Manual Decision Research Articles

Money laundering transaction detection with classification tree models

The detection of money laundering is a very important problem, especially in the financial sector. We propose a mathematical specification of the problem in terms of a classification tree model that ”automates” expert based manual decisions. We operationally validate the model on a concrete application that originates from a large Italian bank. The application of the model to the data shows a good predictive accuracy and, even more importantly, the reduction of false positives, with respect to the ”manual” expert based activity. From an interpretational viewpoint, while some drivers of suspicious laundering activity are in line with the daily business practices of the bank’s anti money laundering operations, some others are new discoveries.

Abstract 9330: Machine Learning Can Enable the Automation of Clinical Endpoint Adjudication

Introduction: Clinical Endpoint Adjudication is a critical component of Cardiovascular (CV) clinical outcome trials and relies on manual decision making to adjudicate clinical events, including CV death. This process is resource intensive and subject to human-driven variance. As part of a broader initiative to automate data collection and analysis of clinical endpoints, we evaluated whether machine learning (ML) algorithms could replicate the outcomes of expert adjudication for CV death. Algorithms were trained on data from THEMIS (NCT01991795), a large, randomized trial comparing ticagrelor to placebo in patients with Type 2 Diabetes Mellitus. Methods: We deployed a deep learning, biomedical named entity (NE) extraction model called BERN to extract relevant NE’s from clinical text of THEMIS events. NE’s were transformed into a sparse, numerical matrix and concatenated with event-level structured features to form a data set of 962 events. Expert adjudicator consensus for CV death was used as ground truth. We trained models using grid search and cross validation, experimenting with XGBoost, Random Forest, Logistic Regression, and Naïve Bayes. Performance was assessed on a 25% (240 of 962) validation subset of the data excluded from training. Metrics used to evaluate performance were Precision, Recall, Accuracy and Area Under the Receiver Operating Characteristic Curve (AUC). Results: Best performance was observed on models trained using Naïve Bayes (>97% ROC AUC on validation data), see Table. Top ranked features for classifying CV death included site investigator decision, sex, and NE’s associated with diagnoses or symptoms, such as “edema” and “chest pain.” Conclusion: With high consistency between automated and expert adjudication, our models demonstrate that machine learning may augment or even replace clinician adjudication in CV outcome trials. Subsequent research will focus on pursuing ML approaches to adjudicate other outcome events.

Asphalt pavement maintenance plans intelligent decision model based on reinforcement learning algorithm

Making a proper maintenance plan for the pavement health is the key to maintain a good service level and bearing capacity. To cope with the increasing demand for pavement maintenance, based on the idea of reinforcement learning, an intelligent decision-making model for pavement maintenance plans based on proximal policy optimization algorithm was proposed in this paper. The decision model fully considers the comprehensive maintenance benefit-cost ratio during the whole life cycle of the road. To overcome the problems of the experience-led manual decision-making, it conducted the decision-making between pavement conditions and maintenance plans based on data mining technique. Besides, a method for constructing a reinforcement learning Environment module based on a deep artificial neural network was proposed, and a reward function is designed for road maintenance decisions. The model was applied to the highway maintenance decision in Jiangsu Province, and verified that the decision overall accuracy of the reinforcement learning model was 82.2%, which was an increase of 17.2% compared with the artificial neural network model.

Research on Fish Slicing Method Based on Simulated Annealing Algorithm

Multiobjective optimization is a common problem in the field of industrial cutting. In actual production settings, it is necessary to rely on the experience of skilled workers to achieve multiobjective collaborative optimization. The process of industrial intelligence is to perceive the parameters of a cut object through sensors and use machines instead of manual decision making. However, the traditional sequential algorithm cannot satisfy multiobjective optimization problems. This paper studies the multiobjective optimization problem of irregular objects in the field of aquatic product processing and uses the information guidance strategy to develop a simulated annealing algorithm to solve the problem according to the characteristics of the object itself. By optimizing the mutation strategy, the ability of the simulated annealing algorithm to jump out of the local optimal solution is improved. The project team developed an experimental prototype to verify the algorithm. The experimental results show that compared with the traditional sequential algorithm method, the simulated degradation algorithm designed in this paper effectively improves the quality of the target solution and greatly enhances the economic value of the product by addressing the multiobjective optimization problem of squid cutting. At the end of the article, the cutting error is analyzed.

