Search results for: hidden markov models (HMM)

Authors: Elisabeth Maidl

Abstract:

The interdependence of climate change and natural hazard goes along with large uncertainties regarding future risks. Regional stakeholders, experts in natural hazards management and scientists have specific knowledge, resp. mental models on such risks. This diversity of views makes it difficult to find common and broadly accepted prevention measures. If the specific knowledge of these types of actors is shared in an interactive knowledge production process, this enables a broader and common understanding of complex risks and allows to agree on long-term solution strategies. Previous studies on mental models confirm that actors with specific vulnerabilities perceive different aspects of a topic and accordingly prefer different measures. In bringing these perspectives together, there is the potential to reduce uncertainty and to close blind spots in solution finding. However, studies that examine the mental models of regional actors on future concrete mass movement risks are lacking so far. The project tests and evaluates the feasibility of knowledge co-creation for the anticipatory prevention of climate change-induced mass movement risks in the Alps. As a key element, mental models of the three included groups of actors are compared. Being integrated into the research program Climate Change Impacts on Alpine Mass Movements (CCAMM2), this project is carried out in two Swiss mountain regions. The project is structured in four phases: 1) the preparatory phase, in which the participants are identified, 2) the baseline phase, in which qualitative interviews and a quantitative pre-survey are conducted with actors 3) the knowledge-co-creation phase, in which actors have a moderated exchange meeting, and a participatory modelling workshop on specific risks in the region, and 4) finally a public information event. Results show that participants' mental models are based on the place of origin, profession, believes, values, which results in narratives on climate change and hazard risks. Further, the more intensively participants interact with each other, the more likely is that they change their views. This provides empirical evidence on how changes in opinions and mindsets can be induced and fostered.

Keywords: climate change, knowledge-co-creation, participatory process, natural hazard risks

Procedia PDF Downloads 50

Authors: Sibylle Tellenbach, Julie Haddock-Millar, Francis Bidault

Abstract:

The aim of this research is to understand the extent to which hybrid working models have influenced employee engagement in the Swiss financial sector. The context for this research is the transition out of the pandemic and the changes that have occurred between 2020 and 2023. Since the pandemic, many financial services companies have had to rethink their working model for office-based employees, as this group of employees has been able to experience a new way of working and, thus, greater freedom and flexibility. For a large number of companies, it was a huge change to shift from the traditional office-based to a new hybrid working model. A heightened focus on employee engagement has become a necessity in order to understand and respond to the challenges presented by the shift in a working model. This new way of working, partly office-based and partly virtual, has led to ambiguities about the impact on the engagement of hybrid teams. Therefore, the research question is: How hybrid working models have influenced employee engagement to what extent? The methodological approach is a narrative inquiry with four similar functional teams within four Swiss financial companies. Semi-structured interviews will be conducted with managers from middle management and their individual team members. The findings will demonstrate whether this shift in the working model influenced individual team members’ engagement and to what extent. The contribution of this research is two-fold. First, the research makes a theoretical contribution, presenting evidence of the impact of hybrid working on individual team members’ engagement in a specific sector and context, enhancing current knowledge on the challenges in working model transition. Second, this research will make a practice-based contribution, recommending ways to enhance the engagement of hybrid teams in a specific context. These recommendations may be applied in wider sectors and teams.

Keywords: employee engagement, hybrid teams, hybrid working models, Swiss financial sector, team engagement

Procedia PDF Downloads 76

Authors: Farhood Rismanchian, Seong Hyeon Park, Young Hoon Lee

Abstract:

During the recent years, demand for healthcare services has dramatically increased. As the demand for healthcare services increases, so does the necessity of constructing new healthcare buildings and redesigning and renovating existing ones. Increasing demands necessitate the use of optimization techniques to improve the overall service efficiency in healthcare settings. However, high complexity of care processes remains the major challenge to accomplish this goal. This study proposes a method based on process mining results to address the high complexity of care processes and to find the optimal layout of the various medical centers in an emergency department. ProM framework is used to discover clinical pathway patterns and relationship between activities. Sequence clustering plug-in is used to remove infrequent events and to derive the process model in the form of Markov chain. The process mining results served as an input for the next phase which consists of the development of the optimization model. Comparison of the current ED design with the one obtained from the proposed method indicated that a carefully designed layout can significantly decrease the distances that patients must travel.

