Hybrid Statistical Model Research Articles

Multi sentence description of complex manipulation action videos

Automatic video description necessitates generating natural language statements that encapsulate the actions, events, and objects within a video. An essential human capability in describing videos is to vary the level of detail, a feature that existing automatic video description methods, which typically generate single, fixed-level detail sentences, often overlook. This work delves into video descriptions of manipulation actions, where varying levels of detail are crucial to conveying information about the hierarchical structure of actions, also pertinent to contemporary robot learning techniques. We initially propose two frameworks: a hybrid statistical model and an end-to-end approach. The hybrid method, requiring significantly less data, statistically models uncertainties within video clips. Conversely, the end-to-end method, more data-intensive, establishes a direct link between the visual encoder and the language decoder, bypassing any statistical processing. Furthermore, we introduce an Integrated Method, aiming to amalgamate the benefits of both the hybrid statistical and end-to-end approaches, enhancing the adaptability and depth of video descriptions across different data availability scenarios. All three frameworks utilize LSTM stacks to facilitate description granularity, allowing videos to be depicted through either succinct single sentences or elaborate multi-sentence narratives. Quantitative results demonstrate that these methods produce more realistic descriptions than other competing approaches.

Differentiating between point and non-point source nutrient loadings of wastewater in an agriculturally impacted area using a hybrid statistical model

Nutrient contamination from point and non-point sources can lead to harmful consequences, such as algal blooms. Point and non-point nutrient loading estimation is determined using modeling approaches and often require an abundance of variables and observations for calibration. Small rural streams that lack water use designations often lack available data to utilize current modeling strategies. This study proposes the use of a 3-phase hybrid stepwise statistical modeling approach using generalized linear mixed models (GLMM) and a reference stream. Two streams in Central Texas were sampled for 13 months between February 2020 and February 2021, one being impacted by a wastewater treatment plant (WWTP). Dissolved phosphorus (PO4-P), ammonia (NH3-N), nitrite/nitrate (NO2 + NO3-N), total nitrogen (TN), and total phosphorus (TP) were sampled in both streams for each month. Non-point sources of contamination, such as land use/land cover and geomorphology composition, were quantified for both sub-basin drainage areas. Phase I models predicted nutrient concentrations in the reference stream using non-point source variables along with discharge and temporal variables. Best fit models were carried forward to phase II and leveraged a point-source variable, which is a naïve estimate of effluent nutrient concentration in the absence of assimilation. Phase II model coefficients highlight the significance of point-source contamination in predicting nutrient concentration, but overall lacked the ability to make future predictions under new hydrologic regimes from WWTP intensification. Phase III models included deterministically calculating an uptake variable using the relationship between discharge and wetted widths, predicting background non-point concentrations by leveraging phase I models, and calculating future nutrient loadings from WWTP intensification. This approach predicted significant increases in nutrient concentrations under planned WWTP intensification scenarios and decreased uptake efficiencies under the new hydrologic regimes.

A Comparative Study Between Bayesian Network and Hybrid Statistical Predictive Models for Proactive Power System Network Resilience Enhancement Operational Planning

Enhancing the operational resilience of the distribution system network (DSN) proactively in a hurricane-prone region requires a pre-hurricane event DSN optimization model, built on accurate hurricane-induced DSN line fault prediction scenarios. In the past, the resilience evaluation methods such as statistical sequential and non-sequential Monte Carlo simulation (MCS) contingency-based technique, and Machine learning-based Bayesian Networks (BN) technique, have been proposed to strengthen the operational resilience of the DSN proactively against the forecasted oncoming hurricane events. However, a comparative study is largely unexplored to evaluate which of these two methods is best for proactive operational planning decision-making against the forecasted oncoming hurricane event. In this paper, the Bayesian network (BN) and combined statistical DSN’s Fragility-curve(FC)-Monte Carlo simulation (MCS)-Scenario reduction (SCENRED) predictive algorithms were developed. The DSN line fault prediction scenarios simulated leveraging the predicted oncoming hurricane Ewiniar data were utilized to perform pre-hurricane DSN optimization to proactively decrease the DSN expected load loss. The pre-event system optimization problems were formulated in a mixed integer linear programming (MILP) approach and solved using a CPLEX solver in general algebraic modelling system (GAMS) on a redesigned 48-bus DSN. The simulated initial expected load loss of 39% of 35 MWh was decreased to 35.34%, and then to 30.71% with the use of combined statistical DSN’s FC-MCS-SCENRED, and the BN-DSN predictive models. These results were validated using the Electrical transient analyzer program (ETAP).This study confirmed that the BN-DSN predictive model is a better operational planning tool compared to combined statistical DSN’s line FC-MCS-SCENRED predictive model.