Governance aspects of large-scale implementation of automated surveillance of healthcare-associated infections

ObjectivesSurveillance of healthcare-associated infections (HAI) is increasingly automated by applying algorithms to routine-care data stored in electronic health records. Hitherto, initiatives have mainly been confined to single healthcare facilities and research settings, leading to heterogeneity in design. The PRAISE network – Providing a Roadmap for Automated Infection Surveillance in Europe – designed a roadmap to provide guidance on how to move automated surveillance (AS) from the research setting to large-scale implementation. Supplementary to this roadmap, we here discuss the governance aspects of automated HAI surveillance within networks, aiming to support both the coordinating centres and participating healthcare facilities as they set up governance structures and to enhance involvement of legal specialists. MethodsThis article is based on PRAISE network discussions during two workshops. A taskforce was installed that further elaborated governance aspects for AS networks by reviewing documents and websites, consulting experts and organizing teleconferences. Finally, the article has been reviewed by an independent panel of international experts. ResultsStrict governance is indispensable in surveillance networks, especially when manual decisions are replaced by algorithms and electronically stored routine-care data are reused for the purpose of surveillance. For endorsement of AS networks, governance aspects specifically related to AS networks need to be addressed. Key considerations include enabling participation and inclusion, trust in the collection, use and quality of data (including data protection), accountability and transparency. ConclusionsThis article on governance aspects can be used by coordinating centres and healthcare facilities participating in an AS network as a starting point to set up governance structures. Involvement of main stakeholders and legal specialists early in the development of an AS network is important for endorsement, inclusivity and compliance with the laws and regulations that apply.

Elastic deployment of container clusters across geographically distributed cloud data centers for web applications

AbstractContainers such as Docker provide a lightweight virtualization technology. They have gained popularity in developing, deploying and managing applications in and across Cloud platforms. Container management and orchestration platforms such as Kubernetes run application containers in virtual clusters that abstract the overheads in managing the underlying infrastructures to simplify the deployment of container solutions. These platforms are well suited for modern web applications that can give rise to geographic fluctuations in use based on the location of users. Such fluctuations often require dynamic global deployment solutions. A key issue is to decide how to adapt the number and placement of clusters to maintain performance, whilst incurring minimum operating and adaptation costs. Manual decisions are naive and can give rise to: over‐provisioning and hence cost issues; improper placement and performance issues, and/or unnecessary relocations resulting in adaptation issues. Elastic deployment solutions are essential to support automated and intelligent adaptation of container clusters in geographically distributed Clouds. In this article, we propose an approach that continuously makes elastic deployment plans aimed at optimizing cost and performance, even during adaptation processes, to meet service level objectives (SLOs) at lower costs. Meta‐heuristics are used for cluster placement and adjustment. We conduct experiments on the Australia‐wide National eResearch Collaboration Tools and Resources Research Cloud using Docker and Kubernetes. Results show that with only a 0.5 ms sacrifice in SLO for the 95th percentile of response times we are able to achieve up to 44.44% improvement (reduction) in cost compared to a naive over‐provisioning deployment approach.

Deep Learning-Based Differentiation between Mucinous Cystic Neoplasm and Serous Cystic Neoplasm in the Pancreas Using Endoscopic Ultrasonography.