Keywords: Mixed Integer programming, Facility layout problem, Process Mining, Healthcare Operation Management

Procedia PDF Downloads 325

Authors: Ismail Aslantas

Abstract:

Value-Added Models (VAMs), the statistical methods for evaluating the effectiveness of teachers and schools based on student achievement growth, has attracted decision-makers’ and researchers’ attention over the last decades. As a result of this attention, many studies have conducted in recent years to discuss these statistical models from different aspects. This research focused on the importance of conceptual variables in VAM estimations; therefor, this research was undertaken to examine the extent to which value-added effectiveness estimates for teachers can be affected by using context predictions. Using longitudinal data over three years from the international school context, value-added teacher effectiveness was estimated by ordinary least-square value-added models, and the effectiveness of the teachers was examined. The longitudinal dataset in this study consisted of three major sources: students’ attainment scores up to three years and their characteristics, teacher background information, and school characteristics. A total of 1,027 teachers and their 35,355 students who were in eighth grade were examined for understanding the impact of model specifications on the value-added teacher effectiveness evaluation. Models were created using selection methods that adding a predictor on each step, then removing it and adding another one on a subsequent step and evaluating changes in model fit was checked by reviewing changes in R² values. Cohen’s effect size statistics were also employed in order to find out the degree of the relationship between teacher characteristics and their effectiveness. Overall, the results indicated that prior attainment score is the most powerful predictor of the current attainment score. 47.1 percent of the variation in grade 8 math score can be explained by the prior attainment score in grade 7. The research findings raise issues to be considered in VAM implementations for teacher evaluations and make suggestions to researchers and practitioners.

Keywords: model specification, teacher effectiveness, teacher performance evaluation, value-added model

Procedia PDF Downloads 111

Authors: Alisha Khanal, Gokhan Saygili

Abstract:

It is commonly observed that aftershocks follow the mainshock. Aftershocks continue over a period of time with a decreasing frequency and typically there is not sufficient time for repair and retrofit between a mainshock–aftershock sequence. Usually, aftershocks are smaller in magnitude; however, aftershock ground motion characteristics such as the intensity and duration can be greater than the mainshock due to the changes in the earthquake mechanism and location with respect to the site. The seismic performance of slopes is typically evaluated based on the sliding displacement predicted to occur along a critical sliding surface. Various empirical models are available that predict sliding displacement as a function of seismic loading parameters, ground motion parameters, and site parameters but these models do not include the aftershocks. The seismic risks associated with the post-mainshock slopes ('damaged slopes') subjected to aftershocks is significant. This paper extends the empirical sliding displacement models for flexible slopes subjected to earthquake mainshock-aftershock sequences (a multi hazard approach). A dataset was developed using 144 pairs of as-recorded mainshock-aftershock sequences using the Pacific Earthquake Engineering Research Center (PEER) database. The results reveal that the combination of mainshock and aftershock increases the seismic demand on slopes relative to the mainshock alone; thus, seismic risks are underestimated if aftershocks are neglected.

Keywords: seismic slope stability, mainshock, aftershock, landslide, earthquake, flexible slopes

Procedia PDF Downloads 129

Authors: Yuliya Radanova, Agnė Tvaronavičienė

Abstract:

With the growing trend for mandating parties of family conflicts to out-of-court processes, the adopted statutory regulations often remain silent on the way the voice of the child is integrated into the procedure. Convention on the Rights of the Child (Art. 12) clearly states the obligation to assure to the child who can form his or her own views the right to express those views freely in all matters affecting him. This article seeks to explore the way children participate in the mandatory mediation schemes applicable to family disputes in the European Union. A review of scientific literature and empirical data has been conducted on those EU Member States that coerce parties to family mediation to establish that different models of practice are deployed, and there is a lack of synchronicity on how children’s role in mediation is viewed. Child-inclusive mediation processes are deemed to produce sustainable results over time but necessitate professional qualifications and skills for the purpose of mediators to accommodate that such discussions are aligned with the best interest of the child. However, there is no unanimous guidance, standards or protocols on the peculiar characteristics and manner through which children are involved in mediation. Herewith, it is suggested that the lack of such rigorous approaches and coherence in an ever-changing mediation setting transitioning towards mandatory mediation models jeopardizes the importance of children’s voices in the process. Thus, it is suggested that there is a need to consider the adoption of uniform guidelines on the specific role children have in mediation, particularly in its mandatory models.

