Completeness of data is vital for the decision making and forecasting on Building Management Systems (BMS) as missing data can result in biased decision making down the line. This study creates a guideline for imputing the gaps in BMS datasets by comparing four methods: K Nearest Neighbour algorithm (KNN), Recurrent Neural Network (RNN), Hot Deck (HD) and Last Observation Carried Forward (LOCF). The guideline contains the best method per gap size and scales of measurement. The four selected methods are from various backgrounds and are tested on a real BMS and metereological dataset. The focus of this paper is not to impute every cell as accurately as possible but to impute trends back into the missing data. The performance is characterised by a set of criteria in order to allow the user to choose the imputation method best suited for its needs. The criteria are: Variance Error (VE) and Root Mean Squared Error (RMSE). VE has been given more weight as its ability to evaluate the imputed trend is better than RMSE. From preliminary results, it was concluded that the best K‐values for KNN are 5 for the smallest gap and 100 for the larger gaps. Using a genetic algorithm the best RNN architecture for the purpose of this paper was determined to be GatedRecurrent Units (GRU). The comparison was performed using a different training dataset than the imputation dataset. The results show no consistent link between the difference in Kurtosis or Skewness and imputation performance. The results of the experiment concluded that RNN is best for interval data and HD is best for both nominal and ratio data. There was no single method that was best for all gap sizes as it was dependent on the data to be imputed.
MULTIFILE
Completeness of data is vital for the decision making and forecasting on Building Management Systems (BMS) as missing data can result in biased decision making down the line. This study creates a guideline for imputing the gaps in BMS datasets by comparing four methods: K Nearest Neighbour algorithm (KNN), Recurrent Neural Network (RNN), Hot Deck (HD) and Last Observation Carried Forward (LOCF). The guideline contains the best method per gap size and scales of measurement. The four selected methods are from various backgrounds and are tested on a real BMS and meteorological dataset. The focus of this paper is not to impute every cell as accurately as possible but to impute trends back into the missing data. The performance is characterised by a set of criteria in order to allow the user to choose the imputation method best suited for its needs. The criteria are: Variance Error (VE) and Root Mean Squared Error (RMSE). VE has been given more weight as its ability to evaluate the imputed trend is better than RMSE. From preliminary results, it was concluded that the best K‐values for KNN are 5 for the smallest gap and 100 for the larger gaps. Using a genetic algorithm the best RNN architecture for the purpose of this paper was determined to be Gated Recurrent Units (GRU). The comparison was performed using a different training dataset than the imputation dataset. The results show no consistent link between the difference in Kurtosis or Skewness and imputation performance. The results of the experiment concluded that RNN is best for interval data and HD is best for both nominal and ratio data. There was no single method that was best for all gap sizes as it was dependent on the data to be imputed.
DOCUMENT
This paper investigate to use of information technology, i.e. machine learning algorithms for water assessment in Timor-Leste. It is essential to access clean water to ensure the safety for humans and others livings in this world. The Water Quality Index (WQI) is the standard tool for assessing water quality, which can be calculated from physicochemical and microbiological parameters. However, in developing countries, it is continuing need to bring water and energy for the most disadvantaged, make it necessary to find new solutions. In such case, missing-value imputation and machine learning models are useful for classifying water samples into suitable or unsuitable with significant accuracy. Some imputation methods were tested, and four machine learning algorithms were explored: logistic regression, support vector machine, random forest, and Gaussian naïve Bayes. We obtained a dataset with 368 observations from 26 groundwater sampling points in Dili city of Timor-Leste. According to experimental results, it is found that 64% of the water samples are suitable for human consumption. We also found k-NN imputation and random forest method were the clear winners, achieving 96% accuracy with three-fold cross validation. The analysis revealed that some parameters significantly affected the classification results.
DOCUMENT
Research on follow-up outcomes of systemic interventions for family members with an intellectual disability is scarce. In this study, short-term and long-term follow-up outcomes of multisystemic therapy for adolescents with antisocial or delinquent behaviour and an intellectual disability (MST-ID) are reported. In addition, the role of parental intellectual disability was examined. Outcomes of 55 families who had received MST-ID were assessed at the end of treatment and at 6-month, 12-month and 18-month follow-up. Parental intellectual disability was used as a predictor of treatment outcomes. Missing data were handled using multiple imputation. Rule-breaking behaviour of adolescents declined during treatment and stabilized until 18 months post-treatment. The presence or absence of parental intellectual disability did not predict treatment outcomes. This study was the first to report long-term outcomes of MST-ID. The intervention achieved similar results in families with and without parents with an intellectual disability.
DOCUMENT
Thirty to sixty per cent of older patients experience functional decline after hospitalisation, associated with an increase in dependence, readmission, nursing home placement and mortality. First step in prevention is the identification of patients at risk. The objective of this study is to develop and validate a prediction model to assess the risk of functional decline in older hospitalised patients.
DOCUMENT
Trustworthy data-driven prognostics in gas turbine engines are crucial for safety, cost-efficiency, and sustainability. Accurate predictions depend on data quality, model accuracy, uncertainty estimation, and practical implementation. This work discusses data quality attributes to build trust using anonymized real-world engine data, focusing on traceability, completeness, and representativeness. A significant challenge is handling missing data, which introduces bias and affects training and predictions. The study compares the accuracy of predictions using Exhaust Gas Temperature (EGT) margin, a key health indicator, by keeping missing values, using KNN-imputation, and employing a Generalized Additive Model (GAM). Preliminary results indicate that while KNN-imputation can be useful for identifying general trends, it may not be as effective for specific predictions compared to GAM, which considers the context of missing data. The choice of method depends on the study’s objective: broad trend forecasting or specific event prediction, each requiring different approaches to manage missing data.