Mucinous cystic neoplasms (MCN) and serous cystic neoplasms (SCN) account for a large portion of solitary pancreatic cystic neoplasms (PCN). In this study we implemented a convolutional neural network (CNN) model using ResNet50 to differentiate between MCN and SCN. The training data were collected retrospectively from 59 MCN and 49 SCN patients from two different hospitals. Data augmentation was used to enhance the size and quality of training datasets. Fine-tuning training approaches were utilized by adopting the pre-trained model from transfer learning while training selected layers. Testing of the network was conducted by varying the endoscopic ultrasonography (EUS) image sizes and positions to evaluate the network performance for differentiation. The proposed network model achieved up to 82.75% accuracy and a 0.88 (95% CI: 0.817–0.930) area under curve (AUC) score. The performance of the implemented deep learning networks in decision-making using only EUS images is comparable to that of traditional manual decision-making using EUS images along with supporting clinical information. Gradient-weighted class activation mapping (Grad-CAM) confirmed that the network model learned the features from the cyst region accurately. This study proves the feasibility of diagnosing MCN and SCN using a deep learning network model. Further improvement using more datasets is needed.

High-frequency forecasting for grocery point-of-sales: intervention in practice and theoretical implications for operational design

Food waste in grocery supply chains may exceed one third of the total volume, depending on the category. To address this problem effectively, grocery retailers are introducing automated systems for more efficient store replenishment and dynamic pricing. The stock keeping unit (SKU) and store level forecast is pivotal in these operations management solutions, but operationally challenging. Large grocery retailers have millions of SKU-store combinations that depending on the operational application would need to be forecasted on a weekly, daily, hourly, or even 15-min frequency. However, in grocery it is challenging to account for demand variation at high frequencies without introducing manual decisions into the process of forecast model configuration. To investigate the limits of current practice and explore opportunities of technology-enabled change, we explore how an advanced forecasting method for electricity demand, called TBATS, can automate daily forecasting for grocery store replenishment. Adopting an interventionist approach, we explore the implications for the design of the operational process in the operational setting provided by the case company. We find that TBATS can produce high frequency base forecasts for the SKU-store level accurately for a period exceeding 3 months. This finding points to the opportunity of shifting operational focus from recalculating forecasts to monitoring forecast errors. Introducing variable, even indefinite re-training frequencies for forecasting models is a significant change of the forecasting process for situations where monitoring requires less computation than retraining, potentially reducing the time and cost associated with increasing the forecast frequency.

The framing of decision making support systems on increasing community resilience in disaster risk reduction efforts: a conceptual approach

This research begins by comprehensively exploring previous research related to community resilience and what steps are used to increase community resilience in reducing disaster risk. Conceptually, it is known that the fatigue model accumulated by the time system, infrastructure system, governance system, regulatory system, and hazard system for disaster risk reduction is often associated with weakening community resilience. It is often associated with catastrophic events, which are sometimes predictable and unpredictable. In manual decision-making, people are aware of the inconsistency of subjective decisions. A decision support system hypothesizes that it will take less time to explore data to make faster and more informed decisions. As a result of this concept, it is possible to reduce the number of wrong choices when dealing with disaster risk reduction issues. In terms of disaster risk reduction, the power of decision support systems is discussed in this paper to find a framework for its effectiveness as relative decision making will differ on different dimensions of Resilience.

StimulEye: An intelligent tool for feature extraction and event detection from raw eye gaze data

Extraction of eye gaze events is highly dependent on automated powerful software that charges exorbitant prices. The proposed open-source intelligent tool StimulEye helps to detect and classify eye gaze events and analyse various metrics related to these events. The algorithms for eye event detection in use today heavily depend on hand-crafted signal features and thresholding, which are computed from the stream of raw gaze data. These algorithms leave most of their parametric decisions on the end user which might result in ambiguity and inaccuracy. StimulEye uses deep learning techniques to automate eye gaze event detection which neither requires manual decision making nor parametric definitions. StimulEye provides an end to end solution which takes raw streams of data from an eye tracker in text form, analyses these to classify the inputs into the events, namely saccades, fixations, and blinks. It provides the user with insights such as scanpath, fixation duration, radii, etc.