Keywords: family mediation, child involvement, mandatory mediation, child-inclusive, child-focused

Procedia PDF Downloads 55

Authors: Zainab Mutlaq Ibrahim, Mohamed Bader-El-Den, Mihaela Cocea

Abstract:

Students' textual feedback can hold unique patterns and useful information about learning process, it can hold information about advantages and disadvantages of teaching methods, assessment components, facilities, and other aspects of teaching. The results of analysing such a feedback can form a key point for institutions’ decision makers to advance and update their systems accordingly. This paper proposes a data mining framework for analysing end of unit general textual feedback using part of speech feature (PoS) with four machine learning algorithms: support vector machines, decision tree, random forest, and naive bays. The proposed framework has two tasks: first, to use the above algorithms to build an optimal model that automatically classifies the whole data set into two subsets, one subset is tailored to assessment practices (assessment related), and the other one is the non-assessment related data. Second task to use the same algorithms to build an optimal model for whole data set, and the new data subsets to automatically detect their sentiment. The significance of this paper is to compare the performance of the above four algorithms using part of speech feature to the performance of the same algorithms using n-grams feature. The paper follows Knowledge Discovery and Data Mining (KDDM) framework to construct the classification and sentiment analysis models, which is understanding the assessment domain, cleaning and pre-processing the data set, selecting and running the data mining algorithm, interpreting mined patterns, and consolidating the discovered knowledge. The results of this paper experiments show that both models which used both features performed very well regarding first task. But regarding the second task, models that used part of speech feature has underperformed in comparison with models that used unigrams and bigrams.

Keywords: assessment, part of speech, sentiment analysis, student feedback

Procedia PDF Downloads 122

Authors: Rafal Bryk, Holger Schmidt, Thomas Mull, Ingo Ganzmann, Oliver Herbst

Abstract:

Prediction of the two-phase water mixture level during fast depressurization of the Reactor Pressure Vessel (RPV) resulting from an accident scenario is an important issue from the view point of the reactor safety. Since the level swell may influence the behavior of some passive safety systems, it has been recognized that an assumption which at the beginning may be considered as a conservative one, not necessary leads to a conservative result. This paper discusses outcomes obtained during simulations of the water dynamics and heat transfer during sudden depressurization of a vessel filled up to a certain level with liquid water under saturation conditions and with the rest of the vessel occupied by saturated steam. In case of the pressure decrease e.g. due to the main steam line break, the liquid water evaporates abruptly, being a reason thereby, of strong transients in the vessel. These transients and the sudden emergence of void in the region occupied at the beginning by liquid, cause elevation of the two-phase mixture. In this work, several models calculating the water collapse and swell levels are presented and validated against experimental data. Each of the models uses different approach to calculate void fraction. The object-oriented models were developed with the Modelica modelling language and the OpenModelica environment. The models represent the RPV of the Integral Test Facility Karlstein (INKA) – a dedicated test rig for simulation of KERENA – a new Boiling Water Reactor design of Framatome. The models are based on dynamic mass and energy equations. They are divided into several dynamic volumes in each of which, the fluid may be single-phase liquid, steam or a two-phase mixture. The heat transfer between the wall of the vessel and the fluid is taken into account. Additional heat flow rate may be applied to the first volume of the vessel in order to simulate the decay heat of the reactor core in a similar manner as it is simulated at INKA. The comparison of the simulations results against the reference data shows a good agreement.

Keywords: boiling water reactor, level swell, Modelica, RPV depressurization, thermal-hydraulics

Procedia PDF Downloads 190

Authors: Aishwarya Ravindra Fursule, Shruti Kshirsagar

Abstract:

In the intersection of artificial intelligence and human-centered computing, this paper delves into speech emotion recognition (SER). It presents a comparative analysis of machine learning models such as K-Nearest Neighbors (KNN),logistic regression, support vector machines (SVM), decision trees, ensemble classifiers, and random forests, applied to SER. The research employs four datasets: Crema D, SAVEE, TESS, and RAVDESS. It focuses on extracting salient audio signal features like Zero Crossing Rate (ZCR), Chroma_stft, Mel Frequency Cepstral Coefficients (MFCC), root mean square (RMS) value, and MelSpectogram. These features are used to train and evaluate the models’ ability to recognize eight types of emotions from speech: happy, sad, neutral, angry, calm, disgust, fear, and surprise. Among the models, the Random Forest algorithm demonstrated superior performance, achieving approximately 79% accuracy. This suggests its suitability for SER within the parameters of this study. The research contributes to SER by showcasing the effectiveness of various machine learning algorithms and feature extraction techniques. The findings hold promise for the development of more precise emotion recognition systems in the future. This abstract provides a succinct overview of the paper’s content, methods, and results.

