Search results for: performance criteria

Authors: Renée Fregosi

Abstract:

The contribution presently proposed will report on the beginnings of a Franco-Chilean study to be launched in 2015 by a multidisciplinary team of Renée Fregosi Political Science University Paris 3 / CECIEC, Norma Muñoz Public Policies University of Santiago of Chile, Jean-Daniel Lelievre, Medicine Paris 11 University, Marcelo WOLFF Medicine University of Chile, Cecilia Blatrix Political Science University Paris-Tech, Ernesto OTTONE, Political Science University of Chile, Paul DENY Medicine Paris 13 University, Rafael Bugueno Medicine Hospital Urgencia Pública of Santiago, Eduardo CARRASCO Political Science Paris 3 University. The problem of hidden populations challenges some criteria and concepts to re-examine: in particular the concept of target population, sampling methods to "snowball" and the cost-effectiveness criterion that shows the connection of political and scientific fields. Furthermore, if the pattern of homosexual transmission still makes up the highest percentage of the modes of infection with HIV, there is a continuous increase in the number of people infected through heterosexual sex, including women and persons aged 50 years and older. This group can be described as " unknown risk people." Access to these populations is a major challenge and raises methodological, ethical and political issues of prevention, particularly on the issue of screening. This paper proposes an inventory of these types of problems and their articulation, to define a new phase in the prevention against HIV refocused on women.

Keywords: HIV testing, hidden populations, difficult to reach PLWHA, women, unknown risk people

Procedia PDF Downloads 506

Authors: Joseph George, Anne Kotteswara Roa

Abstract:

Skin disease is one of the most common and popular kinds of health issues faced by people nowadays. Skin cancer (SC) is one among them, and its detection relies on the skin biopsy outputs and the expertise of the doctors, but it consumes more time and some inaccurate results. At the early stage, skin cancer detection is a challenging task, and it easily spreads to the whole body and leads to an increase in the mortality rate. Skin cancer is curable when it is detected at an early stage. In order to classify correct and accurate skin cancer, the critical task is skin cancer identification and classification, and it is more based on the cancer disease features such as shape, size, color, symmetry and etc. More similar characteristics are present in many skin diseases; hence it makes it a challenging issue to select important features from a skin cancer dataset images. Hence, the skin cancer diagnostic accuracy is improved by requiring an automated skin cancer detection and classification framework; thereby, the human expert’s scarcity is handled. Recently, the deep learning techniques like Convolutional neural network (CNN), Deep belief neural network (DBN), Artificial neural network (ANN), Recurrent neural network (RNN), and Long and short term memory (LSTM) have been widely used for the identification and classification of skin cancers. This survey reviews different DL techniques for skin cancer identification and classification. The performance metrics such as precision, recall, accuracy, sensitivity, specificity, and F-measures are used to evaluate the effectiveness of SC identification using DL techniques. By using these DL techniques, the classification accuracy increases along with the mitigation of computational complexities and time consumption.

Keywords: skin cancer, deep learning, performance measures, accuracy, datasets

Procedia PDF Downloads 112

Authors: Prachi Bhatt

Abstract:

Motivated workforce is an important asset of an organisation. Intrinsic motivation is one of the key aspects of people operations and performance. Researches have emphasized the significance of internal factors in individuals’ motivation. In the changing business scenario, it is a challenge for the organizations’ leaders to inspire and motivate their workforce. The present study deals with the intrinsic motivation potential of an individual which govern the innate capability of an individual driving him or her to behave or perform in the changing work environment, tasks, teams. Differences at individual level significantly influence differences in levels of motivation. In the above context, the present research attempts to explore behavioral trait dimensions which influence motivational potential of an individual. The present research emphasizes the significance of intrinsic motivational potential and the significance of exploring the differences in the intrinsic motivational potential levels of individuals at work places. Thus, this paper empirically tests the framework of behavioral traits which affects motivational potential of an individual. With the help of two studies i.e., Study 1 and Study 2, exploratory factor analysis and confirmatory factor analysis, respectively, indicated a reliable measure assessing intrinsic motivational potential of an individual. Given the variety of challenges of motivating contemporary workforce, and with increasing importance of intrinsic motivation, the paper discusses the relevance of the findings and of the measure assessing intrinsic motivational potential. Assessment of such behavioral traits would assist in the effective realization of intrinsic motivational potential of individuals. Additionally, the paper discusses the practical implications and furnishes scope for future research.

