Search

In this era, research and development are becoming a continuous and accelerating process because technology changes rapidly with a short lifecycle. As a result, various methodologies are being developed to monitor these rapidly changing research trends; In particular, clustering method-related studies in science and technology documents are being developed with a variety of approaches. However, previous studies on document clustering methods focus on a specific field or language but do not take into consideration certain important pieces of information in science and technology documents. Therefore, this study proposes an embedding methodology that uses important content from scientific and technical documents. We took into consideration the importance of information containing core structures in science and technology documents and proposed a clustering methodology that analyzes structured and unstructured data, such as textual information, author information, and citation information. The proposed method combines both textual and structural data from the paper, using a method that focuses on screening important information by sections in science and technology documents. Then, Girvan-Newman clustering and Louvain clustering models are applied to generate embedding vectors and show evaluation results through the clustering indices. As a practical example, we applied the proposed methodology using paper data from the field of hydrogen cell vehicles. The results of this study will be effective in identifying gaps in technology for new technological development, identifying technology trends, and presenting directional information for future technology development.

For organizing and analyzing massive amounts of data and revealing hidden patterns and structures, clustering is a crucial approach. This paper examines unique strategies for rapid clustering, highlighting the problems and possibilities in this area. The paper includes a brief introduction to clustering, discussing various clustering algorithms, improvements in handling various data types, and appropriate evaluation metrics. It then highlights the unsupervised nature of clustering and emphasizes its importance in many different fields, including customer segmentation, market research, and anomaly detection. This review emphasizes ongoing efforts to address these issues through research and suggests exciting directions for future investigations. By examining the advancements, challenges, and future opportunities in clustering, this research aims to increase awareness of cutting-edge approaches and encourage additional innovations in this essential field of data analysis and pattern identification. It highlights the need for resilience to noise and outliers, domain knowledge integration, scalable and efficient algorithms, and interpretable clustering technologies. In addition to managing high-dimensional data, creating incremental and online clustering techniques, and investigating deep learning-based algorithms, the study suggests future research areas. Additionally featured are real-world applications from several sectors. Although clustering approaches have made a substantial contribution, more research is necessary to solve their limitations and fully realize their promise for data analysis.

Large blocks of data must be analyzed and explored by utilizing the data mining procedures in order to uncover significant patterns and trends. Medical databases are one area where the data mining procedures can be utilized. Many people all over the world are struggling with their health and medical diagnoses. Massive amounts of data are produced by hospital information systems (HIS), yet it might be difficult to extract knowledge from diagnosis case data. By just giving the symptoms they are experiencing, patients can quickly learn about the sickness they are experiencing and the medication that can assist, treat it using the approaches utilized in this paper. In this paper, we give drug recommendations relied on ratings and conditions to customers. Four distinct prototypes are utilized to predict the diseases. The Vader tool and sentiment analysis relied on NLP are utilized to analyze the reviews. And finally, probabilistic and weighted average methodologies are utilized to recommend the medications. Each model and strategy utilized in this paper is described in detail. The experimental findings presented in this work can be utilized in future studies and for a variety of different medicinal applications.

Given the proliferation of sensors and actuators, evaluating the network performance of current and forthcoming city-wide Internet of Things (IoT) applications is a challenging task. To overcome this challenge, we created a large-scale simulator called PIoT over the years that can assess the performance of multiple millions of mobile/static IoT devices using a 4G/5G cellular infrastructure. In PIoT, different Key Performance Indicators (KPIs) are defined and collected to produce data to evaluate applications and network performance. PIoT is an on-going academic simulator project, and its most recent version is accessible to the public through the user interface found on https://www.piotsimulation.com without any installation requirements. It uses a realistic database that contains the real locations and features of Base Stations (BSs) and the real locations of IoT user equipment devices. The interface can be used by operators and researchers to understand network behavior when deploying new applications or to gather data to feed artificial intelligence and machine learning algorithms. The objective of this paper is to present a detailed description of the PIoT modeling architecture, as well as some use cases to help potential users understand the type of capabilities that are available when using the simulator. Limitations and comparisons with other popular engines are also included in the paper.

Students learning performance is one of the core components for assessing any educational systems. Students performance is very crucial in tackling issues of learning process and one of the important matters to measure learning outcomes. The ability to use data knowledge to improve education systems has led to the development of the field of research known as educational data mining (EDM). EDM is the creation of techniques to investigate data gathered from educational settings, allowing for a more thorough and accurate understanding of students and the improvement of educational outcomes for them. The use of machine learning (ML) technology has increased significantly in recent years. Researchers and teachers can use the measurements of success, failure, dropout, and more provided by the discipline of data mining in education to predict and simulate education processes. Therefore, this work presents an analysis of students performance using data mining methods. The paper presents both clustering and classification techniques to identify the impact of students performance at early stage with on the GPA. For the clustering technique, the paper uses dimensionality reduction mechanism by T-SNE algorithm with various factors at early stage such as admission scores and first level courses, academic achievement tests (AAT) and general aptitude tests (GAT) in order to explore the relationship between these factors and GPA’s. For the classification technique, the paper presents experiments on different machine learning models on predicting student performance at early stages using different features including courses’ grades and admission tests’ scores. We use different assessment metrics to evaluate the quality of the models. The results suggest that educational systems can mitigate the risks of students failures at the early stages.

In information retrieval and text mining, document clustering is a big challenge because the amount of document collection has been increasing, day by day. The problem of clustering is NP-hard, use of meta-heuristic algorithms to solve these problems could be an effective method. When the solution space is large, traditional methods are unable to find a solution in a reasonable amount of time. K-means is a heuristic clustering algorithm, two main issues with heuristic algorithms are the early convergence and trapping in local optima. Moreover, finding the right number of clusters is one of the main drawbacks of the k-means algorithm. The correct value of k is always confusing, different researchers used different methods to solve this problem. To overcome these mentioned problems, this study presents a novel Hybrid approach for document clustering. One of the challenges in existing BH algorithm is the input data type. Recently, the algorithm was only accepting textual data. Another flaw in the existing model is that it doesn’t choose how many clusters k to form automatically, and the centroids are chosen at random in it. In this paper, we have constructed a Hybrid cluster identification approach which consists of the Elbow method and Silhouette score for cluster k identification. This paper mainly offers three novel combination of model to represent text documents, namely i) K-mean++ - BH + TF-IDF with fix k ii) K-mean++ - BH + W2V with fix k iii) Hybrid Black Hole with automated k. The proposed improvements have validated on the document clustering problem. Cluster analysis based on two evaluation measures, external (Purity) and internal measures (Silhouette score) are used to report the findings. Experiments have been carried out on the four al-phanumeric datasets (Doc50, Reuters, WebKB and News20) as well as on two numeric datasets (Iris and Wine) respectively. The complete result analysis is reported in detail with respect to each research contribution to compare the performance of the proposed algorithm with existing clustering methods. Result shows that the proposed Hybrid BH algorithm outperforms better than the existing clustering methods for all datasets. The clustering of data with and without stop words is examined; additionally, the two alternative word embedding used for data exploration in conjunction with proposed model are also evaluated. In the present study, proposed Hybrid BH algorithm handles the optimal value of k efficiently. This is one of the major contributions of the paper, concluded that Hybrid Black Hole is an effective algorithm for cluster analysis.