Modeling SMEs Credit Default Risk: The Case of Saudi Arabia

This study assesses the credit risk of small and medium-sized enterprises (SMEs) to minimize unexpected risk events. We construct a hybrid statistical model based on factor analysis and logistic regression to predict enterprise default on loans and determine the factors predicting SMEs default. We assess the credit risk of SMEs listed on the Saudi stock market. The results indicate that the SMEs acid-test ratios are the most influential factors in predicting SMEs credit risk. Therefore, the designed logistic model can be used by financial institutions during the decision-making process of granting loans to SMEs. This study sheds light on challenging access to bank credits due to the lack of financial transparency of most Saudi SMEs.

A novel statistical downscaling approach for analyzing daily precipitation and extremes under the impact of climate change: Application to an arid region

Statistical downscaling of daily precipitation is important for assessing the impact of climate change. However, some issues (e.g., the underestimation of extremes) limit statistical downscaling methods to project daily precipitation accurately. This study develops an integrated hybrid statistical downscaling model combined with bias correction and Bayesian model averaging (named as SDBC-BMA) for improving the ability of downscaling methods. The performance of SDBC is compared with regression based statistical downscaling methods (RSDM) and hybrid statistical downscaling methods (HSDM) based on different performance metrics and extreme indices. SDBC-BMA is applied to the Amu Darya River Basin (ADRB) to demonstrate its feasibility and capability, where three GCMs and two CMIP6 scenarios are considered. Several findings can be summarized: (1) compared to RSDM and HSDM, SDBC is more robust with a higher computation ability of daily precipitation and a lower mean bias (i.e., +0.10 mm/day); (2) compared with the simple model average, the biases of extreme indices and mean daily precipitation obtained from SDBC-BMA could decrease by 49 % and 88 %, respectively; (3) the mean precipitation values would increase by 56 ± 41 mm/year (under SSP245) and 126 ± 55 mm/year (under SSP585) during 2076–2100, higher than mean precipitation in the historical period (1979–2005); (4) extreme indices under multi-GCMs (except for consecutive wet days) would increase by 20 % (under SSP245) and 22 % (under SSP585), implying that extreme events in ADRB are more frequent in the 21st century. The increasing risk of precipitation extremes would exacerbate threats to agricultural and ecological security.

A comprehensive analysis to assess the impact of nano MoS2 on the wear characteristic of Al-TiB2-Gr composite

The present study evaluates the influence of nano MoS2 particles on the wear properties of newly developed TiB2 and Graphite reinforced Aluminium composite. Micro particles of MoS2 are downsized into nano levels using a planetary ball mill. Three different compositions of specimens are fabricated through stir casting, with a constant 10% of weight of TiB2, graphite, and nano MoS2 is kept at varying amounts of 10, 15, and 20%. The dry sliding wear tests are executed by following Taguchi’s design of experiment. The wear rate and coefficient of friction are considered as responses, whereas the normal load, sliding distance and the composition of MoS2 are considered as the chief parameters with three levels. The hybridized Taguchi-Grey Relational -Principal Component Analysis mathematical model is implemented to study the effect of wear parameters and inclusion of MoS2 on the wear behavior. Mathematical and experimental results explore the increasing nano MoS2 content reduces the wear rate and coefficient of friction of composites. Analysis of variance results also acknowledge that nano MoS2 content in the composite is a remarkable parameter to impact the tribological property. The hybrid statistical model results explore that the optimum parameter to yield better tribological property are 30 N normal load, 2 km sliding distance and 20% of MoS2. Worn surfaces are analyzed using scanning electron microscopy to picturize the wear mechanism concerning the varying content of reinforcement.

