Incremental Algorithm Research Articles

DBSP: Automatic Incremental View Maintenance for Rich Query Languages

Incremental view maintenance (IVM) has long been a central problem in database theory. Many solutions have been proposed for restricted classes of database languages, such as the relational algebra, or Datalog. These techniques do not naturally generalize to richer languages. In this paper we give a general, heuristic-free solution to this problem in 3 steps: (1) we describe a simple but expressive language called DBSP for describing computations over data streams; (2) we give a new mathematical definition of IVM and a general algorithm for solving IVM for arbitrary DBSP programs, and (3) we show how to model many rich database query languages using DBSP (including the full relational algebra, queries over sets and multisets, arbitrarily nested relations, aggregation, flatmap (unnest), monotonic and non-monotonic recursion, streaming aggregation, and arbitrary compositions of all of these). SQL and Datalog can both be implemented in DBSP. As a consequence, we obtain efficient incremental view maintenance algorithms for queries written in all these languages.

Broad Minimax Probability Learning System and its Application in Regression Modeling

Broad learning system (BLS), a novel incremental learning algorithm, has attracted more and more attention and application. BLS achieves a perfect balance between learning performance and modeling efficiency based on the mean square error (MSE) criterion. However, MSE assumes that a random error between an observed value and real value is a Gaussian distribution, which may be inconsistent to the real-world situation. Moreover, the distributional assumptions may cast doubt on the generalization and validity of the built BLS model. In this article, we propose a novel distribution-free BLS for regression analysis, called broad minimax probability learning system (BMPLS). Learning from the idea of minimax probability machine regression, BMPLS builds a regression model by maximizing the worst probability of the regression function being within the allowed error range. It adequately utilizes the mean and covariance information of the hidden-space feature without making any distributional assumptions of the random error. In order to further improve the learning performance of the model, a regularized BMPLS (RBMPLS) is proposed by applying elastic-net regularization term to BMPLS. The corresponding optimization algorithm is also designed for calculating the output weights of the model. Multiple experiments based on public datasets were carried out to demonstrate the feasibility of the proposed algorithms.

Self-Aware SGD: Reliable Incremental Adaptation Framework For Clinical AI Models.

Healthcare is dynamic as demographics, diseases, and therapeutics constantly evolve. This dynamic nature induces inevitable distribution shifts in populations targeted by clinical AI models, often rendering them ineffective. Incremental learning provides an effective method of adapting deployed clinical models to accommodate these contemporary distribution shifts. However, since incremental learning involves modifying a deployed or in-use model, it can be considered unreliable as any adverse modification due to maliciously compromised or incorrectly labelled data can make the model unsuitable for the targeted application. This paper introduces self-aware stochastic gradient descent (SGD), an incremental deep learning algorithm that utilises a contextual bandit-like sanity check to only allow reliable modifications to a model. The contextual bandit analyses incremental gradient updates to isolate and filter unreliable gradients. This behaviour allows self-aware SGD to balance incremental training and integrity of a deployed model. Experimental evaluations on the Oxford University Hospital datasets highlight that self-aware SGD can provide reliable incremental updates for overcoming distribution shifts in challenging conditions induced by label noise.

Design of Double Closed Loop Control System for DC Motor Based on TX8C1010 Single Chip

In order to improve the operational strength and control precision of the DC motor, a double closed-loop control system based on TAIXIN TX8C1010 MCU is proposed. The single chip is based on a high-speed 8051 core, which is applied to home appliances and can realize the double closed-loop motor control algorithm. The given speed value provided by the potentiometer is compared with the real-time speed value and inputs the error to the speed loop function, which is based on the incremental PID algorithm, to calculate the reference current value. Then it is compared with the real-time current value and inputs the error into the current loop. The PWM signal duty cycle is determined by the output value of the current loop to adjust and stabilize the motor speed. The drive circuit outputs the PWM signal in the inverse amplification way to control the operation of the DC motor, and finally tests the function on the MAGTROL M-TEST7 motor test platform. The test results show that the control system has fast response speed, strong load capacity, good robustness and safety. Hence, TAIXIN TX8C1010 single-chip machine is feasible in the DC motor control. It can be improved to strengthen its load characteristics according to the parameters and applied to a variety of power DC motor control.