Response modality-dependent categorical choice representations for vibrotactile comparisons

Previous electrophysiological studies in monkeys and humans suggest that premotor regions are the primary loci for the encoding of perceptual choices during vibrotactile comparisons. However, these studies employed paradigms wherein choices were inextricably linked with the stimulus order and selection of manual movements. It remains largely unknown how vibrotactile choices are represented when they are decoupled from these sensorimotor components of the task. To address this question, we used fMRI-MVPA and a variant of the vibrotactile frequency discrimination task which enabled the isolation of choice-related signals from those related to stimulus order and selection of the manual decision reports. We identified the left contralateral dorsal premotor cortex (PMd) and intraparietal sulcus (IPS) as carrying information about vibrotactile choices. Our finding provides empirical evidence for an involvement of the PMd and IPS in vibrotactile decisions that goes above and beyond the coding of stimulus order and specific action selection. Considering findings from recent studies in animals, we speculate that the premotor region likely serves as a temporary storage site for information necessary for the specification of concrete manual movements, while the IPS might be more directly involved in the computation of choice. Moreover, this finding replicates results from our previous work using an oculomotor variant of the task, with the important difference that the informative premotor cluster identified in the previous work was centered in the bilateral frontal eye fields rather than in the PMd. Evidence from these two studies indicates that categorical choices in human vibrotactile comparisons are represented in a response modality-dependent manner.

An Efficient and Securable Networked Voting System based on Blockchain

Modern period of trend setting innovation where online framework supports work speed, lessens botches and advances the age of exact outcomes, having manual political decision framework turns into a mishap. An open political decision framework establishes the foundation of a vote based system where the individuals need to choose their state's chief. India as of now utilizes a manual political decision framework, which causes a few sorts of issues. Because of this paper polling form based political race framework, a few issues are looked by voters previously or during races and others are looked by the organization when the democratic. An online framework, which includes strategies like enlistment of voters, vote throwing, vote tallying, and announcing results would comprise a decent answer for supplant current framework and the proposed framework right now be useful for the voters by utilizing any assets like their own framework or masterminded by Government. Additionally, the proposed framework will likewise diminish the hazard for defilement. The framework is proposed subsequent to talking authorities of two offices, the Country Database and Enrollment Authority India (NADRA) and the Political decision Commission of India (ECP). NADRA has an online database of the residents of India, and is giving the Modernized National Personality Cards (CNIC) and furthermore supporting various associations with their online framework. Along these lines, by utilizing NADRA's framework it turns out to be anything but difficult to enroll all voters of the age above 18, and besides to check and verify their information

Recognization and Systematization of MR Imagesusing K Means Clustering and DNN

Brain tumors are the result of unusual growth and unrestrained cell disunity in the brain. Most of the medical image application lack in segmentation and labeling. Brain tumors can lead to loss of lives if they are not detected early and correctly. Recently, deep learning has been an important role in the field of digital health. One of its action is the reduction of manual decision in the diagnosis of diseases specifically brain tumor diagnosis needs high accuracy, where minute errors in judgment may lead to loss therefore, brain tumor segmentation is an necessary challenge in medical side. In recent time numerous ,methods exist for tumor segmentation with lack of accuracy. Deep learning is used to achieve the goal of brain tumor segmentation. In this work, three network of brain MR images segmentation is employed .A single network is compared to achieve segmentation of MR images using separate network .In this paper segmentation has improved and result is obtained with high accuracy and efficiency.

A Generic Approach to Covariance Function Estimation Using ARMA-Models

Covariance function modeling is an essential part of stochastic methodology. Many processes in geodetic applications have rather complex, often oscillating covariance functions, where it is difficult to find corresponding analytical functions for modeling. This paper aims to give the methodological foundations for an advanced covariance modeling and elaborates a set of generic base functions which can be used for flexible covariance modeling. In particular, we provide a straightforward procedure and guidelines for a generic approach to the fitting of oscillating covariance functions to an empirical sequence of covariances. The underlying methodology is developed based on the well known properties of autoregressive processes in time series. The surprising simplicity of the proposed covariance model is that it corresponds to a finite sum of covariance functions of second-order Gauss–Markov (SOGM) processes. Furthermore, the great benefit is that the method is automated to a great extent and directly results in the appropriate model. A manual decision for a set of components is not required. Notably, the numerical method can be easily extended to ARMA-processes, which results in the same linear system of equations. Although the underlying mathematical methodology is extensively complex, the results can be obtained from a simple and straightforward numerical method.

