Incomplete Data Sets Research Articles

Data Imputation of Soil Pressure on Shield Tunnel Lining Based on Random Forest Model.

With the advancement of engineering techniques, underground shield tunneling projects have also started incorporating emerging technologies to monitor the forces and displacements during the construction and operation phases of shield tunnels. Monitoring devices installed on the tunnel segment components generate a large amount of data. However, due to various factors, data may be missing. Hence, the completion of the incomplete data is imperative to ensure the utmost safety of the engineering project. In this research, a missing data imputation technique utilizing Random Forest (RF) is introduced. The optimal combination of the number of decision trees, maximum depth, and number of features in the RF is determined by minimizing the Mean Squared Error (MSE). Subsequently, complete soil pressure data are artificially manipulated to create incomplete datasets with missing rates of 20%, 40%, and 60%. A comparative analysis of the imputation results using three methods-median, mean, and RF-reveals that this proposed method has the smallest imputation error. As the missing rate increases, the mean squared error of the Random Forest method and the other two methods also increases, with a maximum difference of about 70%. This indicates that the random forest method is suitable for imputing monitoring data.

Curation of myeloma observational study MALIMAR using XNAT: solving the challenges posed by real-world data

ObjectivesMAchine Learning In MyelomA Response (MALIMAR) is an observational clinical study combining “real-world” and clinical trial data, both retrospective and prospective. Images were acquired on three MRI scanners over a 10-year window at two institutions, leading to a need for extensive curation.MethodsCuration involved image aggregation, pseudonymisation, allocation between project phases, data cleaning, upload to an XNAT repository visible from multiple sites, annotation, incorporation of machine learning research outputs and quality assurance using programmatic methods.ResultsA total of 796 whole-body MR imaging sessions from 462 subjects were curated. A major change in scan protocol part way through the retrospective window meant that approximately 30% of available imaging sessions had properties that differed significantly from the remainder of the data. Issues were found with a vendor-supplied clinical algorithm for “composing” whole-body images from multiple imaging stations. Historic weaknesses in a digital video disk (DVD) research archive (already addressed by the mid-2010s) were highlighted by incomplete datasets, some of which could not be completely recovered. The final dataset contained 736 imaging sessions for 432 subjects. Software was written to clean and harmonise data. Implications for the subsequent machine learning activity are considered.ConclusionsMALIMAR exemplifies the vital role that curation plays in machine learning studies that use real-world data. A research repository such as XNAT facilitates day-to-day management, ensures robustness and consistency and enhances the value of the final dataset. The types of process described here will be vital for future large-scale multi-institutional and multi-national imaging projects.Critical relevance statementThis article showcases innovative data curation methods using a state-of-the-art image repository platform; such tools will be vital for managing the large multi-institutional datasets required to train and validate generalisable ML algorithms and future foundation models in medical imaging.Key points• Heterogeneous data in the MALIMAR study required the development of novel curation strategies.• Correction of multiple problems affecting the real-world data was successful, but implications for machine learning are still being evaluated.• Modern image repositories have rich application programming interfaces enabling data enrichment and programmatic QA, making them much more than simple “image marts”.Graphical

Dorsal-to-palmar branch neuroanastomosis in horses undergoing palmar digital neurectomy does not reduce neuroma formation or improve outcome.

