Data Mining Problems Research Articles

A secure Multi-Frequency Computation Protocol in 2-Part Fully Distributed Setting

Secure frequency computation protocols have been researched and applied to address various privacy-preserving problems in machine learning, data mining, and data analysis. However, these protocols only allow for the calculation of a single frequency value in one execution. In this paper, we propose a protocol for calculating multiple secure frequency values concurrently for the 2PFD data model, which is a relatively new distributed data model that has appeared in many practical problems but has not received much attention. The proposed protocol not only maintains the accuracy of the output results and a high level of security but also exhibits good performance.

On Reporting Durable Patterns in Temporal Proximity Graphs

Finding patterns in graphs is a fundamental problem in databases and data mining. In many applications, graphs are temporal and evolve over time, so we are interested in finding durable patterns, such as triangles and paths, which persist over a long time. While there has been work on finding durable simple patterns, existing algorithms do not have provable guarantees and run in strictly super-linear time. The paper leverages the observation that many graphs arising in practice are naturally proximity graphs or can be approximated as such, where nodes are embedded as points in some high-dimensional space, and two nodes are connected by an edge if they are close to each other. We work with an implicit representation of the proximity graph, where nodes are additionally annotated by time intervals, and design near-linear-time algorithms for finding (approximately) durable patterns above a given durability threshold. We also consider an interactive setting where a client experiments with different durability thresholds in a sequence of queries; we show how to compute incremental changes to result patterns efficiently in time near-linear to the size of the changes.

Are simple technical trading rules profitable in bitcoin markets?

This paper examines the profitability of simple technical trading rules in bitcoin markets comprehensively, by taking into account realistic investor behaviour and transaction costs, and data mining problems. Realistic investor behaviour is replicated by first employing 75,360 simple technical trading rules, divided over 6 commonly used trading rule classes and daily and intraday frequencies. Next, we select the best performing rules after transaction costs using a multiple hypothesis procedure. Finally, we form portfolios combining the selected rules and analyse their out-of-sample performance. We find that, especially risk–return wise, simple technical trading rules can outperform a buy-and-hold strategy in the bitcoin market out-of-sample.

High utility itemsets mining based on hybrid harris hawk optimization and beluga whale optimization algorithms

The commonly used high utility itemsets mining method for massive data is the intelligent optimization algorithm. In this paper, the WHO (Whale-Hawk Optimization) algorithm is proposed by integrating the harris hawk optimization (HHO) algorithm with the beluga whale optimization (BWO) algorithm. Additionally, a whale initialization strategy based on good point set is proposed. This strategy helps to guide the search in the initial phase and increase the diversity of the population, which in turn improve the convergence speed and algorithm performance. By applying this improved algorithm to the field of high utility itemsets mining, it provides new solutions to optimization problems and data mining problems. To evaluate the performance of the proposed WHO, a large number of experiments are conducted on six datasets, chess, connect, mushroom, accidents, foodmart, and retail, in terms of convergence, recall rates, and runtime. The experimental results show that the convergence of the proposed WHO is optimal in five datasets and has the shortest runtime in all datasets. Compared to PSO, AF, BA, and GA, the average recall rate in the six datasets increased by 32.13%, 49.95%, 12.15%, and 16.24%, respectively.

Application of support vector machine system introducing multiple submodels in data mining

As the information age develops, the scale and form of data become more diverse and diverse. Therefore, people need to use effective means to process information. For large-scale data mining problems, a clustering-based kernel matrix inner product filtering method is introduced to decompose the original quadratic programming problem into multiple sub-problems to support parallel training. And a Spark-based multiple submodels parallel support vector machine is proposed. By introducing open-source tools such as OpenCV, image feature extraction can be performed on large-scale video data. Finally, combined with the designed parallel support vector machine algorithm, video facial and expression recognition is carried out. These experiments confirmed that the research method achieved a maximum acceleration ratio of 2090 times when processing Covtype datasets. The research model could achieve an accuracy of over 99 %. Under the maximum data scale experiment, the research model improved prediction accuracy by 21 percentage points with an acceptable additional time cost of about 4 min only. Task parallel processing could more fully utilize cluster performance, increasing by approximately 3.5 m/s2 from 30 to 150 cores. The research model had the highest recognition accuracy for facial expressions, further demonstrating the effectiveness and superiority of this method. The research method has improved the efficiency of big data analysis and mining, and is of great significance in parallel analysis of video data.