Keywords: behavioral traits, individual differences, intrinsic motivational potential, intrinsic motivation, motivation, workplace motivation

Procedia PDF Downloads 182

Authors: Blessing Ojeme, Frederick Quinn, Russell Karls, Shannon Quinn

Abstract:

The detection and segmentation of mitochondria from fluorescence microscopy are crucial for understanding the complex structure of the nervous system. However, the constant fission and fusion of mitochondria and image distortion in the background make the task of detection and segmentation challenging. In the literature, a number of open-source software tools and artificial intelligence (AI) methods have been described for analyzing mitochondrial images, achieving remarkable classification and quantitation results. However, the availability of combined expertise in the medical field and AI required to utilize these tools poses a challenge to its full adoption and use in clinical settings. Motivated by the advantages of automated methods in terms of good performance, minimum detection time, ease of implementation, and cross-platform compatibility, this study proposes a fully automated framework for the detection and segmentation of mitochondria using both image shape information and descriptive statistics. Using the low-cost, open-source python and openCV library, the algorithms are implemented in three stages: pre-processing, image binarization, and coarse-to-fine segmentation. The proposed model is validated using the mitochondrial fluorescence dataset. Ground truth labels generated using a Lab kit were also used to evaluate the performance of our detection and segmentation model. The study produces good detection and segmentation results and reports the challenges encountered during the image analysis of mitochondrial morphology from the fluorescence mitochondrial dataset. A discussion on the methods and future perspectives of fully automated frameworks conclude the paper.

Keywords: 2D, binarization, CLAHE, detection, fluorescence microscopy, mitochondria, segmentation

Procedia PDF Downloads 347

Authors: Van Thang Nguyen, Amelie Danlos, Richard Paridaens, Farid Bakir

Abstract:

This article presents a study of the effect of a contra-rotating component on the efficiency of centrifugal compressors. A contra-rotating centrifugal compressor (CRCC) is constructed using two independent rotors, rotating in the opposite direction and replacing the single rotor of a conventional centrifugal compressor (REF). To respect the geometrical parameters of the REF one, two rotors of the CRCC are designed, based on a single rotor geometry, using the hub and shroud length ratio parameter of the meridional contour. Firstly, the first rotor is designed by choosing a value of length ratio. Then, the second rotor is calculated to be adapted to the fluid flow of the first rotor according aerodynamics principles. In this study, four values of length ratios 0.3, 0.4, 0.5, and 0.6 are used to create four configurations CF1, CF2, CF3, and CF4 respectively. For comparison purpose, the circumferential velocity at the outlet of the REF and the CRCC are preserved, which means that the single rotor of the REF and the second rotor of the CRCC rotate with the same speed of 16000rpm. The speed of the first rotor in this case is chosen to be equal to the speed of the second rotor. The CFD simulation is conducted to compare the performance of the CRCC and the REF with the same boundary conditions. The results show that the configuration with a higher length ratio gives higher pressure rise. However, its efficiency is lower. An investigation over the entire operating range shows that the CF1 is the best configuration in this case. In addition, the CRCC can improve the pressure rise as well as the efficiency by changing the speed of each rotor independently. The results of changing the first rotor speed show with a 130% speed increase, the pressure ratio rises of 8.7% while the efficiency remains stable at the flow rate of the design operating point.