Chronic foot pain, a common cause of forelimb lameness, can be treated by palmar digital neurectomy (PDN). Complications include neuroma formation and lameness recurrence. In humans, neuroanastomoses are performed to prevent neuroma formation. The aim of the study was to evaluate the outcome of horses undergoing dorsal-to-palmar branch neuroanastomosis following PDN. Retrospective case series. Eighty-five horses with PDN and dorsal-to-palmar branch neuroanastomosis. Medical records for horses undergoing this procedure at two hospitals between 2015 and 2020 were reviewed. Palmar and dorsal nerve branches of the PDN were transected and end-to-end neuroanastomosis was performed by apposition of the perineurium. Follow-up was obtained from medical records and telephone interviews. Success was defined as resolution of lameness for at least one year. Lameness resolved following surgery in 81/85 (95%) horses with 57/84 (68%) sound at one year. Postoperative complications occurred in 19/85 (22%) cases. The main limitations of the study were an incomplete data set, inaccurate owner recall, and variations in procedure. Compared to previous studies, this technique resulted in similar numbers of horses sound immediately after surgery, a comparable rate of postoperative neuroma formation but a higher recurrence of lameness rate at 1 year postoperatively. End-to-end neuroanastomosis of the dorsal and palmar branches of the PDN does not reduce the rate of neuroma formation in horses. Long-term outcome was less favorable compared to previously reported PDN techniques.

Explainable Learning-Based Timeout Optimization for Accurate and Efficient Elephant Flow Prediction in SDNs

Accurately and efficiently predicting elephant flows (elephants) is crucial for optimizing network performance and resource utilization. Current prediction approaches for software-defined networks (SDNs) typically rely on complete traffic and statistics moving from switches to controllers. This leads to an extra control channel bandwidth occupation and network delay. To address this issue, this paper proposes a prediction strategy based on incomplete traffic that is sampled by the timeouts for the installation or reactivation of flow entries. The strategy involves assigning a very short hard timeout (Tinitial) to flow entries and then increasing it at a rate of r until flows are identified as elephants or out of their lifespans. Predicted elephants are switched to an idle timeout of 5 s. Logistic regression is used to model elephants based on a complete dataset. Bayesian optimization is then used to tune the trained model Tinitial and r over the incomplete dataset. The process of feature selection, model learning, and optimization is explained. An extensive evaluation shows that the proposed approach can achieve over 90% generalization accuracy over 7 different datasets, including campus, backbone, and the Internet of Things (IoT). Elephants can be correctly predicted for about half of their lifetime. The proposed approach can significantly reduce the controller–switch interaction in campus and IoT networks, although packet completion approaches may need to be applied in networks with a short mean packet inter-arrival time.

Hotter is not (always) better: Embracing unimodal scaling of biological rates with temperature.

Rate-temperature scaling relationships have fascinated biologists for nearly two centuries and are increasingly important in our era of global climate change. These relationships are hypothesized to originate from the temperature-dependent kinetics of rate-limiting biochemical reactions of metabolism. Several prominent theories have formalized this hypothesis using the Arrhenius model, which characterizes a monotonic temperature dependence using an activation energy E. However, the ubiquitous unimodal nature of biological temperature responses presents important theoretical, methodological, and conceptual challenges that restrict the promise for insight, prediction, and progress. Here we review the development of key hypotheses and methods for the temperature-scaling of biological rates. Using simulations, we examine the constraints of monotonic models, illustrating their sensitivity to data nuances such as temperature range and noise, and their tendency to yield variable and underestimated E, with critical consequences for climate change predictions. We also evaluate the behaviour of two prominent unimodal models when applied to incomplete and noisy datasets. We conclude with recommendations for resolving these challenges in future research, and advocate for a shift to unimodal models that better characterize the full range of biological temperature responses.

Refining Graph Structure for Incomplete Multi-View Clustering.

As a challenging problem, incomplete multi-view clustering (MVC) has drawn much attention in recent years. Most of the existing methods contain the feature recovering step inevitably to obtain the clustering result of incomplete multi-view datasets. The extra target of recovering the missing feature in the original data space or common subspace is difficult for unsupervised clustering tasks and could accumulate mistakes during the optimization. Moreover, the biased error is not taken into consideration in the previous graph-based methods. The biased error represents the unexpected change of incomplete graph structure, such as the increase in the intra-class relation density and the missing local graph structure of boundary instances. It would mislead those graph-based methods and degrade their final performance. In order to overcome these drawbacks, we propose a new graph-based method named Graph Structure Refining for Incomplete MVC (GSRIMC). GSRIMC avoids recovering feature steps and just fully explores the existing subgraphs of each view to produce superior clustering results. To handle the biased error, the biased error separation is the core step of GSRIMC. In detail, GSRIMC first extracts basic information from the precomputed subgraph of each view and then separates refined graph structure from biased error with the help of tensor nuclear norm. Besides, cross-view graph learning is proposed to capture the missing local graph structure and complete the refined graph structure based on the complementary principle. Extensive experiments show that our method achieves better performance than other state-of-the-art baselines.

