Small Data Research Articles

Novices’ reasoning with big data and data-ing through the lens of uncertainty

AbstractAdvancements in computing and statistical analyses have significantly transformed mathematics education, integrating it with data and computer science. As big data becomes increasingly prevalent, understanding its complexities, including its unique uncertainties, is crucial. This study explores the uncertainty that novice students can articulate during their initial engagement with big data, including both data preparation activities or data-ing as well as data analysis activities. Consistent with prior research on students’ engagement with small data, we offer a case study revealing that novices can also exhibit extreme views of uncertainty while engaging with big data. Contrary to the context of small data engagement, our analysis revealed different sources for these extreme views. The classification we used can offer means to identify students’ uncertainty views and support them in developing a more balanced, mature, perception of uncertainty. Moreover, the analysis underscores students’ inclination towards high-uncertainty articulations during their initial encounters with big data-ing activities.

Neural network–based transfer learning to improve stiffness modeling of industrial robots with small experimental data sets

AbstractStiffness modeling is an essential subject for the composition of robot control. Accurate stiffness modeling is helpful for improving the control accuracy of industrial robots, particularly under dynamic load circumstances. The classic virtual joint modeling (VJM) method is challenging in predicting the deformation of the end-effector throughout the full workspace due to the nonlinear deformation of the robot joint and its serial articulated structure. This paper proposes a full-space stiffness modeling method for robots based on the integration of a multi-layer perceptual (MLP) model and VJM. To provide enough training data for the MLP model, VJM is used to build a stiffness model with a small set of experimental data to generate 106,400 training data. A model-based transfer learning approach is proposed to improve the model’s accuracy and generalization regarding the difference between generated training data and actual experimental data. The VJM stiffness model is compared with the MLP stiffness model and the existing CNN-based transfer learning model based on the same experimental data. Considering the deformation prediction in the three directions in Cartesian space, the mean absolute error, standard deviation, and maximum error of the MLP model are decreased by at least 24.90%, 14.20%, and 8.50%, respectively, than the VJM. These prediction results demonstrate that the proposed modeling technique can significantly increase the accuracy of robot stiffness modeling, which is essential for position compensation in precise motion control of robots under dynamic load.

Single Well Production Prediction Model of Gas Reservoir Based on CNN-BILSTM-AM

In the prediction of single-well production in gas reservoirs, the traditional empirical formula of gas reservoirs generally shows poor accuracy. In the process of machine learning training and prediction, the problems of small data volume and dirty data are often encountered. In order to overcome the above problems, a single-well production prediction model of gas reservoirs based on CNN-BILSTM-AM is proposed. The model is built by long-term and short-term memory neural networks, convolutional neural networks and attention modules. The input of the model includes the production of the previous period and its influencing factors. At the same time, the fitting production and error value of the traditional gas reservoir empirical formula are introduced to predict the future production data. The loss function is used to evaluate the deviation between the predicted data and the real data, and the Bayesian hyperparameter optimization algorithm is used to optimize the model structure and comprehensively improve the generalization ability of the model. Three single wells in the Daniudi D28 well area were selected as the database, and the CNN-BILSTM-AM model was used to predict the single-well production. The results show that compared with the prediction results of the convolutional neural network (CNN) model, long short-term memory neural network (LSTM) model and bidirectional long short-term memory neural network (BILSTM) model, the error of the CNN-BILSTM-AM model on the test set of three experimental wells is reduced by 6.2425%, 4.9522% and 3.0750% on average. It shows that on the basis of coupling the empirical formula of traditional gas reservoirs, the CNN-BILSTM-AM model meets the high-precision requirements for the single-well production prediction of gas reservoirs, which is of great significance to guide the efficient development of oil fields and ensure the safety of China’s energy strategy.

