Stochastic Decision Tree Research Articles

Integrating stochastic programs and decision trees in capacitated barge planning with uncertain container arrivals

In this paper, we propose an approach that combines optimization techniques with machine learning to improve capacitated barge planning with uncertain container arrivals. The main idea is to use the predictions of a decision tree in the scenario generation of a 2-stage stochastic program to plan the barge calls (i.e. visits) of a barge operator. The predictions of container arrivals help to generate more accurate scenarios, which in turn leads to more informed decisions and less costs. The approach is tested with an iterative method of periodic planning and simulation for a one year duration so that the long term performance is evaluated. A computational experiment is conducted using the historical data of an inland terminal and the Port of Rotterdam. The results show that the proposed approach improves total costs up to 2.07% over the traditional stochastic approach, and up to 4.57% over the current method used in industry.

Trace-elemental and multi-isotopic (Sr-Nd-Pb) discrimination of jade in the circum-Caribbean: Implications for pre-colonial inter-island exchange networks

Dense and strong, hydrothermal-metasomatic jadeitite and jadeite-omphacite rocks were used as tools and adornments throughout the wider Caribbean since initial inhabitation. Regionally, rich sources of jadeitite and jadeite-omphacite jade are known only in Guatemala (north and south of the Motagua Fault Zone), eastern Cuba and the northern Dominican Republic, establishing that humans transported jadeitic material over vast distances. This study validates that geochemical fingerprinting is a viable provenance method for Caribbean pre-colonial jadeitic lithologies. An assemblage of 101 source rocks has been characterised for trace element and combined Sr-Nd-Pb isotope compositions. Four statistical approaches (Principal Component Analysis, t-Distributed Stochastic Neighbour Embedding, Decision Tree, and Multiclass Regression) were assessed, employing source-distinct trace element ratios. A multiclass regression technique based on trace element ratios of immobile high field strength, light to medium rare earth and fluid-mobile, large-ion-lithophile elements is shown to be most effective in discriminating the four source regions.Ninety-one % of the Guatemalan samples can be discriminated from the Dominican and Cuban sources using La/Th, Zr/Hf and Y/Th ratios. Jadeitic rocks cropping out in the Dominican Republic can be distinguished from Cuban jades employing Er/Yb, Nb/Ta and Ba/Rb ratios with 71% certainty. Furthermore, the two Guatemala sources, north and south of the Motagua Fault Zone, can be discriminated by using (among others) Zr/Hf, Ta/Th, La/Sm and Dy/Y ratios with an 89% success rate. This raises the possibility of determining, in detail, former trading and mobility networks between different islands and the Meso- and Central American mainland within the Greater Caribbean.The provenance technique was applied to 19 pre-colonial jade celts excavated from the Late Ceramic Age Playa Grande archaeological site in the northern Dominican Republic. Three artefacts are discriminated as derived from the Guatemalan source, indicating that, despite a source of jade within 25 km, material was traded from Guatemala. The presence of Guatemalan jade in the Playa Grande lithic assemblage provides further evidence of large scale (>3000 km), regional trading and indigenous knowledge transfer networks.

An in-depth exploration of Bangla blog post classification

Bangla blog is increasing rapidly in the era of information, and consequently, the blog has a diverse layout and categorization. In such an aptitude, automated blog post classification is a comparatively more efficient solution in order to organize Bangla blog posts in a standard way so that users can easily find their required articles of interest. In this research, nine supervised learning models which are Support Vector Machine (SVM), multinomial naïve Bayes (MNB), multi-layer perceptron (MLP), k-nearest neighbours (k-NN), stochastic gradient descent (SGD), decision tree, perceptron, ridge classifier and random forest are utilized and compared for classification of Bangla blog post. Moreover, the performance on predicting blog posts against eight categories, three feature extraction techniques are applied, namely unigram TF-IDF (term frequency-inverse document frequency), bigram TF-IDF, and trigram TF-IDF. The majority of the classifiers show above 80% accuracy. Other performance evaluation metrics also show good results while comparing the selected classifiers.