Effect of body mass index on survival after spinal cord injury.

Increased mortality after acute and chronic spinal cord injury (SCI) remains a challenge and mandates a better understanding of the factors contributing to survival in these patients. This study investigated whether body mass index (BMI) measured after acute traumatic SCI is associated with a change in mortality. A prospective longitudinal cohort study was conducted with 742 patients who were admitted to the Acute Spine Unit of the Vancouver General Hospital between 2004 and 2016 with a traumatic SCI. An investigation of the association between BMI on admission and long-term mortality was conducted using classification and regression tree (CART) and generalized additive models (spline curves) from acute care up to 7.7 years after SCI (chronic phase). Multivariable models were adjusted for (i) demographic factors (e.g., age, sex, and Charlson Comorbidity Index) and (ii) injury characteristics (e.g., neurological level and severity and Injury Severity Score). After the exclusion of incomplete datasets (n = 602), 643 patients were analyzed, of whom 102 (18.5%) died during a period up to 7.7 years after SCI. CART identified three distinct mortality risk groups: (i) BMI: > 30.5 kg/m2, (ii) 17.5-30.5 kg/m2, and (iii) < 17.5 kg/m2. Mortality was lowest in the high BMI group (BMI > 30.5 kg/m2), followed by the middle-weight group (17.5-30.5 kg/m2), and was highest in the underweight group (BMI < 17.5 kg/m2). High BMI had a mild protective effect against mortality after SCI (hazard ratio 0.28, 95% CI: 0.09-0.88, p = 0.029), concordant with a modest "obesity paradox". Moreover, being underweight at admission was a significant risk factor for mortality up to 7.7 years after SCI (hazard ratio 5.5, 95% CI: 2.34-13.17, p < 0.001). Mortality risk (1 month to 7.7 years after SCI) was associated with differences in BMI at admission. Further research is needed to better understand the underlying mechanisms. Given an established association of BMI with metabolic determinants, these results may suggest unknown neuro-metabolic pathways that are crucial for patient survival.

Emotional dysregulation and its pathways to suicidality in a community-based sample of adolescents

ObjectiveEffective suicide prevention for adolescents is urgently needed but difficult, as suicide models lack a focus on age-specific influencing factors such as emotional dysregulation. Moreover, examined predictors often do not specifically consider the contribution to the severity of suicidality.To determine which adolescents are at high risk of more severe suicidality, we examined the association between emotional dysregulation and severity of suicidality directly as well as indirectly via depressiveness and nonsuicidal self-injury.MethodAdolescents from 18 high schools in Bavaria were included in this cross-sectional and questionnaire-based study as part of a larger prevention study. Data were collected between November 2021 and March 2022 and were analyzed from January 2023 to April 2023.Students in the 6th or 7th grade of high school (11–14 years) were eligible to participate. A total of 2350 adolescents were surveyed and data from 2117 students were used for the analyses after excluding incomplete data sets. Our main outcome variable was severity of suicidality (Paykel Suicide Scale, PSS). Additionally, we assessed emotional dysregulation (Difficulties in Emotion Regulation Scale, DERS-SF), depressiveness (Patient Health Questionnaire, PHQ-9) and nonsuicidal self-injury (Deliberate Self-Harm Inventory, DSHI).ResultsIn total, 2117 adolescents (51.6% female; mean age, 12.31 years [standard deviation: 0.67]) were included in the structural equation model (SEM). Due to a clear gender-specific influence, the model was calculated separately for male and female adolescents. For male adolescents, there was a significant indirect association between emotional dysregulation and severity of suicidality, mediated by depressiveness (β = 0.15, SE = .03, p = .008). For female adolescents, there was a significant direct path from emotional dysregulation to severity of suicidality and also indirect paths via depressiveness (β = 0.12, SE = .05, p = 0.02) and NSSI (β = 0.18, SE = .04, p < .001).ConclusionsOur results suggest that gender-related risk markers in 11–14-year-olds need to be included in future suicide models to increase their predictive power. According to our findings, early detection and prevention interventions based on emotion regulation skills might be enhanced by including gender-specific adjustments for the co-occurrence of emotional dysregulation, depressiveness, and nonsuicidal self-injury in girls and the co-occurrence of emotional dysregulation and depressiveness in boys.