Keywords: comparison, ML classifiers, KNN, decision tree, SVM, random forest, logistic regression, ensemble classifiers

Procedia PDF Downloads 26

Authors: Aparna Soni

Abstract:

The much-celebrated foretold story of Indian city engines driving the growth of India has been scrutinized to have serious consequences. A wide spectrum of scholarship has brought to light the un-equalizing effects and the need to adopt a rights-based approach to development planning in India. Notably, these concepts and discourses ubiquitously entail the study of land struggles in the making of Urban. In fact, the very progression of the primitive accumulation theory to accumulation by dispossession, followed by ‘dispossession without development,’ thereafter Development without dispossession and now as Dispossession by financialization noticeably the last three developing in a span of mere three decades, is evidence enough to trace the centrality and evolving role of land in the making of urban India. India, in the last decade, has seen its regional governments actively experimenting with alternative models of land assembly (Amaravati and Delhi land pooling models, the loudly advertised ones). These are publicized as a replacement to the presumably cost and time antagonistic, prone to litigation land acquisition act of 2013. It has been observed that most of the literature treats these models as a generic large bracket of land expropriation and do not, in particular, try to differentially analyse to granularly find a pattern in these alternatives. To cater to this gap, this research comparatively studies these alternative land, assembly models. It categorises them based on their basic architecture, spatial and sectoral application, and governance frameworks. It is found that these alternatives are ad-hoc and fragmented pieces of legislation. These are fit for profit models commodifying land to ease its access by the private sector for real estate led growth. The research augments the literature on the privatization of land use planning in India. Further, it attempts to discuss the increasing role a landowner is expected to play in the future and suggests a way forward to safeguard them from market risks. The study involves a thematic analysis of the policy elements contained in legislative/policy documents, notifications, office orders. The study also derives from the various widely circulated print media information. With the present field-visit limitations, the study relies on documents accessed open-source in the public domain.

Keywords: commodification, dispossession, land acquisition, landowner

Procedia PDF Downloads 145

Authors: Jacob Hatherell, Jason Matthews, Arnaud Marmier

Abstract:

Anvil Staking is a cold-forming process that is used in the assembly of plain spherical bearings into a rod-end housing. This process ensures that the bearing outer lip conforms to the chamfer in the matching rod end to produce a lightweight mechanical joint with sufficient strength to meet the pushout load requirement of the assembly. Finite Element (FE) analysis is being used extensively to predict the behaviour of metal flow in cold forming processes to support industrial manufacturing and product development. On-going research aims to validate FE models across a wide range of bearing and rod-end geometries by systematically isolating and understanding the uncertainties caused by variations in, material properties, load-dependent friction coefficients and strain rate sensitivity. The improved confidence in these models aims to eliminate the costly and time-consuming process of experimental trials in the introduction of new bearing designs. Previous literature has shown that friction coefficients do not remain constant during cold forming operations, however, the understanding of this phenomenon varies significantly and is rarely implemented in FE models. In this paper, a new approach to evaluate the normal contact pressure versus friction coefficient relationship is outlined using friction calibration charts generated via iterative FE models and ring compression tests. When compared to previous research, this new approach greatly improves the prediction of forming geometry and the forming load during the staking operation. This paper also aims to standardise the FE approach to modelling ring compression test and determining the friction calibration charts.

Keywords: anvil staking, finite element analysis, friction coefficient, spherical plain bearing, ring compression tests

Procedia PDF Downloads 191

Authors: Keyi Wang

Abstract:

Similar to the touchscreen, hand gesture based human-computer interaction (HCI) is a technology that could allow people to perform a variety of tasks faster and more conveniently. This paper proposes a training method of an image-based hand gesture image and video clip recognition system using a CNN (Convolutional Neural Network) with a dataset. A dataset containing 6 hand gesture images is used to train a 2D CNN model. ~98% accuracy is achieved. Furthermore, a 3D CNN model is trained on a dataset containing 4 hand gesture video clips resulting in ~83% accuracy. It is demonstrated that a Cozmo robot loaded with pre-trained models is able to recognize static and dynamic hand gestures.