Optimization-Based Scheduling for the Process Industries: From Theory to Real-Life Industrial Applications

Scheduling is a major component for the efficient operation of the process industries. Especially in the current competitive globalized market, scheduling is of vital importance to most industries, since profit margins are miniscule. Prof. Sargent was one of the first to acknowledge this. His breakthrough contributions paved the way to other researchers to develop optimization-based methods that can address a plethora of process scheduling problems. Despite the plethora of works published by the scientific community, the practical implementation of optimization-based scheduling in industrial real-life applications is limited. In most industries, the optimization of production scheduling is seen as an extremely complex task and most schedulers prefer the use of a simulation-based software or manual decision, which result to suboptimal solutions. This work presents a comprehensive review of the theoretical concepts that emerged in the last 30 years. Moreover, an overview of the contributions that address real-life industrial case studies of process scheduling is illustrated. Finally, the major reasons that impede the application of optimization-based scheduling are critically analyzed and possible remedies are discussed.

RETRACTED ARTICLE: A weighted graph-oriented ontology matching algorithm for enhancing ontology mapping and alignment in Semantic Web

Ontologies support the intelligent systems in which information needs to be handled, reused and transferred. This paper is about a weighted graph-oriented ontology matching algorithm, namely the WeGO++ algorithm that helps ontology matching in a pragmatic manner. It considers predominantly the graph-structured nature of ontology models. The methods and approaches in obtaining matching across ontologies use the basic principles of similarity identifications among graph structures. WeGO++ ontology matching frame work is able to successfully match all the data sets. The algorithmic process that is contained in the work looks for nine types of Similarity Scores between members of the ontologies to be matched. Each score is given a specific weight for arriving at the final Similarity Score for each element, and the overall Similarity Score for the ontology is computed using these values. The pragmatic nature of this matching solution is an advantage which approaches the task in a simplistic and efficient manner. The computations are not very complicated and do not use large computing power. It is possible to add more features such as incorporating a sophisticated GUI, changing the matching threshold at runtime and adding a new module to introduce a graphical representation of Similarity Scores of the ontology elements. Threshold values can be fixed to accept levels of matching between elements and classify the quality of match that exists between the ontologies concerned. This provides some room for the manual decision making which can be efficient in deciding matching levels in certain hesitant matches. The algorithms have been tested with OAEI test data sets and some popular ontology library data sets apart from ontologies built from conventional database data sets.

Automated Clustering for Data Analytics

ABSTRACTToday, auditors must consider the risks of material misstatement due to fraud during the financial statement audit (Messier, Glover, and Prawitt 2016). Current audit guidance recommends the use of data mining methods such as clustering to improve the likelihood of discovering irregularities during fraud risk assessment (ASB 2012). Unfortunately, significant challenges exist relative to using clustering in practice, including data preprocessing, model construction, model selection, and outlier detection. The traditional auditor is not trained to effectively address these complexities. One solution entails automation of clustering, thus eliminating the difficult, manual decision points within the clustering process. This would allow practitioners to focus on problem investigation and resolution, rather than being burdened with the technical aspects of clustering. In this paper, automated clustering is explored. In the process, each manual decision point is addressed, and a suitable automated solution is developed. Upon conclusion, a clustering application is formulated and demonstrated.

Behavioral classification of low-frequency acceleration and temperature data from a free-ranging small mammal.