An Optimal Transport View for Subspace Clustering and Spectral Clustering

Clustering is one of the most fundamental problems in machine learning and data mining, and many algorithms have been proposed in the past decades. Among them, subspace clustering and spectral clustering are the most famous approaches. In this paper, we provide an explanation for subspace clustering and spectral clustering from the perspective of optimal transport. Optimal transport studies how to move samples from one distribution to another distribution with minimal transport cost, and has shown a powerful ability to extract geometric information. By considering a self optimal transport model with only one group of samples, we observe that both subspace clustering and spectral clustering can be explained in the framework of optimal transport, and the optimal transport matrix bridges the spaces of features and spectral embeddings. Inspired by this connection, we propose a spectral optimal transport barycenter model, which learns spectral embeddings by solving a barycenter problem equipped with an optimal transport discrepancy and guidance of data. Based on our proposed model, we take advantage of optimal transport to exploit both feature and metric information involved in data for learning coupled spectral embeddings and affinity matrix in a unified model. We develop an alternating optimization algorithm to solve the resultant problems, and conduct experiments in different settings to evaluate the performance of our proposed methods.

Spatio-Temporal Pivotal Graph Neural Networks for Traffic Flow Forecasting

Traffic flow forecasting is a classical spatio-temporal data mining problem with many real-world applications. Recently, various methods based on Graph Neural Networks (GNN) have been proposed for the problem and achieved impressive prediction performance. However, we argue that the majority of existing methods disregarding the importance of certain nodes (referred to as pivotal nodes) that naturally exhibit extensive connections with multiple other nodes. Predicting on pivotal nodes poses a challenge due to their complex spatio-temporal dependencies compared to other nodes. In this paper, we propose a novel GNN-based method called Spatio-Temporal Pivotal Graph Neural Networks (STPGNN) to address the above limitation. We introduce a pivotal node identification module for identifying pivotal nodes. We propose a novel pivotal graph convolution module, enabling precise capture of spatio-temporal dependencies centered around pivotal nodes. Moreover, we propose a parallel framework capable of extracting spatio-temporal traffic features on both pivotal and non-pivotal nodes. Experiments on seven real-world traffic datasets verify our proposed method's effectiveness and efficiency compared to state-of-the-art baselines.

Choosing allowability boundaries for describing objects in subject areas

Anomaly detection is one of the most promising problems for study and can be used as independent units and preprocessing tools before solving any fundamental data mining problems. This article proposes a method for detecting specific errors with the involvement of experts from subject areas to fill knowledge. The proposed method about outliers hypothesizes that they locate closer to logical boundaries of intervals derived from pair features, and the interval ranges vary in different domains. We construct intervals leveraging pair feature values. While forming knowledge in a specific field, a domain specialist checks the logical allowability of objects based on the range of the intervals. If the objects are logical outliers, the specialist ignores or corrects them. We offer the general algorithm for the formation of the database based on the proposed method in the form of a pseudo-code, and we provide comparison results with existing methods.

Empirical Evaluation of the Time Series Forecasting Method by Combining ARIMA with RBFNN under the Additive Model

Time series data is a series of values observed through repeated measurements at different times. Time series data is a type of data present in almost all different fields of life. Time series prediction is an significant problem in time series data mining. Accurate forecasting is crucial to support decision making in many areas of life. Therefore, improving the precision of time series predicting is a interesting mission for experts in this field. Many models for predicting time series have been proposed from traditional time series models as Auto Regressive Integrated Moving Average (ARIMA) model to artificial neural network (ANN) models. ARIMA is a linear model therefore it can only take the linear characteristics in time series. In contrast, Radial Basis Function Neural Network (RBFNN) is a non-linear model therefore it can not predict effectively seasonal or trend changes in time series. To combine the strengths of these two models, in this study, we experimentally evaluate the hybrid method between ARIMA and RBFNN on real time series data from different fields. Experimental results demonstrate that the combined method outperforms each model used individually in terms of accuracy.