The paper deals with the challenge of inconsistency in multichannel emotion recognition. The focus of the paper is to explore factors that might influence the inconsistency. The paper reports an experiment that used multi-camera facial expression analysis with multiple recognition systems. The data were analyzed using a multidimensional approach and data mining techniques. The study allowed us to explore camera location, occlusions and algorithm factors in the late fusion of emotion recognition results. We proposed to use a multidimensional data model for mining the various interdependencies between the factors of inconsistency. The study allowed the exploration of challenges in multichannel emotion recognition. It was achieved by comparing the consistency of obtained emotions and identification of rules determining conditions when the obtained emotions are consistent. However, the main novelty of the paper is the method of mining the inconsistencies. The study might be interesting both for researchers dealing with integration in emotion recognition, as well as for practitioners who use automatic emotion analysis software and expect to get valid results.

System logs play an important role in software development and system maintenance. Many system software programs continuously generate system logs during software runtimes for failure detection and diagnosis purposes. Currently, the analysis of system log data is mainly a manual process that highly depends on human knowledge and experience. This time-consuming task has become a problem because of the ever-increasing volume of log data. Existing studies have investigated machine learning and deep learning techniques to automate the failure detection task. This paper takes the deep learning approach and proposes two detection structures based on recurrent and convolutional neural networks. More importantly, this paper takes a step further by closely examining the timestamps of log data which existing studies have generally ignored. This study found that time information can be a distinguishing factor between regular and abnormal log sequences. Inspired by this observation, a novel method is proposed to integrate log timestamps in deep learning models using interpolation techniques. The evaluation results show that the log timestamps can significantly improve the performance of failure detection. Cross-comparison of the different models demonstrates that the proposed network structure can successfully utilize the timestamp information. The code is available on GitHub: https://github.com/hfyxin/Ts-models-log-data-analysis.git.

Artificial Intelligence (AI) has become a core feature of today’s real-world applications, making it a trending topic within the software engineering (SE) community. The rise in the availability of AI techniques encompasses the capability to make rapid, automated, impactful decisions and predictions, leading to the adoption of AI techniques in SE. With industry revolution 4.0, the role of software engineering has become critical for developing productive, efficient, and quality software. Thus, there is a major need for AI techniques to be applied to enhance and improve the critical activities within the software engineering phases. Software is developed through intelligent software engineering phases. This paper concerns a systematic mapping study that aimed to characterize the publication landscape of AI techniques in software engineering. Gaps are identified and discussed by mapping these AI techniques against the SE phases to which they contributed. Many systematic mapping review papers have been produced only for a specific AI technique or a specific SE phase or activity. Hence, to our best of knowledge within the last decade, there is no systematic mapping review that has fully explored the overall trends in AI techniques and their application to all SE phases.

Nowadays, the use of technology in continuously increasing, making a significant impact in almost every area, including education. New areas have gained much popularity in the last years in educational technology (EdTech), such as Massive Open Online Courses (MOOCs) or computer-supported collaborative learning. In addition, research and interest in this area have also been growing over the years. The quantity of research and scientific publications in EdTech is constantly increasing, and trying to analyze and extract information from a set of research papers is often a very time-consuming task. To make this process easier and solve these limitations, we present Fontana, a framework that can quickly perform trend and social network analysis using any corpus of documents and its metadata. Specifically, the framework can: 1) Discover the latest trends given any corpus of documents, using Natural Language Processing (NLP) analysis and keywords (bibliometric approach); 2) Discover the evolution of the trends previously identified over the years; 3) Discover the primary authors and papers, along with hidden relationships between existing communities. To test its functionality, we evaluated the framework using a corpus of papers from the EdTech research field. We also followed an open science methodology making the entire framework available in Open Science Framework (OSF) easy to access and use. The case study successfully proved the capabilities of the framework, revealing some of the most frequent topics in the area, such as “EDM,” “learning analytics,” or “collaborative learning.” We expect our work to help identifying trends and patterns in the EdTech area, using natural language processing and social network analysis to objectively process large amounts of research.

With the continuous improvement of modern scientific construction, urban development is moving in the direction of digitization. Under the background of digital technology of Internet of things, many cities have new challenges and opportunities in the construction of a new generation. Urban furniture design plays an important role in public areas and activity spaces. It can bring many interactive and spatial elements to people’s daily life. At the same time, the functionality and reliability of urban furniture design are also the main contents we pay attention to. How to reflect the applicability of design and evaluate the performance of products is the research content of this paper under the background of Internet of things and digitization. Firstly, it analyzes the current situation of urban furniture design and the specific needs in the digital background. This paper explores the impact of the above factors on the safety of urban furniture from the aspects of urban development, economic situation and user groups. Using Internet of things data modeling and estimation methods to analyze the data changes contained in urban furniture design. Judge the performance changes of furniture design in actual use according to the data information, and analyze the feasibility of people in use. Finally, in the big data environment, with the support of Internet of things technology, this paper studies the sharing function in urban furniture design, and evaluates the feedback effect of shared urban furniture. The results show that urban furniture design under the digital background of Internet of things reflects good feasibility and applicability in modern urban construction. Digital urban furniture has gradually become a new symbol of a city. This paper uses data mining to obtain and analyze the data in the process of urban furniture design, and realizes the standardized processing of user behavior data. It not only meets people’s daily needs and the needs of intelligent equipment, but also strengthens the importance of urban furniture design and use in modern urban construction. Compared with the traditional Internet, the Internet of things can combine the wireless transmission network with the actual objects to achieve the purpose of accurately transmitting information.