Usage of Particle Swarm Optimization to Improve the Performance of Supervised Classifiers

Representing the data appropriately will have a significant effect on the outcome produced by the classifier. Transforming the feature will help to represent the data points in a more suitable way for the classifier. Particle swarm optimization belongs to the Swarm Optimization techniques category and it is generally used for solving numerical optimization problems, weight updation in neural networks, and feature selection. This research work proposes a Particle swarm optimization-based transformation technique for increasing the classification metrics of popular classification algorithms namely K-Nearest Neighbor, Stochastic Gradient Descent Classifier, Decision trees, and linear discriminant analysis classifier. Experiments are conducted using the SONAR dataset and the highest accuracy of 82.69% is attained for Stochastic Gradient Descent classifier when Particle swarm optimization is used as a transformation technique

Modeling the Synchronization of Multimodal Perceptions as a Basis for the Emergence of Deterministic Behaviors.

Living organisms have either innate or acquired mechanisms for reacting to percepts with an appropriate behavior e.g., by escaping from the source of a perception detected as threat, or conversely by approaching a target perceived as potential food. In the case of artifacts, such capabilities must be built in through either wired connections or software. The problem addressed here is to define a neural basis for such behaviors to be possibly learned by bio-inspired artifacts. Toward this end, a thought experiment involving an autonomous vehicle is first simulated as a random search. The stochastic decision tree that drives this behavior is then transformed into a plastic neuronal circuit. This leads the vehicle to adopt a deterministic behavior by learning and applying a causality rule just as a conscious human driver would do. From there, a principle of using synchronized multimodal perceptions in association with the Hebb principle of wiring together neuronal cells is induced. This overall framework is implemented as a virtual machine i.e., a concept widely used in software engineering. It is argued that such an interface situated at a meso-scale level between abstracted micro-circuits representing synaptic plasticity, on one hand, and that of the emergence of behaviors, on the other, allows for a strict delineation of successive levels of complexity. More specifically, isolating levels allows for simulating yet unknown processes of cognition independently of their underlying neurological grounding.

NETWORK TRAFFIC ANOMALIES DETECTION USING AN ENSEMBLE OF CLASSIFIERS

Network technologies have been steadily developing and their application has been expanding. One of the aspects of the development is a modification of the current network attacks and the appearance of new ones. The anomalies that can be detected in network traffic conform with such attacks. Development of new and improvement of the current approaches to detect anomalies in network traffic have become an urgent task. The article suggests a hybrid approach to detect anomalies on the basis of the combined signature approach and computationally effective classifiers of machine learning: logistic regression, stochastic gradient descent and decision tree with accuracy increase due to weighted voting. The choice of the classifiers is explained by the admissible complexity of the algorithms that allows detection of network traffic events for the time close to real. Signature analysis is carried out with the help of the Zeek IDS (Intrusion Detection System) signature base. Learning is fulfilled by preliminary prepared (by excluding extra recordings and parameters) CICIDS2017 (Canadian Institute for Cybersecurity Intrusion Detection System) signature set by cross validation. The set is roughly divided into ten parts that allows us to increase the accuracy. Experimental evaluation of the developed approach comparing with individual classifiers and with other approaches by such criteria as part of type I and II errors, accuracy and level of detection, has proved the approach suitable to be applied in network attacks detection systems. It is possible to introduce the developed approach into both existing and new anomaly detection systems.

Reliability of Supervised Machine Learning Using Synthetic Data in Health Care: Model to Preserve Privacy for Data Sharing.

BackgroundThe exploitation of synthetic data in health care is at an early stage. Synthetic data could unlock the potential within health care datasets that are too sensitive for release. Several synthetic data generators have been developed to date; however, studies evaluating their efficacy and generalizability are scarce.ObjectiveThis work sets out to understand the difference in performance of supervised machine learning models trained on synthetic data compared with those trained on real data.MethodsA total of 19 open health datasets were selected for experimental work. Synthetic data were generated using three synthetic data generators that apply classification and regression trees, parametric, and Bayesian network approaches. Real and synthetic data were used (separately) to train five supervised machine learning models: stochastic gradient descent, decision tree, k-nearest neighbors, random forest, and support vector machine. Models were tested only on real data to determine whether a model developed by training on synthetic data can used to accurately classify new, real examples. The impact of statistical disclosure control on model performance was also assessed.ResultsA total of 92% of models trained on synthetic data have lower accuracy than those trained on real data. Tree-based models trained on synthetic data have deviations in accuracy from models trained on real data of 0.177 (18%) to 0.193 (19%), while other models have lower deviations of 0.058 (6%) to 0.072 (7%). The winning classifier when trained and tested on real data versus models trained on synthetic data and tested on real data is the same in 26% (5/19) of cases for classification and regression tree and parametric synthetic data and in 21% (4/19) of cases for Bayesian network-generated synthetic data. Tree-based models perform best with real data and are the winning classifier in 95% (18/19) of cases. This is not the case for models trained on synthetic data. When tree-based models are not considered, the winning classifier for real and synthetic data is matched in 74% (14/19), 53% (10/19), and 68% (13/19) of cases for classification and regression tree, parametric, and Bayesian network synthetic data, respectively. Statistical disclosure control methods did not have a notable impact on data utility.ConclusionsThe results of this study are promising with small decreases in accuracy observed in models trained with synthetic data compared with models trained with real data, where both are tested on real data. Such deviations are expected and manageable. Tree-based classifiers have some sensitivity to synthetic data, and the underlying cause requires further investigation. This study highlights the potential of synthetic data and the need for further evaluation of their robustness. Synthetic data must ensure individual privacy and data utility are preserved in order to instill confidence in health care departments when using such data to inform policy decision-making.