DOCUMENT
BACKGROUND: Medication-related problems are common after hospitalization, for example when changes in patients' medication regimens are accompanied by insufficient patient education, poor information transfer between healthcare providers, and inadequate follow-up post-discharge. We investigated the effect of a pharmacy-led transitional care program on the occurrence of medication-related problems four weeks post-discharge.METHODS: A prospective multi-center before-after study was conducted in six departments in total of two hospitals and 50 community pharmacies in the Netherlands. We tested a pharmacy-led program incorporating (i) usual care (medication reconciliation at hospital admission and discharge) combined with, (ii) teach-back at hospital discharge, (iii) improved transfer of medication information to primary healthcare providers and (iv) post-discharge home visit by the patient's own community pharmacist, compared with usual care alone. The difference in medication-related problems four weeks post-discharge, measured by means of a validated telephone-interview protocol, was the primary outcome. Multiple logistic regression analysis was used, adjusting for potential confounders after multiple imputation to deal with missing data.RESULTS: We included 234 (January-April 2016) and 222 (July-November 2016) patients in the usual care and intervention group, respectively. Complete data on the primary outcome was available for 400 patients. The proportion of patients with any medication-related problem was 65.9% (211/400) in the usual care group compared to 52.4% (189/400) in the intervention group (p = 0.01). After multiple imputation, the proportion of patients with any medication-related problem remained lower in the intervention group (unadjusted odds ratio 0.57; 95% CI 0.38-0.86, adjusted odds ratio 0.50; 95% CI 0.31-0.79).CONCLUSIONS: A pharmacy-led transitional care program reduced medication-related problems after discharge. Implementation research is needed to determine how best to embed these interventions in existing processes.
DOCUMENT
Background—Self-management interventions are widely implemented in care for patients with heart failure (HF). Trials however show inconsistent results and whether specific patient groups respond differently is unknown. This individual patient data meta-analysis assessed the effectiveness of self-management interventions in HF patients and whether subgroups of patients respond differently. Methods and Results—Systematic literature search identified randomized trials of selfmanagement interventions. Data of twenty studies, representing 5624 patients, were included and analyzed using mixed effects models and Cox proportional-hazard models including interaction terms. Self-management interventions reduced risk of time to the combined endpoint HF-related all-0.71- in Conclusions—This study shows that self-management interventions had a beneficial effect on time to HF-related hospitalization or all-cause death, HF-related hospitalization alone, and elicited a small increase in HF-related quality of life. The findings do not endorse limiting selfmanagement interventions to subgroups of HF patients, but increased mortality in depressed patients warrants caution in applying self-management strategies in these patients.
DOCUMENT
Background: Knee and hip osteoarthritis (OA) among older adults account for substantial disability and extensive healthcare use. Effective pain coping strategies help to deal with OA. This study aims to determine the long-term relationship between pain coping style and the course of healthcare use in patients with knee and/or hip OA over 10 years. Methods: Baseline and 10-year follow-up data of 861 Dutch participants with early knee and/or hip OA from the Cohort Hip and Cohort Knee (CHECK) cohort were used. The amount of healthcare use (HCU) and pain coping style were measured. Generalized Estimating Equations were used, adjusted for relevant confounders. Results: At baseline, 86.5% of the patients had an active pain coping style. Having an active pain coping style was significantly (p = 0.022) associated with an increase of 16.5% (95% CI, 2.0–32.7) in the number of used healthcare services over 10 years. Conclusion: Patients with early knee and/or hip OA with an active pain coping style use significantly more different healthcare services over 10 years, as opposed to those with a passive pain coping style. Further research should focus on altered treatment (e.g., focus on self-management) in patients with an active coping style, to reduce HCU.
DOCUMENT
Objectives: Decline in the performance of instrumental activities of daily living (IADL) and mobility may be preceded by symptoms the patient experiences, such as fatigue. The aim of this study is to investigate whether self-reported non-task-specific fatigue is a long-term risk factor for IADL-limitations and/or mobility performance in older adults after 10 years. Methods: A prospective study from two previously conducted cross-sectional studies with 10-year follow-up was conducted among 285 males and 249 females aged 40–79 years at baseline. Fatigue was measured by asking “Did you feel tired within the past 4 weeks?” (males) and “Do you feel tired?” (females). Self-reported IADLs were assessed at baseline and follow-up. Mobility was assessed by the 6-minute walk test. Gender-specific associations between fatigue and IADL-limitations and mobility were estimated by multivariable logistic and linear regression models. Results: A total of 18.6% of males and 28.1% of females were fatigued. After adjustment, the odds ratio for fatigued versus non-fatigued males affected by IADL-limitations was 3.3 (P=0.023). In females, the association was weaker and not statistically significant, with odds ratio being 1.7 (P=0.154). Fatigued males walked 39.1 m shorter distance than those non-fatigued (P=0.048). For fatigued females, the distance was 17.5 m shorter compared to those non-fatigued (P=0.479). Conclusion: Our data suggest that self-reported fatigue may be a long-term risk factor for IADL-limitations and mobility performance in middle-aged and elderly males but possibly not in females.
DOCUMENT