Keywords: deep learning, hand gesture recognition, computer vision, image processing

Procedia PDF Downloads 117

Authors: Vanessa Matias Leite, Jandira Guenka Palma, Flávio Henrique de Oliveira

Abstract:

This paper presents a model-oriented development approach to software development in the Model-View-Controller (MVC) architectural standard. This approach aims to expose a process of extractions of information from the models, in which through rules and syntax defined in this work, assists in the design of the initial model and its future conversions. The proposed paper presents a syntax based on the natural language, according to the rules agreed in the classic grammar of the Portuguese language, added to the rules of conversions generating models that follow the norms of the Object Management Group (OMG) and the Meta-Object Facility MOF.

Keywords: BNF Syntax, model driven architecture, model-view-controller, transformation, UML

Procedia PDF Downloads 376

Authors: Ryspek Usubamatov, Tan San Chin, Sarken Kapaeva

Abstract:

Productivity of the manufacturing systems depends on technological processes, a technical data of machines and a structure of systems. Technology is presented by the machining mode and data, a technical data presents reliability parameters and auxiliary time for discrete production processes. The term structure of manufacturing systems includes the number of serial and parallel production machines and links between them. Structures of manufacturing systems depend on the complexity of technological processes. Mathematical models of productivity rate for manufacturing systems are important attributes that enable to define best structure by criterion of a productivity rate. These models are important tool in evaluation of the economical efficiency for production systems.

Keywords: productivity, structure, manufacturing systems, structural design

Procedia PDF Downloads 566

Authors: Onur Karaman, Ceren Karaman

Abstract:

In this study, the adsorption of Methylene Blue (MB), a cationic dye, onto Orange Peel Derived Activated Carbon (OPAC) and chitosan(OPAC/Chitosan composite) composite (a low-cost absorbent) was carried out using a batch system. The composite was characterised using IR spectra, XRD, FESEM and Pore size studies. The effects of initial pH, adsorbent dose rate and initial dye concentration on the initial adsorption rate, capacity and dye removal efficiency were investigated. The Langmuir and Freundlich adsorption models were used to define the adsorption equilibrium of dye-adsorbent system mathematically and it was decided that the Langmuir model was more suitable to describe the adsorption equilibrium for the system. In addition, first order, second order and saturation type kinetic models were applied to kinetic data of adsorption and kinetic constants were calculated. It was concluded that the second order and the saturation type kinetic models defined the adsorption data more accurately. Finally, the evaluated thermodynamic parameters of adsorption show a spontaneous and exothermic behavior. Overall, this study indicates OPAC/Chitosan composite as an effective and low-cost adsorbent for the removal of MB dye from aqueous solutions.

Keywords: activated carbon, adsorption, chitosan, methylene blue, orange peel

Procedia PDF Downloads 270

Authors: Harold Vásquez, Cesar Hernández, Ingrid Páez

Abstract:

This paper approach cognitive radio technic and applied pure proactive handoff Model to decrease interference between PU and SU and comparing it with reactive handoff model. Through the study and analysis of multivariate models SAW and TOPSIS join to 3 dynamic prediction techniques AR, MA ,and ARMA. To evaluate the best model is taken four metrics: number failed handoff, number handoff, number predictions, and number interference. The result presented the advantages using this type of pure proactive models to predict changes in the PU according to the selected channel and reduce interference. The model showed better performance was TOPSIS-MA, although TOPSIS-AR had a higher predictive ability this was not reflected in the interference reduction.

Keywords: cognitive radio, spectrum handoff, decision making, time series, wireless networks

Procedia PDF Downloads 466

Authors: A. Marolleau, F. Salaun, D. Dupont, H. Gidik, S. Ducept.

Abstract:

The goal of this study is to analyze the hydric behaviour of textiles which can impact significantly the comfort of the wearer. Indeed, fabrics can be adapted for different climate if hydric and thermal behaviors are known. In this study, fabrics are only submitted to hydric variations. Sorption and desorption isotherms obtained from the dynamic vapour sorption apparatus (DVS) are fitted with the parallel exponential kinetics (PEK), the Hailwood-Horrobin (HH) and the Brunauer-Emmett-Teller (BET) models. One of the major finding is the relationship existing between PEK and HH models. During slow and fast processes, the sorption of water molecules on the polymer can be in monolayer and multilayer form. According to the BET model, moisture regain, a physical property of textiles, show a linear correlation with the total amount of water taken in monolayer. This study provides potential information of the end uses of these fabrics according to the selected activity level.