Reservoir and Economic-Uncertainties Assessment for Recovery-Strategy Selection Using Stochastic Decision Trees

Summary The field–development phase before investment approval is characterized by relatively large uncertainties at the time important decisions have to be made. For instance, it is crucial to select an appropriate recovery strategy (depletion or injection) to obtain optimal hydrocarbon cumulative production while ensuring good profitability of the project. Evaluation of the reservoir along with economic uncertainties and quantification of their impact are needed before the field–development concept selection. In this paper we describe how to stochastically assess reservoir and economic uncertainties and the screening process used to select the best recovery strategy. Our chosen methodology is a combination of uncertainty studies, including continuous, discrete, and controllable parameters. The different screened scenarios are combined in a stochastic decision tree, built up through decision and chance nodes, to establish a distribution of recoverable volumes and rank the recovery strategies given a chosen criterion. A second uncertainty study is performed by adding economic uncertainties to the initial set of reservoir uncertain parameters. Eventually, a new decision tree is established, and scenarios are ranked using economic criteria. We present an application of this methodology to an oil field from the Norwegian continental shelf and how recovery strategies are ranked in this paper. The described methodology has exhibited the risks and uncertainties carried by the project, as it was possible to rank the different solutions on the basis of the dispersion of the recoverable volumes distribution and/or on the net present value (NPV). In the context of a marginal– or large–capital–expenditure (CAPEX) project, a robust P90 case is required, and this might, therefore, influence the choice of the recovery strategy. For instance, a scenario yielding the largest hydrocarbon volume might not be selected because it requires too many wells and/or too large an investment if one of these criteria is defined as the most important. In addition, the combination of uncertainty studies enabled a full economic evaluation covering the entire recoverable–volumes distribution whereas in many projects, economic evaluation is focused on the P90, mean, and P10 scenarios. The two–step integrated approach allows a decision to be made while taking into account both reservoir and economic aspects. A combined stochastic approach to the reservoir and economic uncertainties avoids a biased decision. All cases are stochastically covered and screened using a systematic and unified methodology that gives the same weight to each scenario.

Cost-effectiveness analysis of oral rehydration therapy compared to intravenous rehydration for acute gastroenteritis without severe dehydration treatment

BackgroundDiarrhea causes, annually, approximately 1.7 billion cases and 760,000 deaths worldwide among children under 5 years of age, although these are preventable and treatable. This study aim to assess the cost-effectiveness for the treatment of diarrhea in emergency services in the management of children of acute gastroenteritis with non-severe dehydration. MethodsA stochastic decision tree model considering the perspective of the Brazilian public health system was used to calculate the cost-effectiveness of the 5 interventions: oral rehydration therapy (ORT) at home, and if it fails supervised ORT; they would receive; ORT at home, and if it fails intravenous rehydration therapy (IVT). ORT at home and if it fails, the half of them will receive supervised ORT, and the other half would receive IVT; Patient receives supervised oral treatment; Patient receives IVT. Quality-adjusted life year (QALY) was used to measure the clinical outcomes. ResultsThe strategy of initiating oral rehydration in children younger than 5 is the most efficient practice with a cost of $14.28 and effectiveness of 0.89 QALYs. ConclusionORT is an underutilized resource for the management of children with non-severe dehydration in emergency services. The overprescribed IVT increases cost without a corresponding significant increase in effectiveness.