A Review of Partial Shading MPPT Algorithm on Speed, Accuracy, and Cost Embedded

This paper describes several Maximum Power Point Tracking algorithms for partial shading conditions that have detrimental effects on photovoltaic systems. The method used is a literature review of articles from reputable publishers. Fifty two articles were obtained after meeting the established criteria for selection. The literature review focused on the ability of the Maximum Power Point Tracking algorithm to overcome partial shading conditions in terms of tracking speed, tracking accuracy, efficiency, and implementation complexity. As the results, some algorithms are recommended to be applied for Maximum Power Point Tracking including the Single Swam Algorithm and the Perturb And Observer algorithm, the Enhanced Adaptive Step Size Perturb and Observe algorithm, the Novel Adaptable Step Incremental Conductance algorithm, the Improved Bat Algorithm and Fuzzy Logic Controller algorithm, and the Particle Swarm Optimization with One Cycle Control algorithm. In terms of implementation complexity, these five algorithms are categorized as medium-complexity, which can be characterized as low cost, high efficiency, and even 100% with high tracking speed and accuracy with a minimum number of sensors used.

Physical Structure Expression for Dense Point Clouds of Magnetic Levitation Image Data.

The research and development of an intelligent magnetic levitation transportation system has become an important research branch of the current intelligent transportation system (ITS), which can provide technical support for state-of-the-art fields such as intelligent magnetic levitation digital twin. First, we applied unmanned aerial vehicle oblique photography technology to acquire the magnetic levitation track image data and preprocessed them. Then, we extracted the image features and matched them based on the incremental structure from motion (SFM) algorithm, recovered the camera pose parameters of the image data and the 3D scene structure information of key points, and optimized the bundle adjustment to output 3D magnetic levitation sparse point clouds. Then, we applied multiview stereo (MVS) vision technology to estimate the depth map and normal map information. Finally, we extracted the output of the dense point clouds that can precisely express the physical structure of the magnetic levitation track, such as turnout, turning, linear structures, etc. By comparing the dense point clouds model with the traditional building information model, experiments verified that the magnetic levitation image 3D reconstruction system based on the incremental SFM and MVS algorithm has strong robustness and accuracy and can express a variety of physical structures of magnetic levitation track with high accuracy.

Dynamic Weights Based Risk Rule Generation Algorithm for Incremental Data of Customs Declarations

Aimed at shortcomings, such as fewer risk rules for assisting decision-making in customs entry inspection scenarios and relying on expert experience generation, a dynamic weight assignment method based on the attributes of customs declaration data and an improved dynamic-weight Can-Tree incremental mining algorithm are proposed. In this paper, we first discretize the customs declaration data, and then form composite attributes by combining and expanding the attributes, which is conducive to generating rules with risk judgment significance. Then, weights are determined according to the characteristics and freshness of the customs declaration data, and the weighting method is applied to the Can-Tree algorithm for incremental association rule mining to automatically and efficiently generate risk rules. By comparing FP-Growth and traditional Can-Tree algorithms experimentally, the improved dynamic-weight Can-Tree incremental mining algorithm occupies less memory space and is more time efficient. The introduction of dynamic weights can visually distinguish the importance level of customs declaration data and mine more representative rules. The dynamic weights combine confidence and elevation to further improve the accuracy and positive correlation of the generated rules.