Hybrid Statistical Models for Forecasting Yield of Mango and Banana in Tamil Nadu, India

Horticulture sector plays a prominent role in economic growth of India. India is the second largest producer of fruits and vegetables in the world next to China. Among the horticultural crops, fruit crops are cultivated in majority of the area in India. Fruit crops play a significant role in the economic development, nutritional security, employment generation, and total growth of country. India is major producer of mango and banana, among fruit crops. The objective of this research paper is to predicate the yield of mango and banana in Tamil Nadu using different models such as linear and nonlinear, parametric, and non-parametric statistical models. In this research, a hybrid model had been proposed, which consists of linear and nonlinear models. In this hybrid model, combination of the Autoregressive Integrated Moving Average (ARIMA) and Regression model were used. The present study was conducted in Tamil Nadu. Since, area and production of Mango and Banana are higher in Tamil Nadu. Based on results obtained production and yield of Mango and Banana were predicted for next four years.

Remanufacturing end-of-life passenger car waste sheet steel into mesh sheet: A sustainability assessment.

This study analysed the business sustainability of remanufacturing waste steel sheet from the shells of end-of-life vehicles into mesh steel sheet for manufacturing sheet-metal products. Hybrid statistical, fuzzy, and overall sustainability-index curve-fitting models were used to analyse the technical, economic, environmental, management, and social feasibility of remanufacturing, where the sales price, eco-cost savings, and CO2 emission reductions were used as typical statistical indicators. The remanufacturing process was optimised to allocate hardware for a plant recovering 480 m2/shift of waste sheet steel and producing 2851-5520 m2/shift of mesh sheet steel. Six scenarios were used to model the sustainability parameters to normalise the sustainability index values. The sustainability index of each parameter was calculated by multiplying its weight of importance by its weight of satisfaction. The highest sustainability index of 0.95 was calculated for the economic feasibility index, while the lowest sustainability index of 0.4 was calculated for the management feasibility. Remanufacturing of waste sheet steel into mesh sheet steel can be applied with an estimated overall sustainability index of 0.88.

A hybrid statistical decision-making optimization approach for groundwater vulnerability considering uncertainty

Recognizing the vulnerable areas for contamination is a feasible way to protect groundwater resources. The main contribution of the paper is developing a hybrid statistical decision-making model for evaluating the vulnerability of Shiraz aquifer, southern Iran, with modified DRASTIC (depth to the water table, net recharge, aquifer media, soil media, topography, impact of the vadose zone, and hydraulic conductivity) by using the genetic algorithm (GA), the analytical hierarchy process (AHP) method, and factorial analysis (FA). First, considering the variation of the uncertain parameters, 32 scenarios were defined to perform factorial analysis. Then using the AHP method and GA, DRASTIC parameters were rated and weighted in all scenarios. To achieve the optimal weights for parameters, the objective function in GA was maximizing the correlation coefficient between the vulnerability index and the nitrate concentration. The single and interactive effects of parameters on groundwater vulnerability were analyzed by factorial analysis. The results revealed that the net recharge had the highest single effect, and the resulted effect between net recharge and hydraulic conductivity was the most significant interactive effect on the objective function. Besides, the variation of aquifer media does not change the objective function. The application of the proposed method leads to a precise groundwater vulnerability map. This research provides valuable knowledge for assessing groundwater vulnerability and enables decision-makers to apply groundwater vulnerability information in future water resources management plans.

The Effect of Aggregating Bootstrap on the Accuracy of Neural Network System for Islamic Investment Prediction