Reliability analysis of DDV and SPM in BOP-based Weibull model and expand fault dataset using deep learning

The hydraulic systems of deepwater Blowout Preventers (BOPs) are crucial for ensuring the reliability and safety of offshore oil and gas extraction operations. A functional failure can lead to uncontrolled blowouts, resulting in casualties and significant economic losses on the rig. The Direct Drive Valve (DDV) and the Subsea Plated Mounted (SPM) valve are key components that help maintain the proper functioning of the hydraulic system in deepwater BOPs. This study begins by utilizing the Weibull analysis method to assess the reliability of the DDV and SPM valves using limited fault data samples. To enhance the accuracy of predictions, Weibull parameters are estimated through various methods, including Maximum Likelihood Estimation (MLE), Least Squares Estimation (LSE), and a combination of Correlation Coefficient Optimization with Support Vector Regression (CCO + SVR).Given the challenges in gathering extensive fault data for DDV and SPM valves—due to complex subsea environments, cost constraints, time limitations, and other factors—this study proposes a method employing a Back Propagation Neural Network (BPNN) model to augment the limited fault data samples. To ensure the reliable operation of the DDV and SPM valves, preventive maintenance cycles are established at 2840 and 7550 operations, respectively. At the same reliability level, as the number of operational cycles increases, the remaining service life of the valves gradually decreases, leading to a higher probability of failure over a shorter timeframe. The Mean Remaining Life (MRL) of the DDV and SPM valves, corresponding to different operational times, is analyzed, providing essential reference points for their usage and maintenance. When the extended data sample is utilized for reliability evaluation, the reliability characteristics of the small fault data samples are effectively reflected, and the parameter prediction error remains low. This indicates that the extended data sample is more suitable for reliability evaluation.

Abstract 4122543: Can We Diagnose Ticagrelor-induced Dyspnea After Urgent Percutaneous Coronary Intervention with Cardiopulmonary Exercise Testing?

Background: Ticagrelor has been reported to cause shortness of breath (SOB) in 6 to 9 % of patients. Precise mechanisms are not enirely understood. We hypothesize that cardiopulmonary exercise testing (CPET) may be helpful in clinical assessment of patients reporting SOB on ticagrelor. Case presentation: 66-yo female with history of arterial hypertension, DM type 2 and prior MI presented with STEMI and successfully underwent urgent PCI of RCA and LAD. The patient received ticagrelor in loading dose followed by 90 mg b.i.d plus low-dose aspirin. Just after PCI she developed SOB with no evidence of acute cardiac or respiratory compromise. After discharge the patient continued to experience SOB impairing daily activities. In 6 months she was re-assessed. Clinical investigation (physical exam, blood tests, echo, CXR, spirometry) showed no abnormalities that could explain SOB. CPET despite good effort showed low VO 2 peak (7,8 ml*min/kg; 49% from predicted), undetectable anaerobic threshold (AT), severely impaired ventilatory efficacy (VE/VCO 2 =50,1) and marked hyperventilation (resting RER=1,0; PetCO 2 =18,0 mmHg at rest and 22,4 mmHg at peak load). Exercise EKG, blood pressure response, SpO 2 and breathing reserve were normal. Once the patient was switched from ticagrelor to clopidogrel she reported significant improvement. CPET one week after showed increase in VO 2 peak up to 12,5 ml/kg/min., AT of 11,1 ml*min/kg, better ventilatory efficacy (VE/VCO 2 =37,1) and less hyperventilation (PetCO 2 28,1 mmHg at rest and 28,8 mmHg at peak load). CPET performed one month later showed further improvement in VO 2 peak to 13,2 ml*min/kg, normal VE/VCO 2 and no signs of hyperventilation. Discussion: it is believed that ticagrelor-induced dyspnea can be driven by 2 potential mechanisms, first related to inhibition of adenosine reuptake, second - to sensory neuron P2Y12 inhibition. It occurs early after starting the medication and is not associated with structural cardiopulmonary abnormalities. Previous small data showed evidence of Cheyne-Stokes respiration during both night and day related to ticagrelor. Symptoms usually are dose-dependent and disappear after discontinuation of ticagrelor. In our case the patient demonstrated reversible hyperventilation and reduced exercise capacity associated with ticagrelor that was confirmed with CPET. Conclusion: CPET allows differentiating SOB in patients treated with ticagrelor. Further investigations are required to confirm this hypothesis.