Diabetes is one of the most rapidly growing chronic diseases, which has affected millions of people around the globe. Its diagnosis, prediction, proper cure, and management are crucial. Data mining based forecasting techniques for data analysis of diabetes can help in the early detection and prediction of the disease and the related critical events such as hypo/hyperglycemia. Numerous techniques have been developed in this domain for diabetes detection, prediction, and classification. In this paper, we present a comprehensive review of the state-of-the-art in the area of diabetes diagnosis and prediction using data mining. The aim of this paper is twofold; firstly, we explore and investigate the data mining based diagnosis and prediction solutions in the field of glycemic control for diabetes. Secondly, in the light of this investigation, we provide a comprehensive classification and comparison of the techniques that have been frequently used for diagnosis and prediction of diabetes based on important key metrics. Moreover, we highlight the challenges and future research directions in this area that can be considered in order to develop optimized solutions for diabetes detection and prediction.

Smart grids are re-engineering the electricity transmission and distribution system throughout the world. It is an amalgam of increased digital information with the electrical power grids. Managing the data generated from the grid efficiently is the key to successful knowledge extraction from the smart grid big data. Most of the scientific advancements are becoming data-driven and becoming an interesting area of research for data scientists. It is challenging the world computationally enough to develop new storage methods and data processing technologies. Managing big data involves data cleaning, integration of varied data sources, and decision-making applications. This paper focuses on the study of big data management and proposes a management process to help manage the data in the grid. Data management tools and techniques have been leveraged in understanding the sources and data types in the grid. The paper emphasizes the limitations of the existing solutions inclined towards applications of the smart grid big data.

The condition of the transformer can be accurately assessed based on dissolved gas online monitoring and analysis technology. With the widespread use of Dissolved Gas Analysis (DGA) technology, a large amount of dissolved gas time series data is obtained and forms DGA big data, which brings challenges to the data selection process of traditional condition assessment methods. Inaccurate results might be generated by selecting excessive or insufficient data from a long-span dissolved gas time series. The shorter dissolved gas time series fails to fully characterize the operation law of transformers, and the longer dissolved gas time series contains redundant information that results in inaccurate analysis models and excessive calculation consumption. This paper attempts to mine the optimal data selection rule of long-span dissolved gas time series for more accurate and efficient analysis. In this paper, C-C method was employed to calculate the phase space parameters of the time series. By analyzing the convergence of the correlation integral, the boundary conditions that maintain the time series phase space stability were revealed, and so was the relationship between the optimal length and the embedding dimension. The real cases show that the optimal data set provides the same accuracy and significantly improves the computation speed by 113.54 times. The optimal data selection rule eliminates redundant information in the long-span dissolved gas time series and retains the “optimal data”, which not only preserves the characteristics of the original data, but also improves the computation speed of transformer condition assessment.

Smart control systems are mostly applied in industry to control the movements of heavy machinery while optimizing overall operational efficiency. Major shipping companies use large quay cranes to load and unload containers from ships and still rely on the experience of on-site operators to perform transportation control procedures using joysticks and visual contact methods. This paper presents the research results of an EU-funded project for the Klaipeda container terminal to develop a novel container transportation security and cargo safety assurance method and system. It was concluded that many risks arise during the container handling procedures performed by the quay cranes and operators. To minimize these risks, the authors proposed controlling the sway of the spreader using a model predictive control method which applies a multi-layer perceptron (MLP) neural network (NN). The paper analyzes current neural network architectures and case studies and provides the engineering community with a unique case study which applies real operation statistical data. Several key training algorithms were tested, and the initial results suggest that the Levenberg–Marquardt (LM) algorithm and variable learning rate backpropagation perform better than methods which use the multi-layer perceptron neural network structure.

The integration of simulation-based optimization and data mining is an emerging approach to support decision-making in the design and improvement of manufacturing systems. In such an approach, knowledge extracted from the optimal solutions generated by the simulation-based optimization process can provide important information to decision makers, such as the importance of the decision variables and their influence on the design objectives, which cannot easily be obtained by other means. However, can the extracted knowledge be directly used during the optimization process to further enhance the quality of the solutions? This paper proposes such an online knowledge extraction approach that is used together with a preference-guided multi-objective optimization algorithm on simulation models of manufacturing systems. Specifically, it introduces a combination of the multi-objective evolutionary optimization algorithm, NSGA-II, and a customized data mining algorithm, called Flexible Pattern Mining (FPM), which can extract knowledge in the form of rules in an online and automatic manner, in order to guide the optimization to converge towards a decision maker’s preferred region in the objective space. Through a set of application problems, this paper demonstrates how the proposed FPM-NSGA-II can be used to support higher quality decision-making in manufacturing.

The physical characteristics of the massive spectrum signals carrying the communication information and the statistical laws of these characteristics also potentially reflect the communication behavior of the communication individuals and the intelligence information related to the communication behavior. Intercepting and cracking signal content usually faces enormous difficulties and costs, and more often, we are not able to crack the encrypted signal content. However, by studying the physical features extracted from the spectrum monitoring signals and the statistical laws of these features, it is also possible to dig out the hidden relationships between communication individuals and even the communication network structure, so as to analyze the communication behaviors of the communication individuals. Based on the characteristics of carrier frequency, bandwidth, power, signal monitoring time and direction information of spectrum monitoring signals, this paper identifies each spectrum signal and studies the distribution characteristics and statistical laws of massive spectrum monitoring signals in the column coordinate system. Due to the clustering of the spectrum signals generated by the sources in the power, monitoring time and direction, and the correlation of the spectrum signals generated by the two parties in the communication process, based on the improved density clustering algorithm, this paper proposes a method for mining the communication relationship between communication individuals from the spectrum monitoring data, and guesses and constructs the communication network structure by matching the communication individual with the communication relationship. Finally, we analyze the communication network structure mined from the spectrum monitoring data.