Parallel incremental association rule mining framework for public opinion analysis

Internet public opinion association rule mining (POARM) has garnered significant attention from a larger group of netizens. However, most POARM methods have been applied to post-event analysis, which has poor timeliness and a low efficiency. Therefore, the real-time monitoring of public opinion association rules is lacking. To address this problem, we propose the parallel Incremental POARM Framework (IPOARM), which improves the timeliness of association rule mining in two ways: 1) using an incremental merge method to consider both inserted and deleted public opinion transaction sets and reuse previous frequent itemsets to reduce redundant computation and 2) employing a parallel implementation of big data process platforms. Moreover, the flexible association rule mining (ARM) algorithm selection structure of IPOARM enables users to freely select suitable ARM algorithms. We represent four classic transaction sets as public opinion transaction sets and compare the IPOARM framework with two novel incremental association rules mining algorithms. Our evaluations indicate that the IPOARM framework can discover Internet public opinion association rules quickly, implying that it can be easily integrated into existing big data processing platforms and that it significantly improves the mining accuracy and efficiency by 12.756% and 29.371%, respectively.

A New Hybrid MPPT Based on Incremental Conductance-Integral Backstepping Controller Applied to a PV System under Fast-Changing Operating Conditions

Maximum power point tracking (MPPT) is becoming more and more important in the optimization of photovoltaic systems. Several MPPT algorithms and nonlinear controllers have been developed for improving the energy yield of PV systems. On the one hand, most of the conventional algorithms such as the incremental conductance (INC) demonstrate a good affinity for the maximum power point (MPP) but often fail to ensure acceptable stability and robustness of the PV system against fast-changing operating conditions. On the other hand, the MPPT nonlinear controllers can palliate the robust limitations of the algorithms. However, most of these controllers rely on expensive solar irradiance measurement systems or complex and relatively less accurate methods to seek the maximum power voltage. In this paper, we propose a new hybrid MPPT based on the incremental conductance algorithm and the integral backstepping controller. The hybrid scheme exploits the benefits of the INC algorithm in seeking the maximum power voltage and feeds a nonlinear integral backstepping controller whose stability was ensured by the Lyapunov theory. Therefore, in terms of characteristics, the overall system is a blend of the MPP-seeking potential of the INC and the nonlinear and robust potentials of the integral backstepping controller (IBSC). It was noted that the hybrid system successfully palliates the conventional limitations of the isolated INC and relieves the PV system from the expensive burden of solar irradiance measurement. The proposed hybrid system increased the operational efficiency of the PV system to 99.94% and was found better than the INC MPPT algorithm and 8 other recently published MPPT methods. An extended validation under experimental environmental conditions showed that the hybrid system is approximately four times faster than the INC in tracking the maximum power with better energy yield than the latter.

Edge Intelligence for Smart EL Images Defects Detection of PV Plants in the IoT-Based Inspection System

Given the huge installed capacity of photovoltaic (PV) worldwide, the traditional defect detection system for PV plants is infeasible, especially for large-scale plants. In this article, unmanned aerial vehicles (UAVs) mounted with several sensors and a computer in the cloud are used cooperatively to establish an Internet of Things-based cloud-edge computing infrastructure, which can automatically detect defects with low latency, low cost, and high accuracy. The pretrained models trained in the cloud server are embedded into the processor in UAVs to implement online detection. Specifically, given the characteristic of defects in electroluminescence images, a two-stage algorithm is proposed to identify the defects with high performance. In the first stage, cells, the basic unit in the PV module, are extracted using an encoder–decoder network. Then, a vision-based incremental defect classification algorithm is proposed for defect detection that integrates deep learning with prior knowledge to maximize computing efficiency. The performance of the proposed system is evaluated through extensive experiments.

Patient Data Analysis with the Quantum Clustering Method

Quantum computing is one of the most promising solutions for solving optimization problems in the healthcare world. Quantum computing development aims to light up the execution of a vast and complex set of algorithmic instructions. For its implementation, the machine learning models are continuously evolving. Hence, the new challenge is to improve the existing complex and critical machine learning training models. Therefore, the healthcare sector is shifting from a classical to a quantum domain to sustain patient-oriented attention to healthcare patrons. This paper presents a hybrid classical-quantum approach for training the unsupervised data models. In order to achieve good performance and optimization of the machine learning algorithms, a quantum k-means (QK-means) clustering problem was deployed on the IBM quantum simulators, i.e.,the IBM QASM simulator. In the first place, the approach was theoretically studied and then implemented to analyze the experimental results. The approach was further tested using small synthetics and cardiovascular datasets on a qsam simulator to obtain the clustering solution. The future direction connecting the dots is the incremental k-means algorithm with the quantum platform, which would open hitherto unimaginable technological doors.