A literature survey on various aspect of class imbalance problem in data mining

A literature survey on various aspect of class imbalance problem in data mining

Anti-Money Laundering by Group-Aware Deep Graph Learning

Anti-money laundering (AML) is a classical data mining problem in finance applications. As well known, money laundering (ML) is critical to the effective operation of transnational and organized crime, which affects a country's economy, government, and social wellbeings. Financial services organizations facilitate the movement of money and have been enlisted by governments to assist with the detection and prevention of money laundering, which is a key tool in the fight to reduce crime and create sustainable economic development. In the application of AML, user identity and financial behavior data are widely used to detect laundering transactions. In recent years, an increasing number of money laundering activities have been conducted by organized criminal gangs while most existing works still treat the actions of each account as independent identity behavior without considering the group-level conspired interactions. Therefore, in this paper, we propose a group-aware deep graph learning-based approach for organized money-laundering detection. In particular, we design a community-centric encoder to represent the nodes and attributes in user transaction graphs and derive the adjacent gang behaviors. Then, we devise a scheme of local enhancement to accommodate nodes with similar transaction features, which are aggregated into gangs for downstream detection. Extensive experiments on the real-world dataset from one of the largest bank card alliances worldwide show that our proposed method outperforms state-of-the-art methods in both offline and online modes, showing the effectiveness of money laundering detection with group-aware deep graph learning.

Performance Analysis of Sketching Methods

One of the problems in data mining or machine learning is the high-dimensional dataset. MinHash can generate sketches of sparse datasets efficiently reducing the dimension to a few thousand. These sketches, then, can be used for different machine-learning applications. It takes O(kd) computations to generate k hash values for a data point with d non-zeros. The Sketch of a data point is the vector of k hash values. Weighted Minwise Hashing is another method to generate a sketch of size k in O(kp/d) computations. Here, p is the size of the universal set. Optimal Densification is the most efficient and accurate method as it can generate k hash values in mere O(d + k) computations. In this paper, we investigate the performance of Optimal Densification, Weighted Minwise Hashing, and Vanilla MinHash by performing two experiments. Firstly, we investigate Jaccard similarity estimation accuracy on six different synthetic datasets. Then, we perform one nearest neighbor classification (1NN) of four real datasets. Optimal Densification outperforms both Weighted Minwise Hashing and Vanilla MinHash in terms of accuracy and time taken to generate the sketches.

Consistent Matrix: A Feature Selection Framework for Large-Scale Datasets

Large-scale data processing based on limited computing resources has always been a difficult problem in data mining, where feature selection is often used as an effective data compressing mechanism. For granular computing of big data, discernibility matrix and dependency degree are the most representative methods for matrix-based and feature importance degree-based feature selection, respectively. However, their temporal and space complexities are high and often lead to poor performance. In this paper, a novel feature selection framework for large scale data processing with linear complexities was proposed for the first time. Firstly, a much more concise fuzzy granule set, called fuzzy arithmetic covering, was introduced to reduce computational costs. Then a new matrix-based feature selection framework, namely consistent matrix, was proposed for general rough set models. As a result, a heuristic attribute reduction algorithm, i.e., HARCM, was designed accordingly. Compared with six state-of-the-art algorithms for feature selection, the average running time of the newly proposed algorithm was reduced up to 2913 times, with a comparable or even better classification performance.

Discovering Interesting Patterns from Hypergraphs

A hypergraph is a complex data structure capable of expressing associations among any number of data entities. Overcoming the limitations of traditional graphs, hypergraphs are useful to model real-life problems. Frequent pattern mining is one of the most popular problems in data mining with a lot of applications. To the best of our knowledge, there exists no flexible pattern mining framework for hypergraph databases decomposing associations among data entities. In this article, we propose a flexible and complete framework for mining frequent patterns from a collection of hypergraphs. To discover more interesting patterns beyond the traditional frequent patterns, we propose frameworks for weighted and uncertain hypergraph mining also. We develop three algorithms for mining frequent, weighted, and uncertain hypergraph patterns efficiently by introducing a canonical labeling technique for isomorphic hypergraphs. Extensive experiments have been conducted on real-life hypergraph databases to show both the effectiveness and efficiency of our proposed frameworks and algorithms.