The miniaturization and affordability of new technology is driving a biologging revolution in wildlife ecology with use of animal‐borne data logging devices. Among many new biologging technologies, accelerometers are emerging as key tools for continuously recording animal behavior. Yet a critical, but under‐acknowledged consideration in biologging is the trade‐off between sampling rate and sampling duration, created by battery‐ (or memory‐) related sampling constraints. This is especially acute among small animals, causing most researchers to sample at high rates for very limited durations. Here, we show that high accuracy in behavioral classification is achievable when pairing low‐frequency acceleration recordings with temperature. We conducted 84 hr of direct behavioral observations on 67 free‐ranging red squirrels (200–300 g) that were fitted with accelerometers (2 g) recording tri‐axial acceleration and temperature at 1 Hz. We then used a random forest algorithm and a manually created decision tree, with variable sampling window lengths, to associate observed behavior with logger recorded acceleration and temperature. Finally, we assessed the accuracy of these different classifications using an additional 60 hr of behavioral observations, not used in the initial classification. The accuracy of the manually created decision tree classification using observational data varied from 70.6% to 91.6% depending on the complexity of the tree, with increasing accuracy as complexity decreased. Short duration behavior like running had lower accuracy than long‐duration behavior like feeding. The random forest algorithm offered similarly high overall accuracy, but the manual decision tree afforded the flexibility to create a hierarchical tree, and to adjust sampling window length for behavioral states with varying durations. Low frequency biologging of acceleration and temperature allows accurate behavioral classification of small animals over multi‐month sampling durations. Nevertheless, low sampling rates impose several important limitations, especially related to assessing the classification accuracy of short duration behavior.

Orchard Free Space and Center Line Estimation Using Naive Bayesian Classifier for Unmanned Ground Self-Driving Vehicle

In the case of autonomous orchard navigation, researchers have developed algorithms that utilize features, such as trunks, canopies, and sky in orchards, but there are still various difficulties in recognizing free space for autonomous navigation in a changing agricultural environment. In this study, we applied the Naive Bayesian classification to detect the boundary between the trunk and the ground and propose an algorithm to determine the center line of free space. The naïve Bayesian classification requires a small number of samples for training and a simple training process. In addition, it was able to effectively classify tree trunk’s points and noise points of the orchard, which are problematic in vision-based processing, and noise caused by small branches, soil, weeds, and tree shadows on the ground. The performance of the proposed algorithm was investigated using 229 sample images obtained from an image acquisition system with a Complementary Metal Oxide Semiconductor (CMOS) Image Sensor (CIS) camera. The center line detected by the unaided-eye manual decision and the results extracted by the proposed algorithm were compared and analyzed for several parameters. In all compared parameters, extracted center line was more stable than the manual center line results.

Computational Testing for Automated Preprocessing 2: Practical Demonstration of a System for Scientific Data-Processing Workflow Management for High-Volume EEG.

Existing tools for the preprocessing of EEG data provide a large choice of methods to suitably prepare and analyse a given dataset. Yet it remains a challenge for the average user to integrate methods for batch processing of the increasingly large datasets of modern research, and compare methods to choose an optimal approach across the many possible parameter configurations. Additionally, many tools still require a high degree of manual decision making for, e.g., the classification of artifacts in channels, epochs or segments. This introduces extra subjectivity, is slow, and is not reproducible. Batching and well-designed automation can help to regularize EEG preprocessing, and thus reduce human effort, subjectivity, and consequent error. The Computational Testing for Automated Preprocessing (CTAP) toolbox facilitates: (i) batch processing that is easy for experts and novices alike; (ii) testing and comparison of preprocessing methods. Here we demonstrate the application of CTAP to high-resolution EEG data in three modes of use. First, a linear processing pipeline with mostly default parameters illustrates ease-of-use for naive users. Second, a branching pipeline illustrates CTAP's support for comparison of competing methods. Third, a pipeline with built-in parameter-sweeping illustrates CTAP's capability to support data-driven method parameterization. CTAP extends the existing functions and data structure from the well-known EEGLAB toolbox, based on Matlab, and produces extensive quality control outputs. CTAP is available under MIT open-source licence from https://github.com/bwrc/ctap.