Evaluation Method of Water Injection Development Effects in Fractured Vuggy Carbonate Reservoirs: Case Study of FI7 Strike-Slip Fault Zone

As the main method of secondary development, water injection has been widely used in fractured vuggy carbonate reservoirs. By using an evaluation index system combined with an integrated approach to objectively evaluate and support targeted and operational adjustments of water injection development effects in fractured vuggy carbonate reservoirs, a comprehensive evaluation method is established in this study. The CRITIC method is used as the main approach, and the analytic hierarchy process, the entropy weight method, and the coefficient of variation method are used as sub-methods. Additionally, the clustering centres are divided using the clustering method to reduce the error caused by the irrational distribution of the relevant evaluation index data. The method is used to evaluate and analyse the effects of water injection development, using the FI7 fault zone of the Halahatang oilfield as an example. The application results show that the evaluation method is feasible and effective for cases with small data volume, fewer computational resources, and less time. This study has a certain reference significance for the evaluation of the effect of water injection development in similar fractured vuggy reservoirs.

Scoping the requirements for a genomic competency framework to inform the integration of genomics into pharmacy undergraduate curricula

Abstract Introduction The application of genomics is due to have a significant impact on the role of pharmacy within future healthcare systems. Since 2021, the updated GPhC education standards have included a requirement to apply the principles of genomics,1 but a lack of consistency across UK pharmacy programmes has led to inequity of genomics literacy for pharmacy students.1 National UK genomic workforce strategies include a pharmacy undergraduate indicative syllabus focussing on upskilling the NHS workforce in readiness for the wider implementation of genomics.2,3 There is a lack of UK research exploring the undergraduate genomic education requirements of pharmacy professionals. Aim Scope the requirements of a genomic competency framework for undergraduate pharmacy curricula. Methods Ethics approval was obtained from Swansea University ethics committee in July 2023. Fifteen multi-professional stakeholders from across the UK, perceived as genomic experts and/or pharmacy educationalists from each devolved nation, were selected using purposive sampling and all were invited to participate within a virtual semi-structured interview (SSI) and a consecutive focus group. SSIs and the focus group were facilitated and recorded using Microsoft® Teams and pre-designed open question topic frameworks were used. All data was anonymously transcribed verbatim and analysed using thematic analysis. Results Twelve participants consented and SSIs were held between September/October 2023. Five participants attended the consecutive focus group in December 2023. Five SSI themes were identified: Competency development - where the need for pharmacists to have comprehensive knowledge of the aspects and factors affecting the use of genomic testing within clinical practice (e.g. research and consent) was identified, whilst others described the need for pharmacists to focus on pharmacogenomics only; Curriculum approach – where there were variations in the depth of general genomic training and appropriate topics needed by pharmacists and the requirement for genomics to be scattered throughout undergraduate programmes; Future practice roles - the current postgraduate pharmacogenomic module could be used to establish the future undergraduate requirements and the need for pharmacy curricula to focus on the application of genomics to medicines rather than as a diagnostic tool; Genomic topics - participants acknowledged the complex nature of determining the genomic undergraduate curricula requirements without routine mainstream use of genomics within clinical practice but suggested pharmacy genomic competencies needing further refinement; Level of knowledge - participants highlighted the need for overall genomic pharmacy knowledge but a deeper level of pharmacogenomic knowledge and recommended that the competency framework should be relevant and achievable across healthcare sectors. Discussion and conclusion Genomic competencies should be established for pharmacy professionals as appropriate to their job roles. Once genomics is widely implemented, clarity around genomic competence requirements will develop, following workforce upskilling. The varied backgrounds of study participants impacted their opinions which were potentially influenced by their knowledge of the pharmacist role across sectors and their lack of familiarity with current pharmacy training. Study limitations are small participant numbers and low data generalisability. A genomic competency framework is needed to standardise pharmacy training and study findings will be used to develop a UK consensus for UK genomic undergraduate pharmacy training.