IG-LSPIA: Least Squares Progressive Iterative Approximation for Isogeometric Collocation Method

The isogeometric collocation method (IGA-C), which is a promising branch of isogeometric analysis (IGA), can be considered fitting the load function with the combination of the numerical solution and its derivatives. In this study, we develop an iterative method, isogeometric least-squares progressive-iterative approximation (IG-LSPIA), to solve the fitting problem in the collocation method. IG-LSPIA starts with an initial blending function, where the control coefficients are combined with the B-spline basis functions and their derivatives. A new blending function is generated by constructing the differences for collocation points (DCP) and control coefficients (DCC), and then adding the DCC to the corresponding control coefficients. The procedure is performed iteratively until the stop criterion is reached. We prove the convergence of IG-LSPIA and show that the computation complexity in each iteration of IG-LSPIA is related only to the number of collocation points and unrelated to the number of control coefficients. Moreover, an incremental algorithm is designed; it alternates with knot refinement until the desired precision is achieved. After each knot refinement, the result of the last round of IG-LSPIA iterations is used to generate the initial blending function of the new round of iteration, thereby saving great computation. Experiments show that the proposed method is stable and efficient. In the three-dimensional case, the total computation time is saved twice compared to the traditional method.

Improved variable-step incremental conductance MPPT algorithm applied to photovoltaic storage systems

For the traditional PV maximum power tracking control method in terms of tracking speed and stability, proposed an improved variable-step incremental conductance method. The method first speeds up the tracking speed by the constant voltage coefficient method to quickly locate around the MPP(Maximum power point,MPP) and dynamically adjusts the step size using the inverse tangent function value when the operating point is located to the left of the maximum power point, and dynamically adjusts the step size using the exponential function value when the operating point is located to the right. Given the characteristics of photovoltaic power generation, it usually constitutes a light storage system with an energy storage battery. To verify the effectiveness, the simulation experiments are carried out in Matlab/Simulink to build a photovoltaic energy storage system model, which contains an energy storage battery. Results indicate that the proposed new method achieves the optimization of tracking precision and speed according to the nature of the function and can make the energy storage battery charge and discharge stably, which verifies the practicality of the proposed improved algorithm.

Linkage Problem in Location Optimization of Dedicated Bus Lanes on a Network

Methods for identifying optimal decisions for dedicated bus lane locations (DBLs) on a network have been extensively studied in the literature. However, the impacts in relation to changes to car and bus delays of deploying a DBL on a given link largely depend on where other DBLs exist on the network. Therefore, for a network-wide location optimization or a bus lane design problem, linkages exist between decision variables. Typically used metaheuristic methods to optimize DBL locations, such as genetic algorithms (GAs), do not perform well for such problems with linkages between decision variables. To this end, this paper has two novel contributions to the literature by (a) demonstrating that the linkage problem exists, and (b) testing different heuristic algorithms that are more suitable than GAs for optimizing the locations of DBL on a network. The linkage problem in the location optimization of DBLs is demonstrated by enumerating all possible bus lane locations in a small grid network. Next, optimization algorithms that do not enumerate all possible bus lane locations that are capable of learning linkages between decision variables, namely Bayesian algorithm and a population-based incremental learning algorithm, are proposed. These algorithms are compared with two types of GAs in relation to consistency and quality of the solutions, and exploration capability. Results show that algorithms that can learn linkages between decision variables perform better than the GAs.

