Data Fusion framework based on cross-validation of features indicated average values of 94.4% for Classification Accuracy, 95.7% for Precision, and 96.4% for Recall. The novelty of the proposed framework includes cost and timesaving advantages for data labelling and preparation, and feature extraction.

Applying modified-data mining techniques to assess public transportation vulnerable urban and suburban city areas

To guarantee the right to move for residents in areas where public transportation is insufficient, research is needed to identify vulnerable areas and prepare measures. This paper defines the vulnerable regions of public transportation within various city types in Korea. In order to identify appropriate areas to apply the Demand Responsive Transit (DRT), the regions with vulnerability were compared with a specific city (Yangsan-si) which already the DRT system was successfully adopted. To collect monthly bus data, web-data crawling method was performed and processed with coordinating program by matching GPS coordinate. The public transportation demand was predicted for each grid cell size (100 m, 250 m, and 500 m) by different methodologies. Various data mining models based on regression were analyzed to predict bus demand of vulnerable areas. Among models, a modified model was suggested to combine Automated machine learning models for high prediction performance. The modified model outperformed other methods as 0.685 and prediction performance was appropriate at 100 m rectangle grid. Regional characters of DRT bus allocation areas were extracted by K-means clustering method and differentiate urban and suburban types. The findings of this study provide valuable insights into conditions that DRT bus stop can be installed. The urban bus stop areas located in metropolitan cities and the suburban bus stop allocation areas located in countryside. The study results can be used as policy data for the successful introduction to prevent social exclusion and improve resident welfare in the future.

An Innovative Discussion on Ideological and Political Education for College Students Based on Digital Mining

Abstract In this paper, a self-organized data mining model for college students’ ideological and political education is constructed. Firstly, a suitable input-output association set is specified to build the model, and then one or more are selected as the outer criterion to filter the model. The models of each layer are evaluated by the outer criterion, and the optimal complexity model is established by finding the model structure and the law of system operation through the selection of different models. Then unknown correlations between factors of ideological and political education were mined from the data samples, and the system variables were predicted and analyzed. Finally, the model is applied to analyze the effect of ideological education and the influencing factors of College A. The p-value of the degree of matching between education level and professional level is 0.456, and the education level will have a significant influence on ideological and political education. The research in this paper is important for the development of ideology and politics in colleges and universities.

Evidence of Students’ Academic Performance at the Federal College of Education Asaba Nigeria: Mining Education Data

One main objective of higher education is to provide quality education to its students. One way to achieve the highest level of quality in the higher education system is by discovering knowledge for prediction regarding enrolment of students in a particular course, alienation of traditional classroom teaching model, detection of unfair means used in online examination, detection of abnormal values in the result sheets of the students, and prediction about students’ performance. The knowledge is hidden among the educational data set and is extractable through data mining techniques. The present paper is designed to justify the capabilities of data mining techniques in the context of higher education by offering a data mining model for the higher education system in the university. In this research, the classification task is used to evaluate student’s performance, and as many approaches are used for data classification, the decision tree method is used here. By this, we extract data that describes students’ summative performance at semester’s end, helps to identify the dropouts and students who need special attention, and allows the teacher to provide appropriate advising/counseling.