Due to the rapid development of information technology and network technology, there is a lot of data, but the phenomenon of lack of knowledge is becoming more and more serious. Data mining technology has developed vigorously in this environment, and it has shown more and more vitality. Based on Spark programming model, this paper designs the parallel extension of fuzzy c-means. In order to enhance the performance of fuzzy c-means parallel expansion, the improvement strategy of k-means during the initialization phase is borrowed, and k-means// is extended to fuzzy c-means to obtain better clustering performance. Combined with Spark's programming model, this paper can obtain extended parallel fuzzy c-means algorithm. Several experiments on the data set of the algorithm proposed in this paper have shown good scalability and parallelism, effectively expanding fuzzy c-means clustering to distributed applications, greatly increasing the scale of the data processed by the algorithm. This improves the robustness of the algorithm and the adaptability of the algorithm to the shape and structure of the data, so that the parallel and scalable clustering algorithm can more effectively perform cluster analysis on big data. Three algorithms were simulated on MATLAB platform. We use simple data sets and complex two-dimensional data sets, and compare with the traditional fuzzy c-means algorithm and fuzzy c-means algorithm based on fuzzy entropy. Experiments show that the scalable parallel fuzzy c-means algorithm not only greatly improves the anti-noise performance, but also improves the convergence speed, and it can automatically determine the optimal number of clusters.

Intrusion detection technology, as an active and effective dynamic network defense technology, has rapidly become a hot research topic in the field of network security since it was proposed. However, current intrusion detection still faces some problems and challenges that affect its detection performance. Especially with the rapid development of the current network, the volume and dimension of network data are increasing day by day, and the network is full of a large number of unlabeled data, which brings great pressure on the data processing methods of IDS. In view of the tremendous pressure of intrusion detection brought by the current complex and high-dimensional network environment, this paper provides a feasible solution. Firstly, this paper briefly outlines the necessity of feature learning, the shortcomings of traditional feature learning methods and the new breakthroughs brought by deep belief network in feature learning, and focuses on the principle and working mechanism of deep belief network and Principal Component Analysis (PCA). Then, it constructs the intrusion detection model based on PCA-BP and DBN respectively. And through the experimental evaluation of the two detection models, a comparative experiment between deep belief network and principal component analysis is constructed. The experimental results show that deep belief network has unique advantages and good performance in feature learning. Therefore, deep belief network can be applied in the field of intrusion detection to extract effective features from the current high-dimensional and redundant network data, thereby improving the detection performance of IDS and its adaptability to the current complex and high-dimensional network environment.

Large scale data and predictive analytics are the most challenging tasks in the field of academic data mining. Academic libraries are a great source of information and knowledge to provide a wide range of services to meet end-user requirements. Due to the rapid changes in the educational environment and availability of huge library rental book data, it is required to utilize data mining and machine learning techniques in the context of the academic library to extract and analyze underlying knowledge from rental book data, which is important to facilitate library administration to drive better future decisions to improve and manage library resources effectively. These are the following resources, such as managing future demands of the library books, selection and arrangement of the books, operational efficiency, and also improve the quality of interaction between the library and end-users, etc. This work uses and analyzes a real dataset collected from the library of Jeju National University, the Republic of Korea. The dataset contains 2,211,413 rental book records including 173671 unique book records, 57203 unique number of the rental user, and 78 data parameters. In this paper, we propose a novel model to analyze and predict library rental book data to facilitates library administration in order to plan and manage library resources effectively and provide better services to end-users. The proposed model consists of two different modules; library data analysis and prediction modules. Firstly, we use data mining techniques to analyze and extract useful underlying patterns from library rental book data, which can lead to plan and manage library resources effectively. Secondly, a novel prediction model is proposed based on Deep Neural Network (DNN), Support Vector Regressor (SVR), and Random Forest (RF) to predict future usage of the academic libraries rental books. The performance results of the implemented regression models are evaluated in terms of MAE, MSE, and RMSE. In this paper, it is found that the DNN model performs significantly better than SVR and RF. The experimentation results show that the proposed model improves the future usage of library books to facilitate library administration to plan and manage library resources effectively. Based on the proposed model results, the academic library administration can easily plan and manage resources effectively to provide quality services to end-users.

Large datasets computing is a research problem as well as a huge challenge due to massive amounts of data that are mined and crunched in order to successfully analyze these massive datasets because they constitute a valuable source of information over different and cross-folded domains, and therefore it represents an irreplaceable opportunity. Hence, the increasing number of environments that use data-intensive computations need more complex calculations than the ones applied to grid-based infrastructures. In this way, this paper analyzes the most commonly used algorithms regarding to this complex problem of handling large datasets whose part of research efforts are focused on reducing dimensional space. Consequently, we present a novel machine learning method that reduces dimensional space in large datasets. This approach is carried out by developing different phases: merging all datasets as a huge one, performing the Extract, Transform and Load (ETL) process, applying the Principal Component Analysis (PCA) algorithm to machine learning techniques, and finally displaying the data results by means of dashboards. The major contribution in this paper is the development of a novel architecture divided into five phases that presents an hybrid method of machine learning for reducing dimensional space in large datasets. In order to verify the correctness of our proposal, we have presented a case study with a complex dataset, specifically an epileptic seizure recognition database. The experiments carried out are very promising since they present very encouraging results to be applied to a great number of different domains.

The number of academic papers being published is increasing rapidly, and recommending sufficient citations to assist researchers in writing papers is a non-trivial task. Conventional recommendation approaches may not be optimal, as the recommended papers may already be known to the users or may be solely relevant to the surrounding context but not to other concepts discussed in the manuscript. In this study, we propose a novel embedding algorithm, namely DocCit2Vec, along with the new concept of “structural context”, to address the aforementioned issues. The proposed models are compared extensively with network-based, document-based, and combined approaches in experiments of citation recommendation and classification tasks. Three implications are concluded. First, the document-based methods demonstrated overwhelmingly superior performances for citation recommendation than the network-based methods, as the latter lack consideration of the word information. Second, DocCit2Vec exhibited significant improvement for citation recommendation among the document-based methods. Third, the ability to conduct classification tasks could be significantly enhanced by adding attention layer to DocCit2Vec.