A study of the role of data and model uncertainty in active learning

Uncertainty-based active learning strategies have demonstrated significant superiority in small data research of materials domain. This study explores the effects of model uncertainty and data uncertainty separately on the performance of active learning strategies, specifically focusing on the number of iterations required to identify the optimal samples. For model uncertainty, three kinds of acquisition functions are compared, including predicted value strategy (PV), ranking of predicted value strategy (PR) and expected improvement strategy (EI). Among these, the active learning model utilizing PR requires the fewest average iterations (1.75). For data uncertainty, we evaluate the iterations of active learning by Gaussian process models that incorporate the uncertainty of the observations and noise samples that takes account into the uncertainty of the input features respectively. The results indicate that the active learning iterations of the three strategies converge to similar at the optimal weighting when the uncertainty of the observations is considered in the model (EI for 1.75, PV for 1.21 and PR for 1.18). In contrast, incorporating noise samples into the augmented dataset after the original samples would severely deteriorate the efficiency of active learning recommendations. Our findings aim to offer guidance for exploring more favorable acquisition functions and methods for active learning strategies.

Model of Dynamic Mechanical Parameters and Vibration Analysis in Hot Rolling

As the core equipment of metal shaping processing, the vibration problem of rolling mill seriously affects the product quality and production efficiency. To address the problem of the limited prediction of mill vibration due to small sample data in non-steady states, we propose to predict dynamic mechanical parameters (rolling force and rolling torque) based on the TCN-LSTM step strategy transfer model. The prediction accuracies of the TCN-LSTM step strategy transfer model under 1000 sets of training data reach 95.8% and 93.2%, respectively, and at the same time, it saves the time of training and regulation of the deep learning model and can better meet the needs of online prediction. The horizontal-torsion hot rolling vibration model is further established to quantitatively analyze the relationship between process parameters and mill vibration. The experimental results show that with the vibration suppression measures of 10% reduction of the rolling speed and 10% reduction of the entrance thickness, the horizontal vibration displacement amplitude is reduced from 0.687 × 10−4 m to 0.229 × 10−4 m, and the rotational displacement amplitude is reduced from 0.0273 m to 0.0117 m, which effectively suppresses the vibration of the mill and improves the stability of the rolling process.

Online and Offline Fault Detection and Diagnostics in a Nuclear Power Plant Condenser

Nuclear power plants (NPPs) face significant financial pressures due to operational and maintenance costs. This research investigates Fault Detection and Diagnostics (FDD) techniques to optimize maintenance schedules and reduce expenses. The NPP condenser plays a critical role in converting turbine exhaust steam back into water for reuse. Condenser tube fouling, a prevalent fault mode, impedes heat transfer efficiency and can lead to decreased plant efficiency and safety risks. This study proposes an FDD framework that leverages raw signal analyses from temperature and pressure monitoring to detect and diagnose condenser tube fouling in both online and offline settings. The online approach facilitates close-to-real-time predictions, enabling proactive maintenance strategies. Additionally, the framework explores incorporating a condenser’s maintenance history for enhanced diagnostics. We employ a dataset obtained from a simulated nuclear power plant condenser using the Asherah Nuclear Power Plant Simulator (ANS). ANS replicates the operational dynamics of a pressurized water reactor (PWR) type NPP. The proposed methodology utilizes an encoder-decoder (E-D) structured 1DCNN model to analyze the raw signals. The research demonstrates consistent and accurate fault detection and diagnostics for condenser tube fouling in both online and offline scenarios. A high potential for generalization to unseen conditions was observed. However, online detection using small data windows necessitates caution due to potential false alarms around the transition points. Our findings pave the way for further exploration of robust diagnostics by accommodating a wider spectrum of fouling rates within degradation classes using ANS. This combined online and offline FDD approach offers a promising solution for promoting operational safety, efficiency, and cost-effectiveness in NPP condensers.