Economic aspects of decision making in production processes with uncertain component quality

Production and service processes always introduce discrepancies between real and expected parameters. Accurate modelling of uncertainties in manufacturing processes has great impact on logistics related factors. In this representation, instead of a deterministic analysis of manufacturing processes with uncertain quality of components, decision making is described with stochastic decision trees. The objective of this work is to show how to include the uncertain quality of components into the optimization of manufacturing processes. This paper proposes a new decision tree model with stochastic parameters that can determine the optimal operations depending on the quality of components to be assembled. After a careful literature review, this paper introduces a model to formulate the problem of decision making in manufacturing and assembly processes depending on the quality of produced batches of components. The model seeks the optimal pre-assembly operation, like testing or repairing while taking into account specific costs, prior and posterior probabilities and likelihoods. Next, we demonstrate an enhanced stochastic decision tree with cumulative hypergeometric probability distributions to find the minimum of expected cost based objective function. Numerical results demonstrate how the proposed model supports the decision making process and increase the efficiency, flexibility and reliability of the manufacturing process. DOI: DOI: http://dx.doi.org/10.5755/j01.ee.30.1.19324

Impact of Machine Learning With Multiparametric Magnetic Resonance Imaging of the Breast for Early Prediction of Response to Neoadjuvant Chemotherapy and Survival Outcomes in Breast Cancer Patients.

The aim of this study was to assess the potential of machine learning with multiparametric magnetic resonance imaging (mpMRI) for the early prediction of pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) and of survival outcomes in breast cancer patients. This institutional review board-approved prospective study included 38 women (median age, 46.5 years; range, 25-70 years) with breast cancer who were scheduled for NAC and underwent mpMRI of the breast at 3 T with dynamic contrast-enhanced (DCE), diffusion-weighted imaging (DWI), and T2-weighted imaging before and after 2 cycles of NAC. For each lesion, 23 features were extracted: qualitative T2-weighted and DCE-MRI features according to BI-RADS (Breast Imaging Reporting and Data System), quantitative pharmacokinetic DCE features (mean plasma flow, volume distribution, mean transit time), and DWI apparent diffusion coefficient (ADC) values. To apply machine learning to mpMRI, 8 classifiers including linear support vector machine, linear discriminant analysis, logistic regression, random forests, stochastic gradient descent, decision tree, adaptive boosting, and extreme gradient boosting (XGBoost) were used to rank the features. Histopathologic residual cancer burden (RCB) class (with RCB 0 being a pCR), recurrence-free survival (RFS), and disease-specific survival (DSS) were used as the standards of reference. Classification accuracy with area under the receiving operating characteristic curve (AUC) was assessed using all the extracted qualitative and quantitative features for pCR as defined by RCB class, RFS, and DSS using recursive feature elimination. To overcome overfitting, 4-fold cross-validation was used. Machine learning with mpMRI achieved stable performance as shown by mean classification accuracies for the prediction of RCB class (AUC, 0.86) and DSS (AUC, 0.92) based on XGBoost and the prediction of RFS (AUC, 0.83) with logistic regression. The XGBoost classifier achieved the most stable performance with high accuracies compared with other classifiers. The most relevant features for the prediction of RCB class were as follows: changes in lesion size, complete pattern of shrinkage, and mean transit time on DCE-MRI; minimum ADC on DWI; and peritumoral edema on T2-weighted imaging. The most relevant features for prediction of RFS were as follows: volume distribution, mean plasma flow, and mean transit time; DCE-MRI lesion size; minimum, maximum, and mean ADC with DWI. The most relevant features for prediction of DSS were as follows: lesion size, volume distribution, and mean plasma flow on DCE-MRI, and maximum ADC with DWI. Machine learning with mpMRI of the breast enables early prediction of pCR to NAC as well as survival outcomes in breast cancer patients with high accuracy and thus may provide valuable predictive information to guide treatment decisions.

Opinion fraud detection via neural autoencoder decision forest

Online reviews play an important role in influencing buyers’ daily purchase decisions. However, fake and meaningless reviews, which cannot reflect users’ genuine purchase experience and opinions, widely exist on the Web and pose great challenges for users to make right choices. Therefore, it is desirable to build a fair model that evaluates the quality of products by distinguishing spamming reviews. We present an end-to-end trainable unified model to leverage the appealing properties from Autoencoder and random forest. A stochastic decision tree model is implemented to guide the global parameter learning process. Extensive experiments were conducted on a large Amazon review dataset. The proposed model consistently outperforms a series of compared methods.