Accurate prediction of the stock price is necessary for efficient financial decision making and reconstruction planning, especially during the COVID-19 crisis. Meanwhile, the ARIMA model with neural network approach is a hybrid statistical model that has been used widely in finance and statistics. It could solve the non-linearity problem in ARIMA. However, based on the unreliable sampling variation, the unbalance between the linearity and nonlinearity parts in the hybrid model could lead to inaccurate prediction. This matter motivates the purpose of this study, where the authors aim to balance the linearity and nonlinearity of the hybrid model towards predicting the Islamic investment during the Malaysia Movement Control Order (MCO). In this study, Islamic investment is analyzed using statistical based model of ARIMA. In order to balance linearity and nonlinearity parts of ARIMA which is outperforming the unreliable sampling variation, the aggregating bootstrap approach is used on ARIMA. The resampled linearity and nonlinearity parts will be declared as inputs of neural network system in order to predict the Islamic investment. Resampled nonlinearity part will be generated in this system and its sampling variation is examined. In addition, the performance of model procedure is estimated. It shows that the aggregating bootstrap gives smaller bias values and generates small weights in neural network system. In terms of prediction, applying the resampled procedure in neural network system eventually increases the precision of prediction estimation where it reduces the error estimation of MAE and MSE. Also, the prediction values continually align with actual values of investment at MCO phase. Balancing the linearity and nonlinearity part using the aggregating bootstrap in major procedure of prediction field contributes to high precise prediction of Islamic investment return estimation. The uncertainty of investment returns during MCO phase is a challenging phase and affects Malaysia financial trading. By considering using alternative procedure proposed in this study, i.e. helps in providing accuracy of investment prediction returns, a well-construct financial decisions and plans could be structured during MCO. However, this study limits to examine the effect of aggregating bootstrap. For further research, it is suggested to apply prediction on 10 years returns of Islamic investment using the proposed method.

A modelling approach and a case study to answer the question: What does it take to retrofit a community to net-zero energy?

In order to have an immediate impact on building energy-related emissions, existing buildings need to be addressed. To inform policy and technical decisions, detailed mathematical models are required which can explore numerous building retrofit solutions and their energy and emissions performance. This paper describes the adaptation of a hybrid statistical and engineering-based model for residential building stock in Canada to analyze community-scale energy retrofits. The modelling domain includes both envelope and mechanical retrofits, as well as district renewable energy systems. This model is then applied to a case study of converting a community of fifty 1980’s vintage single-detached homes to achieve net-zero energy. The case study demonstrated that deep envelope retrofits and fuel switching from natural gas to electric heat pump systems reduce community energy demand by 69%. Saturating available roof area with photovoltaics was able to achieve net-zero balance. By considering net-zero at the community-scale, individual buildings that did not achieve net-zero were offset by net-exporting neighbours. Annual community emissions of the retrofit reduced emissions by 95%. The analysis also highlights the significant impact on electrical infrastructure due to solar generation and energy demand mismatch.

A Multiscale Precipitation Forecasting Framework: Linking Teleconnections and Climate Dipoles to Seasonal and 24‐hr Extreme Rainfall Prediction

AbstractWe develop a hybrid statistical forecasting model for the simultaneous season‐ahead forecasting of both seasonal rainfall and the 24‐hr maximum rainfall for the upcoming season, using predictors identified through the Shared Reciprocal Nearest Neighbor approach. The model uses a generalized linear regression and a four‐parameter Beta distribution model for downscaling extremes using the predictors that were identified. A cross‐validation experiment for the last four decades in both Han‐River and Geum‐River watersheds, South Korea was performed to test the efficacy of the model. The leave‐one‐out cross‐validated seasonal precipitation forecast demonstrates a correlation that ranges from 0.69–0.78 to 0.68–0.76 for the Han‐River and Geum‐River watershed, respectively. Similarly, for the 24‐hr maximum rainfalls in the upcoming season, the cross‐validated correlations between the predicted and the observed values range from 0.67–0.73 to 0.50–0.63, for the two river basins. A discussion of the potential causes of the skill is offered.

A hybrid statistical model for ecological risk integral assessment of PAHs in sediments