Keywords: centrifugal compressor, contra-rotating, interaction rotor, vacuum

Procedia PDF Downloads 121

Authors: Tamar Lazariashvili

Abstract:

There are a number of factors that determine the growth and development of the country's economy; however, trade and economic relations with other countries are the most important of all these factors. The paper analyzes the trade and economic relations between Georgia and Germany, identifies the impediments caused by the Covid pandemic, and substantiates the need for further economic cooperation between the countries. Research objectives. The objective of the research is to develop recommendations and reveal the prospects of further cooperation between Georgia and Germany based on identifying the problems in the field of trade and economy in the post-crisis situation. The research object is Georgian German economic relations. Germany is Georgia's largest trading partner in the European Union. Georgia and Germany actively cooperate within the framework of international organizations as well. The paper analyzes the multilateral and intensive economic relations between Germany and Georgia; evaluates the investments of German companies in Georgia and the activities of Georgian companies in Germany. Research methods. The paper uses general and specific research methods; in particular, analysis, synthesis, induction, deduction, comparison, statistical (selection, grouping, observation, trend), and other research methods.SWOT analysis is used to determine development opportunities between countries. As a result of the research economic ranking of Georgia and Germany are determined according to the above criteria, the causes of the impediments due to the pandemic are studied; the main problems in the field of trade and economy are identified. The paper provides conclusions on the problems in the trade relations between Georgia and Germany and suggests recommendations regarding the prospects for improving these relations.

Keywords: georgia-germany, trade and economic relations, economic ranking, perspective directions

Procedia PDF Downloads 145

Authors: Mehmet Demir Kaya, Engin Gökhan Kulan, Onur İleri, Süleyman Avcı

Abstract:

Due to variation in seed size, shape and oil content of safflower cultivars, germination and emergence performance have been severely influenced by seed characteristics. This study aimed to determine genotypic variation among safflower genotypes for one thousand seed weight, oil content, germination and seed vigor using electrical conductivity (EC) and cold test. In the study, safflower lines ES37-5, ES38-4, ES43-11, ES55-14 and ES58-11 which were developed by single seed selection method, and Dinçer and Remzibey-05 were used as standard varieties. The genotypes were grown under rainfed conditions in Eskişehir, Turkey with four replications. The seeds of each genotype were subjected to standard germination and emergence test at 25°C for 10 days with four replications and 50 seeds per replicate. Electrical conductivity test was performed at 25°C for 24 h to assess the seed vigor. Also, cold test were applied to each safflower genotype at 10°C for 4 days and 25°C for 6 days. Results showed that oil content of the safflower genotypes were different. The highest oil content was determined in ES43-11 with 36.6% while the lowest was 25.9% in ES38-4. Higher germination and emergence rate were obtained from ES55-14 with 96.5% and 73.0%, respectively. There was no significant difference among the safflower genotypes for EC values. Cold test showed that ES43-11 and ES55-14 gave the maximum germination percentages. It was concluded that genotypic factors except for soil and climatic conditions play an important role for determining seed vigor because safflower genotypes grown at the same condition produced various seed vigor values.

Keywords: Carthamus tinctorius L., germination, emergence, cold test, electrical conductivity

Procedia PDF Downloads 349

Authors: Z. Gharaee

Abstract:

Recognizing actions performed by others is important in our daily lives since it is necessary for communicating with others in a proper way. We perceive an action by observing the kinematics of motions involved in the performance. We use our experience and concepts to make a correct recognition of the actions. Although building the action concepts is a life-long process, which is repeated throughout life, we are very efficient in applying our learned concepts in analyzing motions and recognizing actions. Experiments on the subjects observing the actions performed by an actor show that an action is recognized after only about two hundred milliseconds of observation. In this study, hierarchical action recognition architecture is proposed by using growing grid layers. The first-layer growing grid receives the pre-processed data of consecutive 3D postures of joint positions and applies some heuristics during the growth phase to allocate areas of the map by inserting new neurons. As a result of training the first-layer growing grid, action pattern vectors are generated by connecting the elicited activations of the learned map. The ordered vector representation layer receives action pattern vectors to create time-invariant vectors of key elicited activations. Time-invariant vectors are sent to second-layer growing grid for categorization. This grid creates the clusters representing the actions. Finally, one-layer neural network developed by a delta rule labels the action categories in the last layer. System performance has been evaluated in an experiment with the publicly available MSR-Action3D dataset. There are actions performed by using different parts of human body: Hand Clap, Two Hands Wave, Side Boxing, Bend, Forward Kick, Side Kick, Jogging, Tennis Serve, Golf Swing, Pick Up and Throw. The growing grid architecture was trained by applying several random selections of generalization test data fed to the system during on average 100 epochs for each training of the first-layer growing grid and around 75 epochs for each training of the second-layer growing grid. The average generalization test accuracy is 92.6%. A comparison analysis between the performance of growing grid architecture and self-organizing map (SOM) architecture in terms of accuracy and learning speed show that the growing grid architecture is superior to the SOM architecture in action recognition task. The SOM architecture completes learning the same dataset of actions in around 150 epochs for each training of the first-layer SOM while it takes 1200 epochs for each training of the second-layer SOM and it achieves the average recognition accuracy of 90% for generalization test data. In summary, using the growing grid network preserves the fundamental features of SOMs, such as topographic organization of neurons, lateral interactions, the abilities of unsupervised learning and representing high dimensional input space in the lower dimensional maps. The architecture also benefits from an automatic size setting mechanism resulting in higher flexibility and robustness. Moreover, by utilizing growing grids the system automatically obtains a prior knowledge of input space during the growth phase and applies this information to expand the map by inserting new neurons wherever there is high representational demand.

Keywords: action recognition, growing grid, hierarchical architecture, neural networks, system performance

Procedia PDF Downloads 148

Authors: Philipp Stehle, Dragoljub Vrankovic, Montaha Anjass

Abstract:

Due to its high availability and extremely high specific capacity, silicon (Si) is the most promising anode material for next generation lithium-ion batteries (LIBs). However, Si anodes are suffering from high volume changes during cycling causing unstable solid-electrolyte interface (SEI). One approach for mitigation of these effects is to embed Si particles into a carbon matrix to create silicon/carbon composites (Si/C). These typically show more stable electrochemical performance than bare silicon materials. Nevertheless, the same failure mechanisms mentioned earlier appear in a less pronounced form. In this work, we further improved the cycling performance of two commercially available Si/C materials by coating thin metal oxide films of different thicknesses on the powders via Atomic Layer Deposition (ALD). The coated powders were analyzed via ICP-OES and AFM measurements. Si/C-graphite anodes with automotive-relevant loadings (~3.5 mAh/cm2) were processed out of the materials and tested in half coin cells (HCCs) and full pouch cells (FPCs). During long-term cycling in FPCs, a significant improvement was observed for some of the ALD-coated materials. After 500 cycles, the capacity retention was already up to 10% higher compared to the pristine materials. Cycling of the FPCs continued until they reached a state of health (SOH) of 80%. By this point, up to the triple number of cycles were achieved by ALD-coated compared to pristine anodes. Post-mortem analysis via various methods was carried out to evaluate the differences in SEI formation and thicknesses.

Keywords: silicon anodes, li-ion batteries, atomic layer deposition, silicon-carbon composites, surface coatings

Procedia PDF Downloads 106

Authors: Jude K. Safo

Abstract:

Knowledge Graphs KG) and their relation to Graph Embeddings (GE) represent a unique data structure in the landscape of machine learning (relative to image, text and acoustic data). Unlike the latter, GEs are the only data structure sufficient for representing hierarchically dense, semantic information needed for use-cases like supply chain data and protein folding where the search space exceeds the limits traditional search methods (e.g. page-rank, Dijkstra, etc.). While GEs are effective for compressing low rank tensor data, at scale, they begin to introduce a new problem of ’data retreival’ which we observe in Large Language Models. Notable attempts by transE, TransR and other prominent industry standards have shown a peak performance just north of 57% on WN18 and FB15K benchmarks, insufficient practical industry applications. They’re also limited, in scope, to next node/link predictions. Traditional linear methods like Tucker, CP, PARAFAC and CANDECOMP quickly hit memory limits on tensors exceeding 6.4 million nodes. This paper outlines a topological framework for linear mapping between concepts in KG space and GE space that preserve cardinality. Most importantly we introduce a traceable framework for composing dense linguistic strcutures. We demonstrate performance on WN18 benchmark this model hits. This model does not rely on Large Langauge Models (LLM) though the applications are certainy relevant here as well.

Keywords: representation theory, large language models, graph embeddings, applied algebraic topology, applied knot theory, combinatorics

Procedia PDF Downloads 57

Authors: Zahra Bahadoran, Parvin Mirmiran, Fereidoun Azizi

Abstract:

Background: Higher consumption of white rice has been suggested as a risk factor for development of metabolic abnormalities. In this study we investigated the association between consumption of white rice and the 3-year occurrence of metabolic syndrome (MetS) in adults with and without abdominal obesity. Methods: This longitudinal study was conducted within the framework of the Tehran Lipid and Glucose Study on 1476 adults, aged 19-70 years. Dietary intakes were measured, using a 168-food items validated semi-quantitative food frequency questionnaire at baseline. Biochemical and anthropometric measurements were evaluated at both baseline (2006-2008) and after 3-year follow-up (2009-2011). MetS and its components were defined according to the diagnostic criteria proposed by NCEP ATP III, and the new cutoff points of waist circumference for Iranian adults. Multiple logistic regression models were used to estimate the occurrence of the MetS in each quartile of white rice consumption. Results: The mean age of participants was 37.8±12.3 y, and mean BMI was 26.0±4.5 kg/m2 at baseline. The prevalence of MetS in subjects with abdominal obesity was significantly higher (40.9 vs. 16.2%, P<0.01). There was no significant difference in white rice consumption between the two groups. Mean daily intake of white rice was 93±59, 209±58, 262±60 and 432±224 g/d, in the first to fourth quartiles of white rice, respectively. Stratified analysis by categories of waist circumference showed that higher consumption of white rice was more strongly related to the risk of metabolic syndrome in participants who had abdominal obesity (OR: 2.34, 95% CI:1.14-4.41 vs. OR:0.99, 95% CI:0.60-1.65) Conclusion: We demonstrated that higher consumption of white rice may be a risk for development of metabolic syndrome in adults with abdominal obesity.

Keywords: white rice, abdominal obesity, metabolic syndrome, food science, triglycerides

Procedia PDF Downloads 432

Authors: Guermache Ahmed Chafik, Khelfellah Ismail, Ait-Ali Takfarines

Abstract:

The Savonius vertical axis wind turbine is characterized by sufficient starting torque at low wind speeds, simple design and does not require orientation to the wind direction; however, the developed power is lower than other types of wind turbines such as Darrieus. To increase these performances several studies and researches have been developed, such as optimizing blades shape, using passive controls and also minimizing power losses sources like the resisting torque due to friction. This work aims to estimate the performance of a Savonius wind turbine introducing a User Defined Function to the CFD model analyzing resisting torque. This User Defined Function is developed to simulate the action of the wind speed on the rotor; it receives the moment coefficient as an input to compute the rotational velocity that should be imposed on computational domain rotating regions. The rotational velocity depends on the aerodynamic moment applied on the turbine and the resisting torque, which is considered a linear function. Linking the implemented User Defined Function with the CFD solver allows simulating the real functioning of the Savonius turbine exposed to wind. It is noticed that the wind turbine takes a while to reach the stationary regime where the rotational velocity becomes invariable; at that moment, the tip speed ratio, the moment and power coefficients are computed. To validate this approach, the power coefficient versus tip speed ratio curve is compared with the experimental one. The obtained results are in agreement with the available experimental results.