Healthcare Text Classification System and its Performance Evaluation: A Source of Better Intelligence by Characterizing Healthcare Text.

A machine learning (ML)-based text classification system has several classifiers. The performance evaluation (PE) of the ML system is typically driven by the training data size and the partition protocols used. Such systems lead to low accuracy because the text classification systems lack the ability to model the input text data in terms of noise characteristics. This research study proposes a concept of misrepresentation ratio (MRR) on input healthcare text data and models the PE criteria for validating the hypothesis. Further, such a novel system provides a platform to amalgamate several attributes of the ML system such as: data size, classifier type, partitioning protocol and percentage MRR. Our comprehensive data analysis consisted of five types of text data sets (TwitterA, WebKB4, Disease, Reuters (R8), and SMS); five kinds of classifiers (support vector machine with linear kernel (SVM-L), MLP-based neural network, AdaBoost, stochastic gradient descent and decision tree); and five types of training protocols (K2, K4, K5, K10 and JK). Using the decreasing order of MRR, our ML system demonstrates the mean classification accuracies as: 70.13 ± 0.15%, 87.34 ± 0.06%, 93.73 ± 0.03%, 94.45 ± 0.03% and 97.83 ± 0.01%, respectively, using all the classifiers and protocols. The corresponding AUC is 0.98 for SMS data using Multi-Layer Perceptron (MLP) based neural network. All the classifiers, the best accuracy of 91.84 ± 0.04% is shown to be of MLP-based neural network and this is 6% better over previously published. Further we observed that as MRR decreases, the system robustness increases and validated by standard deviations. The overall text system accuracy using all data types, classifiers, protocols is 89%, thereby showing the entire ML system to be novel, robust and unique. The system is also tested for stability and reliability.

The economic value of information provided by milk biomarkers under different scenarios: Case-study of an ex-ante analysis of fat-to-protein ratio and fatty acid profile to detect subacute ruminal acidosis in dairy cows

Monitoring systems (MS) provide additional information that many developers and researchers expect will reduce the uncertainty surrounding decision-making in livestock production and therefore enhance management decisions. However, the actual economic value of the information (VoI) yielded by MS has hardly been investigated. The aim of this study was to fill that void based on two objectives. The first is to estimate the VoI of MS prior to implementation using decision analysis based on scarce data from different sources. The second objective is to identify which factors most influence the VoI of MS and to develop recommendations about the focus of future MS development. To illustrate our objectives, we used a case study of two milk biomarkers used to monitor subacute ruminal acidosis (SARA) in dairy cows: fat-to-protein ratio (FPR) and the fatty acid profile (FAP). FPR is presently used to monitor SARA, while FAP is a newly developed test, currently in the pre-commercial phase, with reports of better accuracy than FPR. A stochastic decision tree model was used to estimate the expected monetary value of three levels of information with regards to SARA: (i) no monitoring, monitoring (ii) with FPR or (iii) with FAP. The VoI of FPR and FAP were calculated as the difference in expected monetary value of monitoring with FPR and FAP as compared with no monitoring, respectively. Several scenarios were modeled using sensitivity and elasticity analyses. The aim was not only to compensate for the scarcity of data for some variables, but also to identify under which conditions decisions based on FAP monitoring were indeed the best. In all the scenarios, monitoring SARA with FPR had the lowest expected monetary value. No monitoring was a better decision in 70% of the iterations in the scenario that described the most probable situation. The VoI of FAP was positive when SARA prevalence was between 0.21 and 0.79 with its maximum value at 0.61, when the treatment costs were lower than €116/case/year and when the disease costs were higher than €260/case/year. Moreover, an increase of specificity of the FAP to 0.95 yielded a positive VoI, whereas an increase of its sensitivity to 1.0 still yielded a negative VoI, suggesting that developers of the FAP should focus on improving its specificity rather than its sensitivity. To avoid suboptimal use of finite resources while developing MS, we recommend ex-ante investigation of the VoI of the MS under development.