A New Approach to Current Transformer Saturation Compensation for Transmission Line Differential Protection

A non-linear autoregressive network with external inputs (NARX) model to reconstruct the current transformer (CT) secondary waveform distortion in the presence of protected transmission lines is presented. The qualitative analysis carried out on the autonomous model reveals the accurate performance of the current differential relay, because of the ability of the NARX to make accurate predictions and signal reconstruction even when faced with noisy or incomplete datasets. The suggested method uses samples of the line-protected remote current signal to produce a desired waveform. The unsaturated section is artificially expanded because it can estimate up to one sample after saturation with a respectable level of accuracy. The proposed method is evaluated by MATLAB and PSCAD software. The effectiveness of the suggested method for different fault types under varied situations has been proved. The simulation outcome confirms the method’s high accuracy concerning the unusual system conditions in the presence of CT saturation.

Skeleton-based coordinate system construction method for non-cooperative targets

Due to the lack of structure and part information in non-cooperative targets, the task of pose estimation suffers from unstable and unreliable initial coordinate system. In this paper, we introduce a curve-skeleton-based method to construct the robust coordinate system automatically for non-cooperative targets. First, a nonmanifold-Laplacian-based method is proposed to extract the curve skeleton from point clouds. The proposed method solves the over-contraction problem while preserves the topology and structure of skeleton well during point clouds contraction. Additionally, we create the skeleton graph to detect the skeleton structure and key points automatically. Then, the detected skeleton feature is used to construct target coordinate system, and improve the stability and robustness. Specifically, a real-scan dataset is constructed in this paper, and the proposed point cloud contraction method outperforms previous methods in structural representation and visualization. The mean rotation error of proposed coordinate system construct method performs 0.1249° and 1.2135° in 0%–30% noise ratio of distortions and outliers, and performs 7.04° and 4.21° in real-scan aircraft and satellite datasets. The results in complete and incomplete datasets verify the robustness and reliability of our skeleton-based coordinate system.

Deep learning based decision tree ensembles for incomplete medical datasets.

In practice, the collected datasets for data analysis are usually incomplete as some data contain missing attribute values. Many related works focus on constructing specific models to produce estimations to replace the missing values, to make the original incomplete datasets become complete. Another type of solution is to directly handle the incomplete datasets without missing value imputation, with decision trees being the major technique for this purpose. To introduce a novel approach, namely Deep Learning-based Decision Tree Ensembles (DLDTE), which borrows the bounding box and sliding window strategies used in deep learning techniques to divide an incomplete dataset into a number of subsets and learning from each subset by a decision tree, resulting in decision tree ensembles. Two medical domain problem datasets contain several hundred feature dimensions with the missing rates of 10% to 50% are used for performance comparison. The proposed DLDTE provides the highest rate of classification accuracy when compared with the baseline decision tree method, as well as two missing value imputation methods (mean and k-nearest neighbor), and the case deletion method. The results demonstrate the effectiveness of DLDTE for handling incomplete medical datasets with different missing rates.