Stocks are the most active part of the securities market, and the analysis of stock generally starts from the price fluctuation. Stock trading data have the characteristics of time series, which make it possible to record the transaction prices in a time-evolving order. Due to the large data and high research complexity of time series, some ideal data are difficult to obtain for analyzing and predicting stock movements. Aiming at the problem, we utilize the piecewise linear representation which combines turning points and maximum absolute deviation points in stock time series to extract sequence features. Firstly, the proposed method finds turning points that satisfy the condition given in this paper, and defines the point distance formula to calculate the fitting errors between the subsequence segment and the fitted straight line, whose average value is set as the threshold P and the subsequence length is as the threshold d. Secondly, if the fitting error or the length of the subsequence segment is greater than P or d, respectively, the maximum absolute deviation point is obtained according to the difference between the fitted value and the original data. Finally, all trend feature points are connected by linear interpolation. In this paper, different sequence lengths, different thresholds d, different methods and industrial data from different fields are discussed and compared in detail to verify the proposed method. The experimental results show that the proposed method gets satisfactory data fitting and expanding effect, and retains the characteristics and integrity of the initial time series.

The existing atmospheric mass density models (AMDM) would produce considerable errors in orbital prediction for Low Earth Orbit (LEO) satellites. In order to reduce these errors and correct the AMDM, this paper presents methods based on data mining with historical data of two-line element (TLE). Starting from a typical LEO satellite, TIANHUI, two orbital dynamical models are firstly proposed as the simulation environment to generate training data. The historical TLE data are regarded as actual space environment and used to generate application data. Secondly, three data mining methods, Random Forest (RF), Artificial Neural Network (ANN) and Support Vector Machine (SVM), are combined with the training data to investigate their feasibility in recovering the known deviation of AMDM under simulation environment. Training results show that RF displays the best performance and achieves the accuracy of 99.99%, while the other two methods only achieve 86.83% and 71.90% respectively. Thirdly, under the actual space environment, this paper uses new training and application data to research the ability of the three methods in recovering the unknown deviation of the AMDM and improve the accuracy of orbital prediction. Numerical results are evidential to the accuracy of the proposed methods based on data mining. It is concluded that the capabilities of the data mining for correction for the atmospheric model are very promising, with great potential to advance practical applications on on-orbit propagation.

Social Media (SM) platforms, particularly Twitter, have become useful tools for startup companies (henceforth startups) which use the latter to support most of their business activities. As a result, there is a need to gauge the performance of specific business initiatives vis-à-vis public sentiment, or more specifically the spread of such initiatives based on Twitter user-generated content. Previous research which makes use of Twitter analysis to analyze the business activities of startups is minimal, especially for Twitter user content in the Arabic language. Consequently, this paper proposes an analytics-based framework called Startup Initiatives Response Analysis (SIRA) designed to assess the performance of initiatives launched by startups via text classification, sentiment analysis, and statistical analysis techniques. To provide empirical evidence for the viability of the proposed research framework, this paper examined the case of an Arab transportation network startup, carrying out a SIRA analysis of an initiative undertaken by Careem to empower women by encouraging them to work for the company. The results confirm the effectiveness of the proposed framework for statistically measuring the initiative spread and the public feedback based on the user-generated content on the Twitter social platform.

With the ever-growing demand of electric power, it is quite challenging to detect and prevent Non-Technical Loss (NTL) in power industries. NTL is committed by meter bypassing, hooking from the main lines, reversing and tampering the meters. Manual on-site checking and reporting of NTL remains an unattractive strategy due to the required manpower and associated cost. The use of machine learning classifiers has been an attractive option for NTL detection. It enhances data-oriented analysis and high hit ratio along with less cost and manpower requirements. However, there is still a need to explore the results across multiple types of classifiers on a real-world dataset. This paper considers a real dataset from a power supply company in Pakistan to identify NTL. We have evaluated 15 existing machine learning classifiers across 9 types which also include the recently developed CatBoost, LGBoost and XGBoost classifiers. Our work is validated using extensive simulations. Results elucidate that ensemble methods and Artificial Neural Network (ANN) outperform the other types of classifiers for NTL detection in our real dataset. Moreover, we have also derived a procedure to identify the top-14 features out of a total of 71 features, which are contributing 77% in predicting NTL. We conclude that including more features beyond this threshold does not improve performance and thus limiting to the selected feature set reduces the computation time required by the classifiers. Last but not least, the paper also analyzes the results of the classifiers with respect to their types, which has opened a new area of research in NTL detection.

In today's world, due to the advancement of technology, predicting the students' performance is among the most beneficial and essential research topics. Data Mining is extremely helpful in the field of education, especially for analyzing students' performance. It is a fact that predicting the students' performance has become a severe challenge because of the imbalanced datasets in this field, and there is not any comparison among different resampling methods. This paper attempts to compare various resampling techniques such as Borderline SMOTE, Random Over Sampler, SMOTE, SMOTE-ENN, SVM-SMOTE, and SMOTE-Tomek to handle the imbalanced data problem while predicting students' performance using two different datasets. Moreover, the difference between multiclass and binary classification, and structures of the features are examined. To be able to check the performance of the resampling methods better in solving the imbalanced problem, this paper uses various machine learning classifiers including Random Forest, K-Nearest-Neighbor, Artificial Neural Network, XG-boost, Support Vector Machine (Radial Basis Function), Decision Tree, Logistic Regression, and Naïve Bayes. Furthermore, the Random hold-out and Shuffle 5-fold cross-validation methods are used as model validation techniques. The achieved results using different evaluation metrics indicate that fewer numbers of classes and nominal features will lead models to better performance. Also, classifiers do not perform well with imbalanced data, so solving this problem is necessary. The performance of classifiers is improved using balanced datasets. Additionally, the results of the Friedman test, which is a statistical significance test, confirm that the SVM-SMOTE is more efficient than the other resampling methods. Moreover, The Random Forest classifier has achieved the best result among all other models while using SVM-SMOTE as a resampling method.

In this paper, we present churn prediction techniques that have been released so far. Churn prediction is used in the fields of Internet services, games, insurance, and management. However, since it has been used intensively to increase the predictability of various industry/academic fields, there is a big difference in its definition and utilization. In this paper, we collected the definitions of churn used in the fields of business administration, marketing, IT, telecommunications, newspapers, insurance and psychology, and described their differences. Based on this, we classified and explained churn loss, feature engineering, and prediction models. Our study can be used to select the definition of churn and its associated models suitable for the service field that researchers are most interested in by integrating fragmented churn studies in industry/academic fields.