Decision Analytic Modelling Methods For The Assessing The Effectiveness Of Treatment Sequences For Clinical And Economic Decision Making: A Methodological Review

Appropriate recognition of treatment sequencing is crucial to many policy decisions and related economic evaluations. The decision problem can become complex when accounting for extensive treatment sequences and pertinent factors relating to the disease. A methodological review was conducted to identify the breadth of decision analytic modelling approaches used for evaluating treatment sequences. A comprehensive literature review of MEDLINE, Embase and Cochrane library. Studies were categorised according to modelling approach and decision problem. Treatment sequencing assumptions were analysed. 70 studies and 48 discrete models were included. A wide range of modelling techniques were identified: cohort-based models (deterministic and stochastic decision trees, Markov, semi-Markov, partitioned survival); individual sampling models (state transition, and discrete event simulation (DES)); and open population-based models (DES and Markov cohort). No study systematically tested different modelling approaches for treatment sequences. Cohort models can be simple and easy to implement. Examples of cohort models adapted successfully to accommodate additional complexity were identified, but these were no longer simple. Individual sampling models are more sophisticated, better able to accommodate greater decision problem complexity, and provide more flexibility. DES appeared the optimum approach in terms of intuitively modelling sequencing algorithms, computational efficacy, and ease of updating, but requires more extensive modelling skills, specialist software, and is data and time intensive. In the absence of sequencing trials, modellers often applied simplifying assumptions to treatment effects obtained from trials of single treatments. These assumptions were frequently not validated, nor their impact assessed: an important limitation of these models. Modelling treatment sequences may require a complex model, which can be time consuming to develop and implement. Using simplistic assumptions, regarding sequencing effects, results in significant uncertainty around the effectiveness and cost-effectiveness estimates, the extent of which is generally unknown. This needs to be recognised in decision making, and further evaluated.

FastBDT: A Speed-Optimized Multivariate Classification Algorithm for the Belle II Experiment

Stochastic gradient-boosted decision trees are widely employed for multivariate classification and regression tasks. This paper presents a speed-optimized and cache-friendly implementation for multivariate classification called FastBDT. The concepts used to optimize the execution time are discussed in detail in this paper. The key ideas include: an equal-frequency binning on the input data, which allows replacing expensive floating-point with integer operations, while at the same time increasing the quality of the classification; a cache-friendly linear access pattern to the input data, in contrast to usual implementations, which exhibit a random access pattern. FastBDT provides interfaces to C/C++, Python and TMVA. It is extensively used in the field of high energy physics (HEP) by the Belle II experiment.

Economic Burden of Inpatient Admission of Ankle Fractures

Category: Trauma Introduction/Purpose: Ankle fractures are one of the most prevalent traumatic orthopaedic injuries. A large proportion of patients sustaining operative ankle fractures are admitted directly from the emergency room prior to surgical management. In our experience, however, many closed ankle injuries may be safely and effectively managed on an outpatient basis. This study aims to characterize the economic impact of routine inpatient admission of ankle fractures. Methods: A retrospective review of all closed ankle fracture surgery performed by a single foot and ankle fellowship trained surgeon at a tertiary level academic center in 2012 was conducted to identify patients requiring postoperative inpatient admission. The National Inpatient Sample was queried for operative lateral malleolus, bimalleolar, and trimalleolar ankle fractures in 2012 to evaluate national estimates and length of stay by age. The maximum allowable Medicare inpatient facility reimbursements and Medicare outpatient facility reimbursements were obtained from the Medicare Acute Inpatient Prospective Pricer and a Medicare Outpatient Pricer Code respectively. Private facility reimbursement rates were estimated at 139% of inpatient Medicare reimbursement and 280% of outpatient reimbursement as described in the literature. Surgeon and anesthesiologist fees were considered similar between both inpatient and outpatient groups. A unique stochastic decision tree model was derived from probabilities and associated costs and evaluated with modified Monte Carlo simulation. Results: Of 76 lateral malleolar, bimalleolar, and trimalleolar ankle fracture ORIF performed in 2012 by the senior author, 7 required admission due to polytrauma, medical comorbiditites or age. Of the 67 outpatient cases, all were discharged home the day of surgery. In the 2012 national cohort analyzed, 48,044 estimated inpatient admissions occurred postoperatively for closed ankle fractures. Median length of stay was three days, and associated with facility reimbursement ranging from $12,920 for Medicare lateral malleolus fractures to $18,613 for private reimbursement of trimalleolar fractures. Outpatient facility reimbursements per case were $4,125 for Medicare patients and $11,459 for private insurance patients. The national cohort’s inpatient admission accounted for $796,033,050 in reimbursements, while outpatient surgery would have been associated with $419,327,612 for treatment of these ankle fractures. Conclusion: Closed lateral malleolus, bimalleolar, and trimalleolar fractures may be safely and effectively treated as outpatient procedures. Routine perioperative admission for ankle fractures results in over $367 million of excess facility reimbursements annually in the US. Even if a conservative 25% necessary admission rate was assumed, routine inpatient admission of ankle fractures results in a $282 million dollar excess economic burden annually. While certain cases necessitate inpatient admission, with value based decision making becoming increasingly the responsibility of the orthopaedic surgeon, understanding cost implications of inpatient ankle fracture management may result in savings to the US healthcare system and patients individually.