For integral ecological risk assessment of polycyclic aromatic hydrocarbons (PAHs), the interdependence is always ignored by assuming concentration additivity (CA) and independent action (IA). This paper proposed a hybrid assessment model, named hierarchical Archimedean copula integral assessment (HACIA), to treat inherent relations and assess the multi-dimensional holistic ecological risks of exposure concentrations of 16 PAHs. The HACIA model couples with hierarchical Archimedean copulas (HACs) and ecological evaluation criteria, sediment quality guidelines (SQGs), which depends on copulas to construe the connotations of correlations among exposure concentrations of PAHs. The two case studies, the exposure concentrations of 16 PAHs from Taihu Lake in China and the Bay of Bengal coast in Bangladesh, respectively, are used to verify the adaptability and feasibility of HACIA model and evaluate their comprehensive environmental risk. In addition, a theoretical and empirical comparison with three frequently-used methods was provided objectively, which focused on presuppositions, principles, evaluations and consequences. The results indicated that the HACIA model has good feasibility and applicability to evaluate the synthetic ecological risk according to PAHs exposure concentrations, remedies for the situation that the interdependence among other methods is neglected, and can produce more deliberate classification of hazards theoretically and better demonstrations of assessment results than the other three. Although the more accurate risk assessments rely on detecting quantity of samples, the integral ecological risk assessment based on the proposed HACIA model is reliable and robust because of its reasonable structure.

Improved DDoS Attacks detection using Hybrid Statistical Model and sparse representation for Cloud Computing

While hosting various cloud based information technology facilities by handling various assets on the internet, Cloud service accessibility has remained one of the chief concerns of cloud service providers (CSP). Several security concerns associated to cloud computing service simulations, and cloud’s major qualities contribute towards its susceptibility of security threats related with cloud service availability, the liability of internet, and the dispense behavior of cloud computing. Distributed Denial of Service (DDoS) attacks is one of the main advanced threats that occur to be extremely problematic and stimulating to stand owing towards its dispersed behavior and resulted in cloud service interruption. Although there exist amount of interruption recognition resolutions anticipated by various investigation groups, there exists not at all such a faultless result that avoids the DDoS attack and cloud service providers (CSP) are presently consuming various detection resolutions by assuring that their product stays well protected. The features of DDoS attack consuming various forms with dissimilar scenarios make it problematic to identify. Inspecting and analyzing various surviving DDoS detecting methods contrary to several factors is accomplished by this paper. To enhance the system performance further, sparse based data optimization is proposed to remove the redundant data. This enhancement reduced the execution time of the system by0.2%.

A Search Method to Support Temporal Transcriptome Analysis.

Recently, there has been an increasing interest in mining time series databases with a focus on data representation. We propose a hybrid statistical and temporal logic model to yield search pattern that can be used for pattern matching and database search to identify novel and common patterns in temporal expression experiments. The method accounts for various challenges that can be found in publicly available gene expression databases.

Statistical and Hybrid Methods Implemented in a Web Application for Predicting Reservoir Inflows during Flood Events

Reservoir management is a critical component of flood management, and information on reservoir inflows is particularly essential for reservoir managers to make real-time decisions given that flood conditions change rapidly. This study's objective is to build real-time data-driven services that enable managers to rapidly estimate reservoir inflows from available data and models. We have tested the services using a case study of the Texas flooding events in the Lower Colorado River Basin in November 2014 and May 2015, which involved a sudden switch from drought to flooding. We have constructed two prediction models: a statistical model for flow prediction and a hybrid statistical and physics-based model that estimates errors in the flow predictions from a physics-based model. The study demonstrates that the statistical flow prediction model can be automated and provides acceptably accurate short-term forecasts. However, for longer term prediction (2 h or more), the hybrid model fits the observations more closely than the purely statistical or physics-based prediction models alone. Both the flow and hybrid prediction models have been published as Web services through Microsoft's Azure Machine Learning (AzureML) service and are accessible through a browser-based Web application, enabling ease of use by both technical and nontechnical personnel.

Hybrid Statistical and Mechanistic Mathematical Model Guides Mobile Health Intervention for Chronic Pain.

Nearly a quarter of visits to the emergency department are for conditions that could have been managed via outpatient treatment; improvements that allow patients to quickly recognize and receive appropriate treatment are crucial. The growing popularity of mobile technology creates new opportunities for real-time adaptive medical intervention, and the simultaneous growth of "big data" sources allows for preparation of personalized recommendations. Here we focus on the reduction of chronic suffering in the sickle cell disease (SCD) community. SCD is a chronic blood disorder in which pain is the most frequent complication. There currently is no standard algorithm or analytical method for real-time adaptive treatment recommendations for pain. Furthermore, current state-of-the-art methods have difficulty in handling continuous-time decision optimization using big data. Facing these challenges, in this study, we aim to develop new mathematical tools for incorporating mobile technology into personalized treatment plans for pain. We present a new hybrid model for the dynamics of subjective pain that consists of a dynamical systems approach using differential equations to predict future pain levels, as well as a statistical approach tying system parameters to patient data (both personal characteristics and medication response history). Pilot testing of our approach suggests that it has significant potential to well predict pain dynamics, given patients reported pain levels and medication usages. With more abundant data, our hybrid approach should allow physicians to make personalized, data-driven recommendations for treating chronic pain.