CREDIBILISTIC ROBUST ONLINE FUZZY CLUSTERING IN DATA STREAM MINING TASKS

Context. The task of clustering-classification without a teacher of data arrays occupies an important place in the general problem of Data Mining, and for its solution there exists currently many approaches, methods and algorithms. There are quite a lot of situations where the real data to be clustered are corrupted with anomalous outliers or disturbances with non-Gaussian distributions. It is clear that “classical” methods of artificial intelligence (both batch and online) are ineffective in this situation. The goal of the paper is to develop a credibilistic robust online fuzzy clustering method that combines the advantages of credibilistic and robust approaches in fuzzy clustering tasks.
 Objective. The goal of the work is online credibilistic fuzzy clustering of distorted data, using of credibility theory in data stream mining.
 Method. The procedure of fuzzy clustering of data using credibilistic approach based on the use of both robust goal functions of a special type, insensitive to outliers and designed to work both in batch and its recurrent online version designed to solve Data Stream Mining problems when data are fed to processing sequentially in real time.
 Results. Analyzing the obtained results overall accuracy of clustering methods and algorithm, proposed method similar with result of credibilistic fuzzy clustering method, but has time superiority regardless of the number observations that fed on clustering process.
 Conclusions. The problem of fuzzy clustering of data streams contaminated by anomalous non-Gaussian distributions is considered. A recurrent credibilistic online algorithm based on the objective function of a special form is introduced, which suppresses these outliers by using the hyperbolic tangent function, which, in addition to neural networks, is used in robust estimation tasks. The proposed algorithm is quite simple in numerical implementation and is a generalization of some well-known online fuzzy clustering procedures intended for solving Data Stream Mining problems.

Intelligent Classification and Diagnosis of Diabetes and Impaired Glucose Tolerance Using Deep Neural Networks

Time series classification is a challenging and exciting problem in data mining. Some diseases are classified and diagnosed based on time series. Such is the case for diabetes mellitus, which can be analyzed based on data from the oral glucose tolerance test (OGTT). Prompt diagnosis of diabetes mellitus is essential for disease management. Diabetes mellitus does not appear suddenly; instead, the patient presents symptoms of impaired glucose tolerance that can also be diagnosed via glucose tolerance testing. This work presents a classification and diagnosis scheme for diseases, specifically diabetes mellitus and poor glucose tolerance, using deep neural networks based on time series data. In addition, data from virtual patients were obtained through the Dalla Man and UVA/Padova models; the validation was carried out with data from actual patients. The results show that deep neural networks have an accuracy of 96%. This indicates that DNNs is a helpful tool that can improve the diagnosis and classification of diseases in early detection.

Analysis and prospects of common problems in clinical data mining of traditional Chinese medicine prescriptions

Mining data from traditional Chinese medicine(TCM) prescriptions is one of the important methods for inheriting the experience of famous doctors and developing new drugs. However, current research work has problems such as to be optimized research plans and non-standard statistics. The main problems and corresponding solutions summarized by the research mainly include four aspects.(1)The research plan design needs to consider the efficacy and quality of individual cases.(2)The significance of the difference in confidence order of association rules needs to be further considered, and the lift should not be ignored.(3)The clustering analysis steps are complex. The selection of clustering variables should comprehensively consider factors such as the frequency of TCM, network topology parameters, and practical application significance. The selection of distance calculation and clustering methods should be improved based on the characteristics of TCM clinical data. Jaccard distance and its improvement plan should be given attention in the future. A single, unexplained clustering result should not be presented, but the final clustering plan should be selected based on a comprehensive consideration of TCM clinical characteristics and objective evaluation indicators for clustering.(4)When calculating correlation coefficients, algorithms that are only suitable for continuous variables should not be applied to binary variables. This article explained the connotations of the above problems based on the characteristics of TCM clinical research and statistical principles and proposed corresponding suggestions to provide important references for future data mining research work.