Keywords: resistant torque modeling, Savonius wind turbine, user-defined function, vertical axis wind turbine performances

Procedia PDF Downloads 144

Authors: Mijoh A. Gbededo, Kapila Liyanage, Ilias Oraifige

Abstract:

Decision making for sustainable manufacturing design and management requires critical considerations due to the complexity and partly conflicting issues of economic, social and environmental factors. Although there are tools capable of assessing the combination of one or two of the sustainability factors, the frameworks have not adequately integrated all the three factors. Case study and review of existing simulation applications also shows the approach lacks integration of the sustainability factors. In this paper we discussed the development of a simulation based framework for support of a holistic assessment of sustainable manufacturing design and management. To achieve this, a strategic approach is introduced to investigate the strengths and weaknesses of the existing decision supporting tools. Investigation reveals that Discrete Event Simulation (DES) can serve as a rock base for other Life Cycle Analysis frameworks. Simio-DES application optimizes systems for both economic and competitive advantage, Granta CES EduPack and SimaPro collate data for Material Flow Analysis and environmental Life Cycle Assessment, while social and stakeholders’ analysis is supported by Analytical Hierarchy Process, a Multi-Criteria Decision Analysis method. Such a common and integrated framework creates a platform for companies to build a computer simulation model of a real system and assess the impact of alternative solutions before implementing a chosen solution.

Keywords: discrete event simulation, life cycle sustainability analysis, manufacturing, sustainability

Procedia PDF Downloads 267

Authors: Ayda Youssef, Mohammed Ali Khalaf, Tarek Rafaat

Abstract:

Background: Langerhans cell histiocytosis (LCH) is a rare multi-systemic disease that shows an abnormal proliferation of these kinds of cells associated with a granular infiltration that affects different structures of the human body, including the lung, liver, spleen, lymph nodes, brain, mucocutaneous, soft tissue (head and neck), and salivary glands. Evaluation of the extent of disease is one of the major predictors of patient outcome. Objectives: To recognize the pathogenesis of Langerhans cell histiocytosis (LCH), describe the radiologic criteria that are suggestive of LCH in different organs rather than the bones and to illustrate the appropriate differential diagnoses for LCH in each of the common extra-osseous sites. Material and methods: A retrospective study was done on 150 biopsy-proven LCH patients from 2007 to 2012. All patients underwent imaging studies, mostly US, CT, and MRI. These patients were reviewed to assess the extra-skeletal manifestations of LCH. Results: In 150 patients with biopsy-proven LCH, There were 33 patients with liver affection, 5 patients with splenic lesions, 55 patients with enlarged lymph nodes, 9 patient with CNS disease and 11 patients with lung involvement. Conclusions: Because of the frequent LCH children and evaluation of the extent of disease is one of the major predictors of patient outcome. Radiologist need to be familiar with its presentation in different organs and regions of body outside the commonest site of affection (bones). A high-index suspicion should be raised a biopsy is recommended in the presence of radiological suspicion. Chemotherapy is the preferred therapeutic modality.

Keywords: langerhans cell histiocytosis, extra-skeletal, pediatrics, radiology

Procedia PDF Downloads 425

Authors: Ana Pego

Abstract:

Environmental technologies were developed for decades. Clean technologies emerged a few years ago. In these perspectives, the use of cleantech technologies has become very important due the fact of new era of environmental feats. As such, the market itself has become more competitive, more collaborative towards a better use of clean technologies. This paper shows the importance of clean technologies in offshore energy sector in Portuguese market, its localization and its impact on economy. Clean technologies are directly related with renewable cluster and concomitant with economic and social resource optimization criteria, geographic aspects, climate change and soil features. Cleantech is related with regional development, socio-technical transitions in organisations. There are an economical and social combinations which allow specialisation of regions in activities, higher employment, reduce of energy costs, local knowledge spillover and, business collaboration and competitiveness. The methodology used will be quantitative (IO matrix for Portugal 2013) and qualitative (questionnaires to stakeholders). The mix of both methodologies will confirm whether the use of technologies will allow a positive impact on economic and social variables used on this model. It is expected a positive impact on Portuguese economy both in investment and employment taking in account the localization of offshore renewable activities. This means that the importance of offshore renewable investment in Portugal has a few points which should be pointed out: the increase of specialised employment, localization of specific activities in territory, and increase of value added in certain regions. The conclusion will allow researchers and organisation to compare the Portuguese model to other European regions in order to a better use of natural and human resources.

Keywords: cleantech, economic impact, localisation, territory dynamics

Procedia PDF Downloads 219

Authors: Jaballah Jamil

Abstract:

KLD (Peter Kinder, Steve Lydenberg and Amy Domini) research & analytics is an independent intermediary of social performance information that adopts an investor-pay model. KLD rating agency does not have an explicit monitoring on the rated firm which suggests that KLD ratings may not include private informations. Moreover, the incapacity of KLD to predict accurately the extra-financial rating of Enron casts doubt on the reliability of KLD ratings. Therefore, we first investigate whether KLD ratings affect investors' perception by studying the effect of KLD rating changes on firms' financial performances. Second, we study the impact of the Enron scandal on investors' perception of KLD rating changes by comparing the effect of KLD rating changes on firms' financial performances before and after the failure of Enron. We propose an empirical study that relates a number of equally-weighted portfolios returns, excess stock returns and book-to-market ratio to different dimensions of KLD social responsibility ratings. We first find that over the last two decades KLD rating changes influence significantly and negatively stock returns and book-to-market ratio of rated firms. This finding suggests that a raise in corporate social responsibility rating lowers the firm's risk. Second, to assess the Enron scandal's effect on the perception of KLD ratings, we compare the effect of KLD rating changes before and after the Enron scandal. We find that after the Enron scandal this significant effect disappears. This finding supports the view that the Enron scandal annihilates the KLD's effect on Socially Responsible Investors. Therefore, our findings may question results of recent studies that use KLD ratings as a proxy for Corporate Social Responsibility behavior.

Keywords: KLD social rating agency, investors' perception, investment decision, financial performance

Procedia PDF Downloads 426

Authors: Jarrod Haar, David Brougham

Abstract:

The present study explores the links between supporting employee well-being and the potential benefits to employee performance. We focus on employee innovative work behaviors (IWBs), which have three stages: (1) development, (2) adoption, and (3) implementation of new ideas and work methods. We explore the role of organizational support focusing on employee well-being via High-Performance Work Systems (HPWS). HPWS are HR practices that are designed to enhance employees’ skills, commitment, and ultimately, productivity. HPWS influence employee performance through building their skills, knowledge, and abilities and there is meta-analytic support for firm-level HPWS influencing firm performance, but less attention towards employee outcomes, especially innovation. We explore HPWS-wellbeing being offered (e.g., EAPs, well-being App, etc.) to capture organizational commitment to employee well-being. Under social exchange theory, workers should reciprocate their firm's offering of HPWS-wellbeing with greater efforts towards IWBs. Further, we explore playful work design as a mediator, which represents employees proactively creating work conditions that foster enjoyment/challenge but don’t require any design change to the job itself. We suggest HPWS-wellbeing can encourage employees to become more playful, and ultimately more innovative. Finally, beyond direct effects, we examine whether these relations are similar by gender and ultimately test a moderated mediation model. Using N=1135 New Zealand employees, we established measures with confirmatory factor analysis (CFA), and all measures had good psychometric properties (α>.80). We controlled for age, tenure, education, and hours worked and analyzed data using the PROCESS macro (version 4.2) specifically model 8 (moderated mediation). We analyzed overall IWB, and then again across the three stages. Overall, we find HPWS-wellbeing is significantly related to overall IWBs and the three stages (development, adoption, and implementation) individually. Similarly, HPWS-wellbeing shapes playful work design and playful work design predicts overall IWBs and the three stages individually. It only partially mediates the effects of HPWS-wellbeing, which retains a significant indirect effect. Moderation effects are supported, with males reporting a more significant effect from HPWS-wellbeing on playful work design but not IWB (or any of the three stages) than females. Females report higher playful work design when HPWS-wellbeing is low, but the effects are reversed when HPWS-wellbeing is high (males higher). Thus, males respond stronger under social exchange theory from HPWS-wellbeing, at least towards expressing playful work design. Finally, evidence of moderated mediation effects is found on overall IWBs and the three stages. Males report a significant indirect effect from HPWS-wellbeing on IWB (through playful work design), while female employees report no significant indirect effect. The benefits of playful work design fully account for their IWBs. The models account for small amounts of variance towards playful work design (12%) but larger for IWBs (26%). The study highlights a gap in the literature on HPWS-wellbeing and provides empirical evidence of their importance towards worker innovation. Further, gendered effects suggest these benefits might not be equal. The findings provide useful insights for organizations around how providing HR practices that support employee well-being are important, although how they work for different genders needs further exploration.