TIWMFLP: Two-Tier Interactive Weighted Matrix Factorization and Label Propagation Based on Similarity Matrix Fusion for Drug-Disease Association Prediction.

Accurately identifying new therapeutic uses for drugs is crucial for advancing pharmaceutical research and development. Matrix factorization is often used in association prediction due to its simplicity and high interpretability. However, existing matrix factorization models do not enable real-time interaction between molecular feature matrices and similarity matrices, nor do they consider the geometric structure of the matrices. Additionally, efficiently integrating multisource data remains a significant challenge. To address these issues, we propose a two-tier interactive weighted matrix factorization and label propagation model based on similarity matrix fusion (TIWMFLP) to assist in personalized treatment. First, we calculate the Gaussian and Laplace kernel similarities for drugs and diseases using known drug-disease associations. We then introduce a new multisource similarity fusion method, called similarity matrix fusion (SMF), to integrate these drug/disease similarities. SMF not only considers the different contributions represented by each neighbor but also incorporates drug-disease association information to enhance the contextual topological relationships and potential features of each drug/disease node in the network. Second, we innovatively developed a two-tier interactive weighted matrix factorization (TIWMF) method to process three biological networks. This method realizes for the first time the real-time interaction between the drug/disease feature matrix and its similarity matrix, allowing for a better capture of the complex relationships between drugs and diseases. Additionally, the weighted matrix of the drug/disease similarity matrix is introduced to preserve the underlying structure of the similarity matrix. Finally, the label propagation algorithm makes predictions based on the three updated biological networks. Experimental outcomes reveal that TIWMFLP consistently surpasses state-of-the-art models on four drug-disease data sets, two small molecule-miRNA data sets, and one miRNA-disease data set.

사전학습 모델 기반 발화 동영상 멀티 모달 감정 인식

Recently, as the demand for non-face-to-face counseling has rapidly increased, the need for emotion recognition technology that combines various aspects such as text, voice, and facial expressions is being emphasized. In this paper, we address issues such as the dominance of non-Korean data and the imbalance of emotion labels in existing datasets like FER-2013, CK+, and AFEW by using Korean video data. We propose methods to enhance multimodal emotion recognition performance in videos by integrating the strengths of image modality with text modality. A pre-trained model is used to overcome the limitations caused by small training data. A GPT-4-based LLM model is applied to text, and a pre-trained model based on VGG-19 architecture is fine-tuned to facial expression images. The method of extracting representative emotions by combining the emotional results of each aspect extracted using a pre-trained model is as follows. Emotion information extracted from text was combined with facial expression changes in a video. If there was a sentiment mismatch between the text and the image, we applied a threshold that prioritized the text-based sentiment if it was deemed trustworthy. Additionally, as a result of adjusting representative emotions using emotion distribution information for each frame, performance was improved by 19% based on F1-Score compared to the existing method that used average emotion values for each frame.

The Effect of Women’s Role on Board, ESG Performance, and Tax Avoidance

This research examines the impact of board gender diversity on tax avoidance, with ESG performance as an intervening variable. Using 87 non-financial companies listed on the Indonesia Stock Exchange (IDX) with ESG scores from Thomson-Reuters ASSET4 during the 2017-2019 period, the study reveals that board gender diversity significantly influences tax avoidance, indicating that diversity does not improve tax compliance. The male-dominated boards in Indonesia limit the effectiveness of monitoring corporate activities, including tax avoidance. Additionally, gender diversity negatively affects ESG performance, suggesting that the small number of women on boards diminishes efforts to enhance sustainability. ESG performance does not mediate the relationship between gender diversity and tax avoidance, though it can reduce tax avoidance when transparency in ESG reporting is increased. Limitations include a small sample size and data restricted to the 2017-2019. Future studies could explore alternative tax avoidance measures and other corporate governance factors.