Forecasting COVID‐19 cases using dynamic time warping and incremental machine learning methods

AbstractThe investment of time and resources for developing better strategies is key to dealing with future pandemics. In this work, we recreated the situation of COVID‐19 across the year 2020, when the pandemic started spreading worldwide. We conducted experiments to predict the coronavirus cases for the 50 countries with the most cases during 2020. We compared the performance of state‐of‐the‐art machine learning algorithms, such as long‐short‐term memory networks, against that of online incremental machine learning algorithms. To find the best strategy, we performed experiments to test three different approaches. In the first approach (single‐country), we trained each model using data only from the country we were predicting. In the second one (multiple‐country), we trained a model using the data from the 50 countries, and we used that model to predict each of the 50 countries. In the third experiment, we first applied clustering to calculate the nine most similar countries to the country that we were predicting. We consider two countries to be similar if the differences between the curve that represents the COVID‐19 time series are small. To do so, we used time series similarity measures (TSSM) such as Euclidean Distance (ED) and Dynamic Time Warping (DTW). TSSM return a real value that represents the distance between the points in two time series which can be interpreted as how similar they are. Then, we trained the models with the data from the nine more similar countries to the one that was predicted and the predicted one. We used the model ARIMA as a baseline for our results. Results show that the idea of using TSSM is a very effective approach. By using it with the ED, the obtained RMSE in the single‐country and multiple‐country approaches was reduced by 74.21% and 74.70%, respectively. And by using the DTW, the RMSE was reduced by 74.89% and 75.36%. The main advantage of our methodology is that it is very simple and fast to apply since it is only based on time series data, as opposed to more complex methodologies that require a deep and thorough study to consider the number of parameters involved in the spread of the virus and their corresponding values. We made our code public to allow other researchers to explore our proposed methodology.

PseU-ST: A new stacked ensemble-learning method for identifying RNA pseudouridine sites.

Background: Pseudouridine (Ψ) is one of the most abundant RNA modifications found in a variety of RNA types, and it plays a significant role in many biological processes. The key to studying the various biochemical functions and mechanisms of Ψ is to identify the Ψ sites. However, identifying Ψ sites using experimental methods is time-consuming and expensive. Therefore, it is necessary to develop computational methods that can accurately predict Ψ sites based on RNA sequence information. Methods: In this study, we proposed a new model called PseU-ST to identify Ψ sites in Homo sapiens (H. sapiens), Saccharomyces cerevisiae (S. cerevisiae), and Mus musculus (M. musculus). We selected the best six encoding schemes and four machine learning algorithms based on a comprehensive test of almost all of the RNA sequence encoding schemes available in the iLearnPlus software package, and selected the optimal features for each encoding scheme using chi-square and incremental feature selection algorithms. Then, we selected the optimal feature combination and the best base-classifier combination for each species through an extensive performance comparison and employed a stacking strategy to build the predictive model. Results: The results demonstrated that PseU-ST achieved better prediction performance compared with other existing models. The PseU-ST accuracy scores were 93.64%, 87.74%, and 89.64% on H_990, S_628, and M_944, respectively, representing increments of 13.94%, 6.05%, and 0.26%, respectively, higher than the best existing methods on the same benchmark training datasets. Conclusion: The data indicate that PseU-ST is a very competitive prediction model for identifying RNA Ψ sites in H. sapiens, M. musculus, and S. cerevisiae. In addition, we found that the Position-specific trinucleotide propensity based on single strand (PSTNPss) and Position-specific of three nucleotides (PS3) features play an important role in Ψ site identification. The source code for PseU-ST and the data are obtainable in our GitHub repository (https://github.com/jluzhangxinrubio/PseU-ST).

Independent MPPT and voltage regulation of dual input DC - DC converter with a voltage multiplier cell

ABSTRACT A dual input boost DC–DC converter (DIBC) with combined voltage multiplier cell, the larger duty ratio (greater than 0.5), and the overlapping switching signal are required to obtain high voltage gain. This paper investigates the independent maximum power point tracking (MPPT) implementation on two solar photovoltaic (PV) energy sources along with the output voltage regulation of the DIBC during load and input voltage disturbances. For maximal power extraction, a single ended primary-inductance converter (SEPIC) is designed at each input, one with the perturb and observe (P and O) approach and the other with the incremental conductance algorithm. The small-signal transfer function model is used to design the voltage regulator. The MPPT algorithms for maximum power extraction as well as the inner current-loop and outer voltage-loop control approach have been implemented simultaneously. Duty ratio of the DIBC MOSFETs as a function of the converter output voltage and duty ratio of the SEPIC MOSFET as a function of MPPT are used to generate the switching signals. An experimental two 100W PV sources are developed using real-time solar simulator with the 200 W DIBC and controller is designed using XILINX SPARTAN-6 FPGA board. MATLAB simulation and experimental results ensure the active current sharing among the sources, reference output voltage (156.6 V) tracking, voltage regulation (below 5%) under different scenarios, and independent MPPT simultaneously.