Error analysis of hybrid photovoltaic power forecasting models: A case study of mediterranean climate

The advancement of photovoltaic (PV) energy into electricity market requires efficient photovoltaic power prediction systems. Furthermore the analysis of PV power forecasting errors is essential for optimal unit commitment and economic dispatch of power systems with significant PV power penetrations. This study is focused on the forecasting of the power output of a photovoltaic system located in Apulia – South East of Italy at different forecasting horizons, using historical output power data and performed by hybrid statistical models based on Least Square Support Vector Machines (LS-SVM) with Wavelet Decomposition (WD). Five forecasting horizons, from 1h up to 24h, were considered. A detailed error analysis, by mean error and statistical distributions was carried out to compare the performance with the traditional Artificial Neural Network (ANN) and LS-SVM without the WD. The decomposition of the RMSE into three contributions (bias, standard deviation bias and dispersion) and the estimation of the skewness and kurtosis statistical metrics provide a better understanding of the differences between prediction and measurement values. The hybrid method based on LS-SVM and WD out-performs other methods in the majority of cases. It is also evaluated the impact of the accuracy of the forecasting method on the imbalance penalties. The most accurate forecasts permit to reduce such penalties and thus maximize revenue.

Wavelet-based hybrid natural image modeling using generalized Gaussian and α-stable distributions

Natural image is characterized by its highly kurtotic and heavy-tailed distribution in wavelet domain. These typical non-Gaussian statistics are commonly described by generalized Gaussian density (GGD) or α-stable distribution. However, each of the two models has its own deficiency to capture the variety and complexity of real world scenes. Considering the statistical properties of GGD and α-stable distributions respectively, in this paper we propose a hybrid statistical model of natural image’s wavelet coefficients which is better in describing the leptokurtosis and heavy tails simultaneously. Based on a clever fusion of GGD and α-stable functions, we establish the optimal parametric hybrid model, and a close-formed Kullback–Leibler divergence of the hybrid model is derived for evaluating model accuracy. Experiment results and comparative studies demonstrate that the proposed hybrid model is closer to the true distribution of natural image’s wavelet coefficients than the single modeling using GGD or α-stable, while is beneficial for applications such as image comparison.

A hybrid statistical downscaling model for prediction of winter precipitation in China

Downscaling techniques can effectively improve the coarse resolution and poor representation of precipitation predicted by general circulation model (GCM). In this study, a statistical downscaling (SD) method, based on the singular value decomposition (SVD), is proposed for better representing the coupled variation between predictors and winter precipitation in China. By comparing current predictors from Climate Forecast System version 2 (CFSv2) of National Centers for Environmental Prediction and previous predictors from observation, the two best appropriate predictors, the winter sea level pressure (SLP) from the CFSv2 and the autumn sea-ice concentration (SIC) from observation, are selected to construct the SD model for prediction of winter precipitation in China. Three downscaling schemes are developed by involving the SLP, SIC, and both of them (i.e. SLP-scheme, SIC-scheme, and SS-scheme), respectively. Validations for the schemes show a considerable improvement of performance in predicting China winter precipitation, compared with the original CFSv2 output. The temporal and spatial anomaly correlation coefficient (ACC) and root mean square errors (RMSE) were estimated. For the cross validation, the spatial ACC are increased from ∼0.01 of the CFSv2 to >0.3 of the downscaling model. For the independent validation, the temporal RMSE from the downscaling schemes are all decreased more than 30%. In particular, the results using the SS-scheme showed relatively smaller RMSE than those of either the SLP-scheme or SIC-scheme, and hence can reproduce the precipitation anomaly in 2011 and 2012 winters more accurately.