Research on the P‐S‐N Curve Fitting Method of Notched Specimens Considering Small Sample Properties

ABSTRACTAiming at the issue of fatigue test data for large‐scale mechanical components of building steel are very limited, a method for fitting P‐S‐N curves under small sample data of notched specimens is proposed to predict fatigue life. First, a fatigue life subsample augmented and its reliability assessment method are established, based on Bayesian hierarchical modeling and modified Monte Carlo method. Second, a clustering combination weighting method is proposed, to define weights of hidden variables of the binomial mixture Weibull distribution, and the expectation–maximization algorithm is used to determine probability density function of the distribution. Finally, the P‐S‐N curves under various failure probabilities are fitted with Weibull distributed life models, and the convergence and prediction accuracy of the different models are compared. The results show that the fatigue data of small samples can be predicted better by using mixed Weibull distribution, and the fitting P‐S‐N curve is more reliable and accurate.

Ethnic inequalities in primary care experiences for people with multiple long-term conditions: Evidence from the general practice patient survey

ObjectivesTo examine the relationship between ethnicity and experiences of primary care for people with multiple long-term conditions (MLTCs) and assess the relative importance of demographic, practice, and area-level factors as influences on primary care experiences across ethnic groups. Study designA retrospective study using 2018–19 GP Patient Survey data linked to General Practice Workforce data and small area data published by the Office for National Statistics. MethodsWe conducted multilevel regression analysis to assess the relationship between ethnicity and experience of accessing primary care and interacting with healthcare professionals. We built separate regression models for each outcome and included (i) each covariate separately, (ii) demographic factors and (iii) demographic, practice, and area-level factors. ResultsUpon full adjustment Arab, Bangladeshi, Chinese, Indian, Pakistani, other Asian, mixed white and Asian, and other white people with MLTCs have lower levels of satisfaction with primary care access and interacting with healthcare professionals compared with white British people. The influence of demographic, practice and area-level factors is not uniform across ethnic groups; demographic factors account for the inequalities in levels of satisfaction with access to primary care between white British people and Black other, mixed other, mixed white & Black Caribbean and Gypsy & Irish Travellers. However, practice and area-level factors strengthen inequalities in the experience of accessing primary care for Bangladeshi, Indian and Pakistani people. ConclusionsGiven the link between patient satisfaction and patient-related health outcomes, the lower levels of satisfaction with accessing primary care and interacting with healthcare professionals among the aforementioned minoritised ethnic groups are concerning and require further scrutiny. Qualitative studies are required to understand and address the sources of poor experiences in primary care for minoritised people with MLTCs to improve patient-centred healthcare and outcomes.

Inverse problems in hydrodynamics lubrication: Parameter identification in the Reynold equation by using physics-informed neural networks

In this work, the inverse problems for hydrodynamic lubrication were solved by using the physics informed neural networks (PINNs) method. Both the conditions without and with the cavitation effects are studied. For the journal bearings without considering the cavitation effects, the eccentricity values can be accurately inverted using only 10 data points with a maximum error of less than 6%. For the problems with the mass-conserving cavitation, the geometry parameters for 1D conditions were estimated and accurate results can be obtained by also using 10 data points. While for the journal bearings with incompressible lubricants, the error was less than 3% with only 50 data points. For the ones with compressible lubricants, the eccentricity ratios can be estimated with 50 data points again, and the maximum error is less than 10%. In addition, the experimental oil pressure data were used to predict the journal bearing eccentricity, and the results were acceptable with only 9 oil pressure data points. The PINNs method presented two significant strengths: its ability to operate effectively with very sparse data points and its precision in delivering results, and it can provide a promising method for identifying unknown parameters in hydrodynamics lubrication with small data.