Comparative Study of Fuzzy Rule-Based Classifiers for Medical Applications.

The use of machine learning in medical decision support systems can improve diagnostic accuracy and objectivity for clinical experts. In this study, we conducted a comparison of 16 different fuzzy rule-based algorithms applied to 12 medical datasets and real-world data. The results of this comparison showed that the best performing algorithms in terms of average results of Matthews correlation coefficient (MCC), area under the curve (AUC), and accuracy (ACC) was a classifier based on fuzzy logic and gene expression programming (GPR), repeated incremental pruning to produce error reduction (Ripper), and ordered incremental genetic algorithm (OIGA), respectively. We also analyzed the number and size of the rules generated by each algorithm and provided examples to objectively evaluate the utility of each algorithm in clinical decision support. The shortest and most interpretable rules were generated by 1R, GPR, and C45Rules-C. Our research suggests that GPR is capable of generating concise and interpretable rules while maintaining good classification performance, and it may be a valuable algorithm for generating rules from medical data.

Incremental feature selection approach to interval-valued fuzzy decision information systems based on [formula omitted]-fuzzy similarity self-information

The relative decision self-information is a crucial evaluation function of feature selection in information system. It encapsulates classification information in upper and lower approximations and pays attention to the boundary region of samples. Nevertheless, with the frequent replacement of data, the static feature selection neglects the previous information of samples, which diminishes the computational efficiency. With the purpose of adapting to the evolution of the era, incremental learning is widely exerted in the field of data mining. In combination with incremental technique, it is not cumbersome to update the reduct in time. Enlightened by this, our work focuses on the mechanism of incremental feature selection due to the variation of objects in IvFDIS. Firstly, we construct λ-fuzzy similarity relation and introduce λ-fuzzy similarity self-information into IvFDIS based on relative decision self-information. Besides, with the assistance of matrix operation, we recommend static feature selection according to λ-fuzzy similarity self-information. Furthermore, two relevant incremental algorithms involving the insertion and removal of objects in IvFDIS are made a research. Finally, some comparative experiments are conducted on twelve public data sets to certify the validity of our incremental algorithms. Experimental results show that comparable to three tested algorithms, the proposed incremental algorithms lessen the computation time greatly, and they select fewer features in most instances without decreasing classification accuracy in IvFDIS.

A Convolutional Neural Network-Based Broad Incremental Learning Filter for Attenuating Physiological Tremors in Telerobot Systems

While master-slave teleoperated robotic systems have extensive applications in practice, the physiological tremors can easily affect the control accuracy and even destroy the stability of the closed-loop control systems during operation. Hence, the development of some effective approaches for counteracting physiological tremors is of both theoretical and practical importance. In this paper, a broad learning network-based filter integrating a deep learning network and modified incremental learning algorithms is proposed to reconstruct and compensate for tremor signals. To strengthen the recognition of correlations between different moments, the lateral connectivity structure is adopted to obtain multi-scale feature maps. Each feature window is obtained from multi-scale feature maps generated by the convolutional neural network, which has an advantage that makes the feature nodes fuse the feature information of long time series and short time series by the lateral connection. The broad learning network is a unique construction, which only needs to obtain the input and the output to conveniently calculate the connection weights by the pseudo-inverse without involving backpropagation. It is known that the relation between the data X and the label Y can be represented as XW=Y, and the solution W can be obtained by the pseudo-inverse W=X+Y. In addition, to guarantee the ill-posed problem, a ridge regression algorithm is used for the pseudo-inverse calculation. The effectiveness of our raised network architecture is illustrated by comparative simulation and experiment results.