Non-contact reconstitution of the traction distribution using incomplete deformation measurements: Methodology and experimental validation

Traction distribution is often unknown in solid mechanics. However, identifying traction distribution with both high accuracy and spatial resolution can sometimes be critical. In this paper, we consider the frequent situation where incomplete deformation measurements, such as partial surface measurements obtained by digital image correlation, are available. For such situations, we propose to achieve the complete identification of boundary tractions, using an optimization based inverse method. The feasibility and robustness of the proposed approach are firstly assessed with numerical simulations. We show that complete traction distributions can be identified with high accuracy even when there is noise added to the deformation measurements. More specifically, the relative error was less than 3% even with 10% noise. Moreover, the effect of noise depends significantly on the level of incompleteness of the deformation data. We show for instance that when the measurement area is 8.3% of the solid, the relative error can be 5 times larger than when the measurement area is 33.3%. However, even with 8.3% deformation measurement area of the solid, the relative error of the mapped traction distribution can be less than 6%. Besides, the accuracy of the mapped traction distribution can be further improved if the region of the deformation measurement is closer to the traction boundary. We eventually validate the proposed inverse approach using incomplete experimental datasets consisting in tensile tests carried out on two elastic samples. We obtain very promising results, with relative errors below 4% using the measured displacements of only 3 points, which is promising to open potential applications in different fields of engineering.

Is Subcapital Realignment for Slipped Capital Femoral Epiphysis a Reproducible Technique? Midterm Results.

The modified Dunn procedure, which is based on the development of an extended retinacular flap containing the blood supply for the femoral head, allows anatomic reestablishment in patients with moderate to severe slipped capital femoral epiphysis (SCFE). Some controversy exists regarding the short-term to midterm risk of avascular necrosis (AVN) and other complications resulting from the surgical technique. (1) What percentage of patients treated with an extended retinacular flap during the modified Dunn procedure for SCFE with a moderate (slip angle from 30° to 60°) or severe slip (slip angle equal or greater than 60°) develop symptomatic AVN, and what percentage underwent further surgery or had other complications? (2) What femoral head-neck alignment and position parameters relative to the greater trochanter are achieved after surgery? (3) Can we identify radiographic signs of osteoarthritis at a minimum of 4 years after surgery? Between January 2006 and December 2018, we treated 61 patients for SCFE. During this time, we generally used the modified Dunn procedure when the slip angle was ≥ 30°. Based on this indication, the modified Dunn procedure was performed in 37 patients (41 hips) during that time period, and those patients were potentially eligible for this retrospective study. Because bilateral hips in the same patient are not statistically independent, for our analyses, we analyzed only the hip with the longer follow-up time. Of those who remained, 11% (4) were lost before the minimum study follow-up of 48 months or had incomplete datasets, leaving 89% (33) for analysis here at a median follow-up of 80 months (range 49 to 208 months). Periacetabular osteotomy or subtrochanteric rotational femoral osteotomy was added under the same anaesthesia time in 33% (11 patients). We added periacetabular osteotomy in 24% (8) when intraoperative anterior instability was present in external rotation. A femoral rotational osteotomy was added in 9% (3) when posterior instability was present in flexion and internal rotation. The mean ± standard deviation age at the time of surgery was 13 ± 1 years, and 33% (11 of 33) of patients were female. The mean slip angle was 51° ± 16º, and 15% (5) of hips had unstable slips, defined as an inability to walk with or without crutches. We documented chronic presentations in 82% (27) of patients, acute and chronic in 12% (4), and acute in 6% (2). The rate of symptomatic AVN was determined by reviewing all radiographs obtained at the latest follow-up interval. Further surgery and other complications were assessed through an electronic medical record review. Radiographic morphologic parameters were measured before surgery and at a minimum follow-up of 4 years by two senior orthopaedic surgeons. Radiographs obtained at the latest follow-up visit were also screened for signs of osteoarthritis by the same surgeons. At the latest follow-up, 3% (1 patient) of patients developed symptomatic AVN and underwent further surgery and 3% (1) underwent revision surgery for screw breakage after a high-energy fall. Postoperatively, the alpha angle was restored to 39º ± 6º, the anterior head-neck offset was restored to 8 ± 3 mm, the neck-shaft angle was 136º ± 6º, and the presence of a positive Klein line decreased from 64% (21 hips) to 0% (0 hips). No patients showed radiographic signs of osteoarthritis at the minimum follow-up of 4 years. In this series, the modified Dunn procedure in moderate and severe slips was a reproducible procedure, and few patients developed symptomatic AVN or experienced other complications. Hip morphology was restored, but a longer follow-up duration and a detailed analysis of the results from other centers is warranted to assess the possible long-term risk of progression to AVN or osteoarthritis. Level IV, therapeutic study.