Keywords: comfort, hydric properties, modelling, underwears

Procedia PDF Downloads 131

Authors: Mengxiao Li, Johnson Zhang

Abstract:

Currently, topology optimization technique is widely used to define the layout design of structures that are presented as truss-like topologies. However, due to the difficulty in combining optimization technique with more realistic material models where their nonlinear properties should be considered, the achieved optimized topologies are commonly unable to apply straight towards the practical design problems. This study presented an optimization procedure of composite structures where different elastic stiffness, yield criteria, and hardening models are assumed for the candidate materials. From the results, it can be concluded that a more explicit modeling has the significant influence on the resulting topologies. Also, the isotropic or kinematic hardening is important for elastoplastic structural optimization design. The capability of the proposed optimization procedure is shown through several cases.

Keywords: topology optimization, material composition, nonlinear modeling, hardening rules

Procedia PDF Downloads 466

Authors: Janet Modu, Jean-Francois Georgin, Laurent Briancon, Eric Antoinet

Abstract:

The current practice during the repowering phase of wind turbines is deconstruction of existing foundations and construction of new foundations to accept larger wind loads or once the foundations have reached the end of their service lives. The ongoing research project FUI25 FEDRE (Fondations d’Eoliennes Durables et REpowering) therefore serves to propose scalable wind turbine foundation designs to allow reuse of the existing foundations. To undertake this research, numerical models and laboratory-scale models are currently being utilized and implemented in the GEOMAS laboratory at INSA Lyon following instrumentation of a reference wind turbine situated in the Northern part of France. Sensors placed within both the foundation and the underlying soil monitor the evolution of stresses from the foundation’s early age to stresses during service. The results from the instrumentation form the basis of validation for both the laboratory and numerical works conducted throughout the project duration. The study currently focuses on the effect of coupled mechanisms (Thermal-Hydro-Mechanical-Chemical) that induce stress during the early age of the reinforced concrete foundation, and scale factor considerations in the replication of the reference wind turbine foundation at laboratory-scale. Using THMC 3D models on COMSOL Multi-physics software, the numerical analysis performed on both the laboratory-scale and the full-scale foundations simulate the thermal deformation, hydration, shrinkage (desiccation and autogenous) and creep so as to predict the initial damage caused by internal processes during concrete setting and hardening. Results show a prominent effect of early age properties on the damage potential in full-scale wind turbine foundations. However, a prediction of the damage potential at laboratory scale shows significant differences in early age stresses in comparison to the full-scale model depending on the spatial position in the foundation. In addition to the well-known size effect phenomenon, these differences may contribute to inaccuracies encountered when predicting ultimate deformations of the on-site foundation using laboratory scale models.

Keywords: cement hydration, early age behavior, reinforced concrete, shrinkage, THMC 3D models, wind turbines

Procedia PDF Downloads 157

Authors: HyunSu Seo, HoYoung Son, Sungjin Kim, WooYoung Jung

Abstract:

In order to understand the seismic behavior of steel frame structure with infill FRP composite panel, simple models for simulation on the steel frame with the panel systems were developed in this study. To achieve the simple design method of the steel framed structure with the damping panel system, 2-D finite element analysis with the springs and dashpots models was conducted in ABAQUS. Under various applied spring stiffness and dashpot coefficient, the expected hysteretic energy responses of the steel frame with damping panel systems we re investigated. Using the proposed simple design method which decides the stiffness and the damping, it is possible to decide the FRP and damping materials on a steel frame system.

Keywords: numerical analysis, FEM, infill, GFRP, damping

Procedia PDF Downloads 403

Authors: Jan Stodt, Christoph Reich

Abstract:

The usage of machine learning models for prediction is growing rapidly and proof that the intended requirements are met is essential. Audits are a proven method to determine whether requirements or guidelines are met. However, machine learning models have intrinsic characteristics, such as the quality of training data, that make it difficult to demonstrate the required behavior and make audits more challenging. This paper describes an ML audit framework that evaluates and reviews the risks of machine learning applications, the quality of the training data, and the machine learning model. We evaluate and demonstrate the functionality of the proposed framework by auditing an steel plate fault prediction model.

Keywords: audit, machine learning, assessment, metrics

Procedia PDF Downloads 249

Authors: Laidi Maamar, Hanini Salah

Abstract:

The present work investigates the potential of artificial neural network (ANN) model to predict the horizontal global solar radiation (HGSR). The ANN is developed and optimized using three years meteorological database from 2011 to 2013 available at the meteorological station of Blida (Blida 1 university, Algeria, Latitude 36.5°, Longitude 2.81° and 163 m above mean sea level). Optimal configuration of the ANN model has been determined by minimizing the Root Means Square Error (RMSE) and maximizing the correlation coefficient (R2) between observed and predicted data with the ANN model. To select the best ANN architecture, we have conducted several tests by using different combinations of parameters. A two-layer ANN model with six hidden neurons has been found as an optimal topology with (RMSE=4.036 W/m²) and (R²=0.999). A graphical user interface (GUI), was designed based on the best network structure and training algorithm, to enhance the users’ friendliness application of the model.