The C-V2V-assisted early small cell data offloading (E-SCDO) method using the multi-access edge computing (MEC) mechanism

The C-V2V-assisted early small cell data offloading (E-SCDO) method using the multi-access edge computing (MEC) mechanism

The impact of fuel poverty on cold-related mortality in England and Wales: disentangling the effects of fuel price, income, and energy efficiency

Abstract Cold temperatures have been widely acknowledged as an important burden of poor health in many parts of the world including the UK. Although the temperature is expected to increase in the future due to climate change, the latest UK Health Effects of Climate Change (HECC) report projected that the number of cold-related deaths in the UK is expected to be more than 2 times of heat in the 2070s even under a high emission climate change scenario (RCP8.5). Therefore, it is still imperative to take intervention to reduce the health impacts of cold. People spent more than 90% of their time indoor and hence being able to stay in warm homes is crucial for reducing the impact of cold. However, some people may not be able to warm their home enough, a problem known as fuel poverty. Multiple factors can contribute to fuel poverty, including low income, poor housing performance, high fuel price as well as competing interest in fuel consumption and other essential living expenses. These challenges have been witnessed markedly during the cost of living crisis since 2021. This research investigates the effect of fuel poverty on cold-related mortality risk in England and Wales using small areal data. The underlying contributing factors of fuel poverty are explored to study the main mechanisms. The findings can support the identification of those who may be mostly affected by fuel poverty, and hence improve policies and interventions on reducing fuel poverty and protecting the public health from low temperatures.

Subspace‐based distributed target detection method with small training data samples

AbstractDetecting distributed targets precisely in homogeneous environments has been a hot topic in radar signal processing. Generally, distributed targets are often modelled with subspace models of unknown coordinates, and clutter is modelled as the complex Gaussian distribution with zero mean and unknown covariance matrix, while covariance matrix is estimated with a set of training data without the target signal. However, in practice, the complexity of the external environment makes the training data that satisfy the condition of independent homogeneous distribution less available. Therefore, it is assumed that the covariance matrix of the clutter is persymmetric structure and the approach of dimensionality reduction using subspace transformations is introduced, two detectors based upon generalised likelihood ratio test criterion and Wald test criterion in homogeneous environments are proposed. Theoretical analyses indicate the constant false alarm rate characteristics of the two proposed detectors for unknown clutter covariance matrices. Simulation analyses indicate that the proposed detector works well even with fewer training data samples, and its detection performance outperforms that of existing contrast detectors.

A multi-step regularity assessment and joint prediction system for ordering time series based on entropy and deep learning

Customer maintenance is of vital importance to the enterprise management. Valuable assessment and efficient prediction for customer ordering behavior can offer better decision-making and reduce business costs significantly. According to existing studies about customer behavior regularity segment and demand prediction most focus on e-commerce and other fields with large amount of data, making them not suitable for small enterprises and data features like sparsity and outliers are not mined when doing regularity quantification. Additionally, more and more complex network structures for demand prediction are proposed, which builds on the assumption that all the samples have predictive value, ignoring the fine-grained analysis of different time series regularity with high cost. To deal with the above issues, a multi-step regularity assessment and joint prediction system for ordering time series is proposed. For extracting features, comprehensive assessment of customer regularity based on entropy weight method with the result of predictability quantification using K-Means clustering algorithm, real entropy, LZW algorithm and anomaly detection adopting Isolation Forest algorithm not only gives an objective result to ‘how high the regularity of customers is’, filling the gap in the field of regularity quantification, but also provides a theoretical basis for demand prediction models selection. Prediction models: Random Forest regression, XGBoost, CNN and LSTM network are experimented with sMAPE and MSLE for performance evaluation to verify the effectiveness of the proposed regularity quantitation method. Moreover, a merged CNN-BiLSTM neural network model is established for predicting those customers with low regularity and difficult to predict by traditional machine leaning algorithms, which performs better on the data set compared to others. Random Forest is still used for prediction of customers with high regularity due to its high training efficiency. Finally, the results of prediction, regularity quantification, and classification are output from the intelligent system, which is capable of providing scientific basis for corporate strategy decision and has highly extendibility in other enterprises and fields for follow-up research.