This paper considers heuristic approaches that can be used to assign stock keeping units (SKU) to individual slots in distribution center. Firstly, we propose two novel strategies, slot selection and frequent itemset grouping. The former is used to find the most suitable slot for a single SKU, while the latter is for sequencing SKUs in an appropriate order. Secondly, we develop several storage location assignment heuristics by applying the two strategies. Especially, the heuristics are designed to assign frequently ordered SKU to a slot close to I/O (input/output) point and SKUs frequently ordered together to slots close to each other. Consequently, the proposed heuristics are helpful to reduce the travel distance of order picker in distribution center. In this paper, travel distance of order picker is calculated based on a routing policy that enables order picker to move along a flexible route. For illustration, we applied the heuristics to real data collected from a large distribution center. The experiment results reveal that slot selection strategy is very helpful to reduce average travel distance of order picker, especially under greedy routing policy. Also, frequent itemset grouping strategy can provide additional reduction in average travel distance if it is applied together with slot selection strategy.

Interactive learning plays a key role in extending the occupant behavior implementation toward smart buildings. Efficient feedbacks can be obtained from the end user by involving occupants and increasing their awareness about energy systems. Working in highly energy-efficient buildings can be a great opportunity, but users need to feel empowered. This means making them aware of the building features and allowing them to manage some of the appliances. In this way, disorientation or annoyance is avoided, and people feel more in control. This paper proposes a solution to interact with occupants to estimate the number of occupants. A novel way of supervised learning is analyzed to estimate the occupancy in a room where actual occupancy is interactively requested to occupants when it is the most relevant to limit the number of interactions. Occupancy estimation algorithm relies on machine learning; it uses information gathered from occupants with the measurements collected from common sensors, for instance, motion detection, power consumption, and CO2 concentration. Two different classifiers are investigated for occupancy estimation with interactions: a decision tree C4.5 and a parameterized rule-based classifier. In this paper, the question of when interacting with occupants is investigated. This approach avoids the usage of a camera to determine the actual occupancy. A complete real-time interaction environment has been developed and is used to estimate the occupancy in an office case study. The graphical user interface has been designed to carry out a real-time experiment.

This paper reviews the use of outlier detection approaches in urban traffic analysis. We divide existing solutions into two main categories: flow outlier detection and trajectory outlier detection. The first category groups solutions that detect flow outliers and includes statistical, similarity and pattern mining approaches. The second category contains solutions where the trajectory outliers are derived, including off-line processing for trajectory outliers and online processing for sub-trajectory outliers. Solutions in each of these categories are described, illustrated, and discussed, and open perspectives and research trends are drawn. Compared to the state-of-the-art survey papers, the contribution of this paper lies in providing a deep analysis of all the kinds of representations in urban traffic data, including flow values, segment flow values, trajectories, and sub-trajectories. In this context, we can better understand the intuition, limitations, and benefits of the existing outlier urban traffic detection algorithms. As a result, practitioners can receive some guidance for selecting the most suitable methods for their particular case.

Feature selection has been an important research area in data mining, which chooses a subset of relevant features for use in the model building. This paper aims to provide an overview of feature selection methods for big data mining. First, it discusses the current challenges and difficulties faced when mining valuable information from big data. A comprehensive review of existing feature selection methods in big data is then presented. Herein, we approach the review from two aspects: methods specific to a particular kind of big data with certain characteristics and applications of methods in classification analysis, which are significantly different to the existing review work. This paper also highlights the current issues of feature selection in big data and suggests the future research directions.

As an unsupervised learning technique, clustering can effectively capture the patterns in a data stream based on similarities among the data. Traditional data stream clustering algorithms either heavily depend on some prior knowledge or predefined parameters while the characteristics of real-time data are considered unknown. Besides, the user-specified threshold is used to overcome the effect of outliers and noises, which significantly affects the clustering performance. The overlap among clusters is another major challenge for the existing stream clustering methods. These constraints strongly limit their real-time applications. In this paper, a two-phase stream clustering algorithm based on fitness proportionate sharing is proposed. It handles streaming data when prior knowledge is not available and maps the clustering problem into a multimodal optimization problem. It introduces a density-based objective function and adopts the fitness proportionate sharing strategy to perform a more effective search for the cluster centers. To capture the dynamic characteristics of streaming data, a recursive formula for the lower bound of the density function is derived, and a summary of historical data is established for the proposed algorithm. The proposed algorithm is applied to different data sets, and a comprehensive comparison between the proposed algorithm and five well-known data stream clustering algorithms in the literature is provided. Results show comparable or better performance relative to five popular data stream clustering algorithms. A scalability analysis of the proposed streaming clustering method is presented in this paper as well.

The communication relationship can reflect the behavior relationship between different communication targets. The in-depth analysis of the communication relationship can obtain the behaviors of communication individuals, and speculate their hierarchical positions in the communication network, so as to provide a basis for further speculation on the structure of the communication network. For massive spectrum signals, we can also obtain important information such as communication relationships and behaviors of communication individuals, without cracking the signal content, but by analyzing the physical characteristics and statistical laws of the spectrum signals. In order to overcome the difficulties and costs of analyzing communication behaviors from cracking the signal content in existing research, this paper studies the physical characteristics and statistical laws of spectrum signals based on the features of frequency hopping period, average power and time of signal occurrence. Because the spectrum signals generated by the communication individuals show clustering characteristics, this paper proposes a communication relationship mining method based on improved DBSCAN (Density-Based Spatial Clustering of Applications with Noise). The method can accurately discover the communication relationship of the radio station from the incomplete spectrum monitoring data, without cracking the content carried by the spectrum signals, which provides a new idea for the mining and analysis of mass spectrum monitoring signals.

As one of the most important techniques in data mining, clustering has always been highly concerned. Most clustering algorithms have encountered challenges, such as the difficulty of cluster centers selection, the artificial determination of the number of clusters K, low accuracy of clustering, and uneven clustering efficiency of different data sets. Considering the difficulty of cluster centers chosen, a new algorithm of fast selecting the initial cluster centers is proposed in this paper. Generally, cluster centers are those data points with higher density, smaller radius threshold and far away from each other, this method uses MNN (M nearest neighbors), density and distance to determine the initial cluster centers. First, the neighborhood radius r of each point is measured by MNN based on distance, and the average value of all r is marked as r̅; second, the densities ρ of each point in the region within r̅ are calculated; and then, factor f is defined to describe the probability that points become cluster centers, based on which, the initial cluster centers are determined by the candidates with bigger f . In the end, the method proposed in this paper is tested by using 12 groups of typical benchmark data sets and applied in the stellar spectral data of LAMOST survey. The experiment results compared with the other six algorithms indicate that the initial cluster centers obtained by this method are of higher quality than that of the six algorithms. Meanwhile, the initial cluster centers of spectral data are of good agreement with the actual stellar classifications.