Analytical Bayesian Copula‐Based Uncertainty Quantification (A‐BASIC‐UQ) Using Data with Missing Values in Structural Health Monitoring

The presence of missing values is common in real‐world datasets, so modeling and uncertainty quantification (UQ) of incomplete datasets have gained increasing attention in various research areas, including structural health monitoring (SHM). However, modeling and UQ utilizing incomplete datasets are nontrivial tasks. On the other hand, prediction based on a set of incomplete measured input variables is also an important task, but most existing methods, which are discriminative models, do not possess this capability. Aiming to tackle these two challenges, we propose the two‐stage analytical Bayesian copula‐based uncertainty quantification (A‐BASIC‐UQ) using incomplete SHM data. In the modeling stage, the copula‐based multivariate joint probability density function (PDF) is modeled directly according to an incomplete dataset without imputation or disposal of any data points. For the univariate marginal PDF, using the measured (nonmissing) values of the corresponding random variable (RV), Bayesian model class selection is conducted to select the most suitable model class. For the Gaussian copula PDF, using the bivariate complete data points of entry‐by‐entry pairwise data, the optimal parameter vector is obtained from the estimation of the Pearson correlation coefficient. In the prediction stage, the analytical expressions of the predictive PDF, the predicted value and the credible region of the output variables are derived according to a set of incomplete measured input variables. The analytical expression of the predictive PDF is obtained based on the analytical operations on the auxiliary RVs and that of the predicted value and the credible region are obtained based on the analysis of multivariate Gaussian distribution. Therefore, the proposed method does not require numerical integration nor Monte Carlo simulation and does not suffer from computational burden even when there are many variables (say 4 or above). Examples using simulated data and real SHM data are presented to illustrate the capability of the proposed A‐BASIC‐UQ.

Taking full advantage of ‘low-quality’ datasets in watershed modeling and management: From a perspective of parameter calibration

The quality of calibration datasets is critical for establishing well-calibrated models for reliable decision-making support. However, the analysis of the influence of calibration dataset quality and the discussion on how to use flawed and/or incomplete datasets are still far from sufficient. An evaluation framework for the impact of model calibration data on parameter identifiability, sensitivity, and uncertainty (ISU) was established. Three quantitative and normalized indicators were designed to describe the magnitude of ISU. With the case study of the upper Daqing River watershed, China and the model SWAT (Soil and Water Assessment Tool), one ideal dataset without quality flaws and 79 datasets with different types of flaws including observation error, low monitoring frequency, short data duration and low data resolution were evaluated. The result showed that 4 of 13 parameters that control canopy, groundwater and channel processes have higher ISU values, indicating the high identifiability, high sensitivity, and low uncertainty. The largest gap of parameter ISU between dataset with quality flaw and ideal dataset was 0.61 due to short data duration, while the smallest gap was −0.28 due to low monitoring data frequency. Although some defective datasets caused unacceptable calibration results and model output, some defective datasets can still be valuable for model calibration which depends on the hydrological processes of interest when applying the model. Equivalent calibration results were yielded by the datasets with similar statistical properties. When using datasets with traditional defective issues for calibration, a new step checking the consistency among decision goal, representative system process, determinative parameters and calibration datasets is suggested. Practices including process-related data selection, dataset regrouping and risk self-reporting when using low-quality datasets are encouraged to increase the reliability of model-based watershed management.