Keywords: artificial neural network, global solar radiation, solar energy, prediction, Algeria

Procedia PDF Downloads 482

Authors: Isik Cetintav, Cenk Misirli, Yilmaz Can

Abstract:

This work investigates upsetting of the tri-metallic cylindrical billets both experimentally and analytically with a reduction ratio 30%. Steel, brass, and copper are used for the outer and outmost rings and aluminum for the inner core. Two different models have been designed to show material flow and the cavity took place over the two interfaces during forming after this reduction ratio. Each model has an outmost ring material as steel. Model 1 has an outer ring between the outmost ring and the solid core material as copper and Model 2 has a material as brass. Solid core is aluminum for each model. Billets were upset in press machine by using parallel flat dies. Upsetting load was recorded and compared for models and single billets. To extend the tests and compare with experimental procedure to a wider range of inner core and outer ring geometries, finite element model was performed. ABAQUS software was used for the simulations. The aim is to show how contact between outmost ring, outer ring and the inner core are carried on throughout the upsetting process. Results have shown that, with changing in height, between outmost ring, outer ring and inner core, the Model 1 and Model 2 had very good interaction, and the contact surfaces of models had various interface behaviour. It is also observed that tri-metallic materials have lower weight but better mechanical properties than single materials. This can give an idea for using and producing these new materials for different purposes.

Keywords: tri-metallic, upsetting, copper, brass, steel, aluminum

Procedia PDF Downloads 325

Authors: Galal Elkobrosy, Amr M. Abdelrazek, Bassuny M. Elsouhily, Mohamed E. Khidr

Abstract:

Crank shaft length, connecting rod length, crank angle, engine rpm, cylinder bore, mass of piston and compression ratio are the inputs that can control the performance of the slider crank mechanism and then its efficiency. Several combinations of these seven inputs are used and compared. The throughput engine torque predicted by the simulation is analyzed through two different regression models, with and without interaction terms, developed according to multi-linear regression using LU decomposition to solve system of algebraic equations. These models are validated. A regression model in seven inputs including their interaction terms lowered the polynomial degree from 3^rd degree to 1^st degree and suggested valid predictions and stable explanations.

Keywords: design of experiments, regression analysis, SI engine, statistical modeling

Procedia PDF Downloads 167

Authors: Shih-Pin Chen, Shih-Syuan You

Abstract:

Fuzzy regression models are useful for investigating the relationship between explanatory variables and responses in fuzzy environments. To overcome the deficiencies of previous models and increase the explanatory power of fuzzy data, the graded mean integration (GMI) representation is applied to determine representative crisp regression coefficients. A fuzzy regression model is constructed based on the modified dissemblance index (MDI), which can precisely measure the actual total error. Compared with previous studies based on the proposed MDI and distance criterion, the results from commonly used test examples show that the proposed fuzzy linear regression model has higher explanatory power and forecasting accuracy.

Keywords: dissemblance index, fuzzy linear regression, graded mean integration, mathematical programming

Procedia PDF Downloads 420

Authors: Ashneel Ajay Singh, Naoki Suzuki, Kazumi Sakuramoto, Swastika Roshni, Paras Nath, Alok Kalla

Abstract:

Albacore tuna (Thunnus alalunga) is one of the commercially important species of tuna in the North Pacific region. Despite the long history of albacore fisheries in the Pacific, its ecological characteristics are not sufficiently understood. The effects of changing climate on numerous commercially and ecologically important fish species including albacore tuna have been documented over the past decades. The objective of this study was to explore and elucidate the relationship of environmental variables with the stock parameters of albacore tuna. The relationship of the North Pacific albacore tuna recruitment (R), spawning stock biomass (SSB) and recruits per spawning biomass (RPS) from 1970 to 2012 with the environmental factors of sea surface temperature (SST), Pacific decadal oscillation (PDO), El Niño southern oscillation (ENSO) and Pacific warm pool index (PWI) was construed. SST and PDO were used as independent variables with SSB to construct stock reproduction models for R and RPS as they showed most significant relationship with the dependent variables. ENSO and PWI were excluded due to collinearity effects with SST and PDO. Model selections were based on R2 values, Akaike Information Criterion (AIC) and significant parameter estimates at p<0.05. Models with single independent variables of SST, PDO, ENSO and PWI were also constructed to illuminate their individual effect on albacore R and RPS. From the results it can be said that SST and PDO resulted in the most significant models for reproducing North Pacific albacore tuna R and RPS time series. SST has the highest impact on albacore R and RPS when comparing models with single environmental variables. It is important for fishery managers and decision makers to incorporate the findings into their albacore tuna management plans for the North Pacific Oceanic region.