The arrival of Information Age, with its rapid development of information technology, has provided a wide space for Data Analysis and Mining. Yet growth in this market could be held back by privacy concerns. This paper addresses the problem of secure association rule mining where transactions are distributed across sources. The existing solutions for distributed data (vertical partition and horizontal partition) have high complexity of encryption and incomplete definition of attributes of multiple parties. In this paper, we study how to maintain differential privacy in distributed databases for mining of association rules without revealing each party's raw transactions despite how strong background knowledge the attackers have. We use a intermediate server for data consolidation without assuming it is safe. Our methods offer enhanced privacy against various attacks model. In addition, it is simpler and is significantly more efficient in terms of communication rounds and computation overhead.

Current artificial intelligence models for evaluating goaf stability in underground metal mines need a large amount of sample data for training, and their accuracy declines with small sample size. With the aim of solving this problem, this paper proposes an improved TrAdaBoost algorithm based on transfer learning theory. The scope of the TrAdaBoost algorithm is extended from the two-level classification to multi-level classification problems, which makes it suitable for evaluating goaf stability. The isolated forest method is used to filter the bad points of the auxiliary training set, thereby eliminating the interference of abnormal data. The dynamic factor concept is introduced to solve the problem that the weight of source domain data decreases too quickly and irreversibly, and this enhances the generalization performance of the algorithm for different goaf samples. To test the accuracy of the proposed model in predicting goaf stability, an evaluation model is constructed and the performance compared with other algorithms in current use. The prediction accuracy and generalization ability of the model are evaluated by mean square error and F1 measurements, which prove that the performance of the model is excellent. The most obvious finding to emerge from this paper is that, with suitable improvements, goaf evaluation models can still maintain a high level of accuracy with small sample size.

Hybrid Anomaly Detection Model (HADM) is a platform that filters network traffic and identifies malicious activities on the network. The platform applies data mining techniques to tackle effectively the security issues in high load communication networks. The platform uses a combination of linear and learning algorithms combined with protocol analyzer. The linear algorithms filter and extract distinctive attributes and features of the cyber-attacks while the learning algorithms use these attributes and features to identify new types of cyber-attacks. The protocol analyzer in this platform classifies and filters vulnerable protocols to avoid unnecessary computation load. The use of linear algorithms in conjunction with learning algorithms and protocol analyzer allows the HADM to achieve improved efficiency in terms of accuracy and computation time to detect cyber-attacks over existing solutions. While authors' previous paper evaluated HADM efficiency (accuracy and computation time) against related studies, this paper, concentrates on HADM robustness and scalability. For this purpose, five datasets, including ISCX-2012, UNSW-NB15 Jan, UNSW-NB15 Feb, ISCX-2017, and MAWILab-2018, with various size and diverse attacks have been used. Different feature selection methods are applied to find the best features. The feature selection methods are selected based on the algorithms' computation time and detection rate. The best algorithms are then selected through a benchmark on applied datasets and based on the metrics such as cross-entropy loss, precision, recall, and computation time. The result of HADM platform shows robustness and scalability against datasets with different size and diverse attacks.

Data mining technology can be used to dig out potential and valuable information from massive data, and support vector machine (SVM) is one of the most widely used and most efficient methods in the field of data mining classification. However, the training set data often contains sensitive attributes, and the traditional training method of SVM reveals the individual privacy information. In view of the low prediction accuracy and poor versatility of the existing SVM classifiers with privacy protection, this paper proposed a new SVM training method for differential privacy protection. The algorithm first solved the dual problem of SVM by using SMO method and the difference Ei between the estimated value and the real value for each support vector was recorded. Then the ratio of the Ei of each support vector to the sum of the Ei of all the support vectors was calculated. Next, different levels of Laplace random noise were added to the corresponding dual variables αi of each support vector to be released, according to the ratio of each support vector. According to the principle of differential privacy protection, the algorithm meets ε-differential privacy which can be used to effectively protect individual privacy. Experimental results on real datasets showed that the algorithm proposed in this paper could be used for classification prediction under a reasonable privacy budget.

Underground strata are reflected in various information sources in petroleum exploration, including good logging and drilling data. Real-time measurement parameters obtained from mud logging can provide data support for the early discovery of oil and gas resources and the prevention of safety accidents. It plays a forward-looking role in the drilling process. In this paper, we aim at the defection of fuzzy and random characteristics of the big data of drilling element parameters in the current drilling process. A new method named grey wolf optimization-support vector machine (GWO-SVM) is proposed by analyzing the relationship between logging data and formation to solve the serious problem of formation misjudgment. Using element content and Gamma-ray value, data mining is performed by a large number of real-time data obtained from the drilling site. The obtained information is used for comprehensive estimation and prediction of strata. First, the data is normalized, and then, the best ζ and σ values are found through the optimization of gray wolf algorithm, next the SVM training is carried out, finally, the formation prediction model is established, and the error analysis of the results was conducted. In the paper, the algorithm model is subsequently applied to three actual wells. The GWO-SVM model based on drilling data is used to predict the formation, and the error analysis showed that the error range of the GWO-SVM algorithm is within 10%. Compared with the GWO-SVM, the model accuracy of SVM, Particle Swarm OptimizationSupport Vector Machine (PSO-SVM) algorithm is lower 53% and 23%, respectively. The GWO-SVM has higher robustness, reliability, and achieves faster convergence speed, stronger generalization effect, and improves the identification accuracy of elements for the formation. The average accuracy of the GWOSVM in stratum dynamic identification is 93.5%. This model is implemented to support the logging system to improve application strength.