A Chaotic Maps-Based Privacy-Preserving Distributed Deep Learning for Incomplete and Non-IID Datasets

A Chaotic Maps-Based Privacy-Preserving Distributed Deep Learning for Incomplete and Non-IID Datasets

Tensor Learning Meets Dynamic Anchor Learning: From Complete to Incomplete Multiview Clustering.

Multiview clustering (MVC), which can dexterously uncover the underlying intrinsic clustering structures of the data, has been particularly attractive in recent years. However, previous methods are designed for either complete or incomplete multiview only, without a unified framework that handles both tasks simultaneously. To address this issue, we propose a unified framework to efficiently tackle both tasks in approximately linear complexity, which integrates tensor learning to explore the inter-view low-rankness and dynamic anchor learning to explore the intra-view low-rankness for scalable clustering (TDASC). Specifically, TDASC efficiently learns smaller view-specific graphs by anchor learning, which not only explores the diversity embedded in multiview data, but also yields approximately linear complexity. Meanwhile, unlike most current approaches that only focus on pair-wise relationships, the proposed TDASC incorporates multiple graphs into an inter-view low-rank tensor, which elegantly models the high-order correlations across views and further guides the anchor learning. Extensive experiments on both complete and incomplete multiview datasets clearly demonstrate the effectiveness and efficiency of TDASC compared with several state-of-the-art techniques.

A Multi-Level Branch Neural Networks With Self-Evolution for Condition Monitoring of Industrial Equipment Under Incomplete Training Dataset Scenario

A Multi-Level Branch Neural Networks With Self-Evolution for Condition Monitoring of Industrial Equipment Under Incomplete Training Dataset Scenario

A conditionally positive definite kernel function for clustering of incomplete data

Clustering of incomplete data sets that contains missing features is one of the most widely studied problems in the literature, and several imputation and non-imputation techniques are used to solve this problem. A weighted sum of the Euclidean distance from the datum to the corresponding clusters is used in Fuzzy c-means clustering. It has been observed that the kernel-based clustering techniques outperform the conventional algorithms in terms of accuracy. This is due to their ability to handle non-linear data and map it to higher dimensional space while preserving its internal structure. Kernel functions are really important when it comes to the performance of kernel-based clustering methods. Choosing the right kernel function isn’t simple. Among the various clustering algorithms that have been examined in the literature, the, Gaussian kernel function has been found to be more useful.. This paper suggests a conditionally positive definite kernel function that can be used in the unsupervised clustering of incomplete data. Numerical analysis shows that the conditionally positive definite kernel function also performs well on datasets with incomplete features.

Generative AI-based Data Completeness Augmentation Algorithm for Data-driven Smart Healthcare.

In the decade, artificial intelligence has achieved great popularity and applications in medicine and healthcare. Various AI-based algorithms have shown astonishing performance. However, in various data-driven smart healthcare algorithms, the problem of incomplete dataset remains a huge challenge. In this paper, we propose a data completeness enhancement algorithm based on generative AI (i.e., GenAI-DAA) to solve the problems of the in-sufficient data for model training, the data imbalance, and the biases of the training samples. We first construct the cognitive field of the generative models and effectively understand the state of incomplete cognition in generative models. Secondly, on this basis, we propose a quest algorithm for abnormal samples in the cognitive field based on local outlier factor. By fine-grained value evaluation, abnormal samples are given more refined attention. Finally, integrating the above process through multiple cognitive adjustments, GenAI-DAA gradually improves the cognitive ability. GenAI-DAA can be summarized as "Quest-→Estimate-→Tune-up". We have conducted extensive experiments to demonstrate the effectiveness of our proposed algorithm, and shown widely applications to some typical data-driven smart healthcare algorithms.