Keywords: Albacore tuna, El Niño southern oscillation, Pacific decadal oscillation, sea surface temperature

Procedia PDF Downloads 214

Authors: Divvela Vishnu Sai Kumar, Deepak Arora, Sheenu Rizvi

Abstract:

Forecasting a cyclone through satellite images consists of the estimation of the intensity of the cyclone and predicting it before a cyclone comes. This research work can help people to take safety measures before the cyclone comes. The prediction of the intensity of a cyclone is very important to save lives and minimize the damage caused by cyclones. These cyclones are very costliest natural disasters that cause a lot of damage globally due to a lot of hazards. Authors have proposed five different CNN (Convolutional Neural Network) models that estimate the intensity of cyclones through INSAT-3D IR images. There are a lot of techniques that are used to estimate the intensity; the best model proposed by authors estimates intensity with a root mean squared error (RMSE) of 10.02 kts.

Keywords: estimating cyclone intensity, deep learning, convolution neural network, prediction models

Procedia PDF Downloads 98

Authors: Cristina da Silva Torres, José Luis Duarte Ribeiro, Maria Auxiliadora Cannarozzo Tinoco

Abstract:

As a result of the rapid growth of the Brazilian Air Transport, many airports are at the limit of their capacities and have a reduction in the quality of services provided. Thus, there is a need of models for assessing the quality of airport services. Because of this, the main objective of this work is to propose a model for the evaluation of quality attributes in airport services. To this end, we used the method composed by literature review and interview. Structured a working method composed by 5 steps, which resulted in a model to evaluate the quality of airport services, consisting of 8 dimensions and 45 attributes. Was used as base for model definition the process mapping of boarding and landing processes of passengers and luggage. As a contribution of this work is the integration of management process with structuring models to assess the quality of services in airport environments.

Keywords: quality airport services, model for identification of attributes quality, air transport, passenger

Procedia PDF Downloads 510

Authors: Qisheng Hu, Jianglei Han, Yue Yang, My Hoa Ha

Abstract:

When a company is considering purchasing software for commercial use, contract risk assessment is critical to identify risks to mitigate the potential adverse business impact, e.g., security, financial and regulatory risks. Contract risk assessment requires reviewers with specialized knowledge and time to evaluate the legal documents manually. Specifically, validating contracts for a software vendor requires the following steps: manual screening, interpreting legal documents, and extracting risk-prone segments. To automate the process, we proposed a framework to assist legal contract document risk identification, leveraging pre-trained deep learning models and natural language processing techniques. Given a set of pre-defined risk evaluation problems, our framework utilizes the pre-trained transformer-based models for question-answering to identify risk-prone sections in a contract. Furthermore, the question-answering model encodes the concatenated question-contract text and predicts the start and end position for clause extraction. Due to the limited labelled dataset for training, we leveraged transfer learning by fine-tuning the models with the CUAD dataset to enhance the model. On a dataset comprising 287 contract documents and 2000 labelled samples, our best model achieved an F1 score of 0.687.

Keywords: contract risk assessment, NLP, transfer learning, question answering

Procedia PDF Downloads 112

Authors: Abdullah A. AlShaher

Abstract:

In this paper, we demonstrate how regression curves can be used to recognize 2D non-rigid handwritten shapes. Each shape is represented by a set of non-overlapping uniformly distributed landmarks. The underlying models utilize 2^nd order of polynomials to model shapes within a training set. To estimate the regression models, we need to extract the required coefficients which describe the variations for a set of shape class. Hence, a least square method is used to estimate such modes. We then proceed by training these coefficients using the apparatus Expectation Maximization algorithm. Recognition is carried out by finding the least error landmarks displacement with respect to the model curves. Handwritten isolated Arabic characters are used to evaluate our approach.

Keywords: character recognition, regression curves, handwritten Arabic letters, expectation maximization algorithm

Procedia PDF Downloads 129