Enterprise Architecture (EA) is defined as a set of principles, methods, and models that support the design of organizational structures, expressing the different concerns of a company and its IT landscape, including processes, services, applications, and data. One role of EA management is to automate modeling tasks and maintain up-to-date EA models while reality changes. However, EA modeling still relies primarily on manual methods. Contributing to EA modeling automation, EA Mining is an approach that uses data mining techniques for EA modeling and management. It automatically captures existing information in operational databases to generate architectural models and views. This paper presents an ontology for EA Mining that focuses on generating architectural models from API gateway log files. An ontology defines the concepts and relationships among them to uniquely describe a domain of interest and specify the meaning of the terms. API Gateways are information technology components that serve as a facade between information systems and enterprise business partners. The ontology development methodology followed the SABiO process, whereas the Unified Foundational Ontology provided the foundations of the ontology and OntoUML, the ontology modeling language. An experiment in an e-commerce application scenario was conducted to evaluate the theoretical feasibility and applicability of the ontology. Automatic semantic and syntactic validation tools and semi-structured expert interviews were used to confirm the desired ontology properties. This study aims to contribute to the evolution of the knowledge base of EA Management.

Spatio-temporal trajectory is a movement of an object in space over a certain time period, represented by a series of nodes composed of geospatial location and corresponding timestamp. A large amount of spatio-temporal trajectory data is being gathered through various trajectory acquiring devices by tracking the movement of objects such as people, animals, vehicles and natural events. Various trajectory data mining techniques have been proposed to discover useful patterns to understand the behaviour of spatio-temporal trajectories. One unexplored pattern is to identify potential contacts of targeted trajectories which can be defined as contact mining, that is useful for many applications. One such example would be to identify potential victims from known infected humans or animals, especially when the victims are asymptomatic in a rapid spread of infectious disease environments. Another one would be to identify individuals who have been close contacts with known terrorist networks or law breakers. This paper proposes a robust contact mining framework to efficiently and effectively mine contacts of multiple trajectories-of-interest from a given set of spatio-temporal trajectories. Experimental results demonstrate the efficiency, effectiveness and scalability of our approach. In addition, parameter sensitivity analysis reveals the robustness and insensitivity of our framework.

Frequent itemset mining (FIM) is a highly resource-demanding data-mining task fundamental to numerous data-mining applications. Support calculation is a frequently performed computation-intensive operation of FIM algorithms, whereas storing transactional data is memory-intensive. The FIM is even more resource-hungry for dense data than for sparse data. The rapidly growing size of datasets further exacerbates this situation and necessitates the design of out-of-the-box highly efficient solutions. This paper proposes a novel approach to frequent itemset mining for dense datasets. This approach, after the initial stage, does not use transactional data, which makes it memory efficient. It also replaces processing-intensive support calculations with efficient support predictions, which are probabilistic and need no transactional data. To predict the support of an itemset, it only needs the support of its subsets. However, this technique works only for itemsets of size three or higher. We also propose an FIM algorithm ProbBF, which incorporates this technique. The ProbBF discards transactional data after it uses it to calculate frequent one and two-size itemsets. For the itemsets of size k, where $k\ge 3$ , ProbBF uses the proposed probabilistic technique to predict their support. It is considered frequent if the predicted support is greater than a given threshold. Our experiments show that ProbBF is efficient in both time and space against state-of-the-art FIM algorithms that use transactional data. The experiments also show that ProbBF can successfully generate the majority of the frequent itemsets on real-world datasets. Since ProbBF is probabilistic, some loss in quality is inevitable.

This paper explores the analysis of user behavior in information systems through audit records, creating a behavior model represented as a graph. The model captures actions over a specified period, facilitating real-time comparison to identify insider threats exploring anomalies detected in behavior models. “e-StepControl,” developed by “ABC software” Ltd., incorporates this approach for monitoring user behavior in different business environments. The study proposes enhancing this solution with automatic user clustering, achieved by grouping individuals exhibiting similar behavior patterns using AI/ML algorithms. The research evaluates various clustering methods, discussing their suitability for grouping users based on their behavior. The subsequent step involves leveraging user class behavior models to identify anomalies by comparing an individual’s actions with the behavior model expected in their specific user group. This extension aims to enhance the system’s ability to detect potentially malicious activities, providing data security administrators with timely alerts in case of deviations from typical behavior.

Decentralized finance (DeFi) offers a range of financial instruments and services that leverage the capabilities of web3 technology. Maker protocol, which enables users to obtain loans backed by cryptocurrencies, is one of them. Unlike traditional banks, Maker’s data is transparently recorded on the Ethereum blockchain. In this research paper, we focus on analyzing the lending aspect of Maker from a traditional finance perspective. To achieve this, we create a unique dataset with loan portfolios from the MakerDAO project, making it the first dataset of its kind in the DeFi field. This publicly available dataset contains essential financial characteristics related to borrowing, including balance, loss given default, annual equivalent rate, and probability of default. Additionally, we develop a specialized mathematical model tailored specifically to this project. This model allows us to estimate the probability of default by considering the presence of crypto-collateral and utilizing Brownian motion passage levels. The results of this study provide valuable insights into lending practices in DeFi projects. They also help bridge the gap between traditional finance and blockchain-based financial services.

The mining and metal processing industries are undergoing a transformation through digitization, with sensors and data analysis playing a crucial role in modernization and increased efficiency. Vibration sensors are particularly important in monitoring production infrastructure in metal processing plants. This paper presents the installation of vibration sensors in an actual industrial environment and the results of spectral vibration data analysis. The study demonstrates that vibration sensors can be installed in challenging environments such as metal processing plants and that analyzing vibration patterns can provide valuable insights into predicting machine failures and different machine states. By utilizing dimensionality reduction and dominant frequency observation, we analyzed vibration data and identified patterns that are indicative of potential machine states and critical events that reduce production throughput. This information can be used to improve maintenance, minimize downtime, and ultimately enhance the production process’s overall efficiency. This study highlights the importance of digitization and data analysis in the mining and metal processing industries, particularly the capability not only to predict critical events before they impact production throughput and take action accordingly but also to identify machine states for legacy equipment and be part of retrofitting strategies.

This paper proposes a variable-length Differential Evolution for Association Rule Mining. The proposed algorithm includes a novel representation of individuals, which can encode both numerical and discrete attributes in their original or absolute complement of the original intervals. The fitness function used is comprised of a weighted sum of Support and Confidence Association Rule Mining metrics. The proposed algorithm was tested on fourteen publicly available, and commonly used datasets from the UC Irvine Machine Learning Repository. It is also compared to the nature inspired algorithms taken from the NiaARM framework, providing superior results. The implementation of the proposed algorithm follows the principles of Green Artificial Intelligence, where a smaller computational load is required for obtaining promising results, and thus lowering the carbon footprint.