Traffic Flow Prediction Based on Spatiotemporal Potential Energy Fields

Traffic flow prediction is a fundamental problem in spatiotemporal data mining. Most of the existing studies focuses on designing statistical models to fit historical traffic data, which are purely data-driven approaches and fail to reveal the underlying mechanisms of urban traffic. To address this issue, we propose the spatiotemporal potential energy field model (ST-PEF+), which applies the field theory for human mobility to interpret the underlying mechanisms of urban traffic, and introduces the theory into data-driven deep learning models. ST-PEF+ consists of a PEF extraction module and a data-driven module. Inspired by the field theory for human mobility, the PEF extraction module adopts an algorithm to decompose the grid-based traffic flow graph into several polytree-based potential energy fields (PEFs), where traffic flows from high potential locations to low potential locations, just as water is driven by the gravity field. We also provide a theoretical analysis to ensure that the polytree decomposition algorithm can decompose any traffic flow graph. In the data-driven module, ST-PEF+ learns a spatiotemporal deep learning model to predict the dynamics of PEFs. The model adopts correlation-adaptive neural network structures, which consists of a temporal component for temporal correlations and a spatial component for spatial correlations. The temporal component employs a GRU and DCN combined structure to capture both short-term autocorrelation and long-term repeating patterns of PEFs. The spatial component extends the GAT using weighted directed attention to model the asymmetric spatial structure in PEFs. The prediction results of traffic flow are finally derived from PEFs that are predicted by the spatiotemporal deep learning model. We conduct extensive evaluations on three real-world traffic datasets. The results show that our model outperforms the state-of-the-art baselines. In addition, case studies confirm that the PEFs learned in our framework can reveal the underlying mechanisms of urban traffic, thus improving the model interpretability.

SufRec, an algorithm for mining association rules: Recursivity and task parallelism

The mining of frequent itemsets and association rules is a core problem in data mining and an essential task in data analysis. In this paper, we present SufRec, a new algorithm for finding frequent itemsets and association rules. We give two versions. In both versions, the mining of the frequent itemsets is decomposed as a sequence of tasks. The first version (the SufRecDep algorithm) proceeds the tasks successively, each one using the results of the previous ones. The second version (the SufRecInd algorithm) performs the tasks independently of each other. Both versions are recursive with respect to the items and have thus the advantage to be particularly efficient for updating the mining process when new items are added to the database or when others are excluded. Moreover, the task-independent processing of SufRecInd makes it very easy to build parallel versions of SufRec. We present two parallel SufRec algorithms. With the first one, P-FIFO-SufRec, the processors perform the tasks according to the order in which the items appear in the database, preserving the recursive nature of SufRec. For second one, P-PAST-SufRec, the set of items is fixed and the tasks are pre-assigned to each processor. In order to evaluate the performance of SufRec, we carry out a consequent experimental study. In particular, we compare its running times with those of well-known parallel and non-parallel frequent itemsets mining algorithms.

Categorization of crowd-sensing streaming data for contextual characteristic detection

The growing reliance on large wireless sensor networks, potentially consisting of hundreds of nodes, to monitor real-world phenomena inevitably results in large, complex datasets that become increasingly difficult to process using traditional methods. The inadvertent inclusion of anomalies in the dataset, resulting from the inherent characteristics of these networks, makes it difficult to isolate interesting events from erroneous measurements. Simultaneously, improvements in data science methods, as well as increased accessibility to powerful computers, lead to these techniques becoming more applicable to everyday data mining problems. In addition to being able to process large amounts of complex streaming data, a wide array of specialized data science methods enables complex analysis not possible using traditional techniques. Using real-world streaming data gathered by a temperature sensor network consisting of approximately 600 nodes, various data science methods were analyzed for their ability to exploit implicit dependencies embedded in unlabelled data to solve the complex task to identify contextual characteristics. The methods identified during this analysis were included in the construction of a software pipeline. The constructed pipeline reduced the identification of characteristics in the dataset to a trivial task, the application of which led to the detection of various characteristics describing the context in which sensors are deployed.