Optimizing Negotiation Conflict in the Cloud Service Negotiation Framework Using Probabilistic Decision Making Model.

Optimization of negotiation conflict in the cloud service negotiation framework is identified as one of the major challenging issues. This negotiation conflict occurs during the bilateral negotiation process between the participants due to the misperception, aggressive behavior, and uncertain preferences and goals about their opponents. Existing research work focuses on the prerequest context of negotiation conflict optimization by grouping similar negotiation pairs using distance, binary, context-dependent, and fuzzy similarity approaches. For some extent, these approaches can maximize the success rate and minimize the communication overhead among the participants. To further optimize the success rate and communication overhead, the proposed research work introduces a novel probabilistic decision making model for optimizing the negotiation conflict in the long-term negotiation context. This decision model formulates the problem of managing different types of negotiation conflict that occurs during negotiation process as a multistage Markov decision problem. At each stage of negotiation process, the proposed decision model generates the heuristic decision based on the past negotiation state information without causing any break-off among the participants. In addition, this heuristic decision using the stochastic decision tree scenario can maximize the revenue among the participants available in the cloud service negotiation framework.

On the relationship between the prices of oil and the precious metals: Revisiting with a multivariate regime-switching decision tree

This study examines the volatility and correlation and their relationships among the euro/US dollar exchange rates, the S&P500 equity indices, and the prices of WTI crude oil and the precious metals (gold, silver, and platinum) over the period 2005 to 2012. Our model links the univariate volatilities with the correlations via a hidden stochastic decision tree. The ensuing Hidden Markov Decision Tree (HMDT) model is in fact an extension of the Hidden Markov Model (HMM) introduced by Jordan et al. (1997). The architecture of this model is the opposite that of the classical deterministic approach based on a binary decision tree and, it allows a probabilistic vision of the relationship between univariate volatility and correlation. Our results are categorized into three groups, namely (1) exchange rates and oil, (2) S&P500 indices, and (3) precious metals. A switching dynamics is seen to characterize the volatilities, while, in the case of the correlations, the series switch from one regime to another, this movement touching a peak during the period of the Subprime crisis in the US, and again during the days following the Tohoku earthquake in Japan. Our findings show that the relationships between volatility and correlation are dependent upon the nature of the series considered, sometimes corresponding to those found in econometric studies, according to which correlation increases in bear markets, at other times differing from them.

William W. Cooper: Innovator, Fighter, and Scholar

W illiam Wager Cooper was born on July 23, 1914 in Birmingham, Alabama, where his father was a bookkeeper and later a distributor for Anheuser-Busch. Bill’s family moved to Chicago when he was three years old. His father owned a chain of gasoline stations, but lost them during the Great Depression. With his father in ill health and no family revenue, Bill left high school after his sophomore year to work at whatever jobs he could find. These included spotting pins in bowling alleys, caddying at golf courses, and even professional boxing, where his record was 58 wins, 3 losses, and 2 draws. While hitchhiking to the golf course one day, Bill met Eric L. Kohler, a principal with Arthur Andersen & Co., who became his life-long mentor. Kohler loaned him money to enroll as a non-degree student at The University of Chicago. Bill liked the academic atmosphere and took the college entrance examinations, intending to become a physical chemist. During his studies, Kohler asked him to review the mathematics used in a patent infringement suit. After Cooper found errors in the mathematics used by the plaintiff’s engineers, Andersen hired him part time during the school year and full time in the summer. This awakened Cooper’s interest in accounting, and he switched his major to economics. While at Chicago, Bill became good friends with a political science student, Herbert A. Simon. Cooper graduated Phi Beta Kappa with his A.B. in 1938.