Keywords: human resource practices, wellbeing, innovation, playful work design

Procedia PDF Downloads 69

Authors: Andrea N. Ofori-Boadu

Abstract:

The purpose is to develop a conceptual model of subcontractor development practices and processes that strengthen the integration of subcontractors into construction supply chain systems for improved subcontractor performance on Leadership in Energy and Environmental Design (LEED) certified building projects. The construction management of a LEED project has an important objective of meeting sustainability certification requirements. This is in addition to the typical project management objectives of cost, time, quality, and safety for traditional projects; and, therefore increases the complexity of LEED projects. Considering that construction management organizations rely heavily on subcontractors, poor performance on complex projects such as LEED projects has been largely attributed to the unsatisfactory preparation of subcontractors. Furthermore, the extensive use of unique and non-repetitive short term contracts limits the full integration of subcontractors into construction supply chains and hinders long-term cooperation and benefits that could enhance performance on construction projects. Improved subcontractor development practices are needed to better prepare and manage subcontractors, so that complex objectives can be met or exceeded. While supplier development and supply chain theories and practices for the manufacturing sector have been extensively investigated to address similar challenges, investigations in the construction sector are not that obvious. Consequently, the objective of this research is to investigate effective subcontractor development practices and processes to guide construction management organizations in their development of a strong network of high performing subcontractors. Drawing from foundational supply chain and supplier development theories in the manufacturing sector, a mixed interpretivist and empirical methodology is utilized to assess the body of knowledge within literature for conceptual model development. A self-reporting survey with five-point Likert scale items and open-ended questions is administered to 30 construction professionals to estimate their perceptions of the effectiveness of 37 practices, classified into five subcontractor development categories. Data analysis includes descriptive statistics, weighted means, and t-tests that guide the effectiveness ranking of practices and categories. The results inform the proposed three-phased LEED subcontractor development program model which focuses on preparation, development and implementation, and monitoring. Highly ranked LEED subcontractor pre-qualification, commitment, incentives, evaluation, and feedback practices are perceived as more effective, when compared to practices requiring more direct involvement and linkages between subcontractors and construction management organizations. This is attributed to unfamiliarity, conflicting interests, lack of trust, and resource sharing challenges. With strategic modifications, the recommended practices can be extended to other non-LEED complex projects. Additional research is needed to guide the development of subcontractor development programs that strengthen direct involvement between construction management organizations and their network of high performing subcontractors. Insights from this present research strengthen theoretical foundations to support future research towards more integrated construction supply chains. In the long-term, this would lead to increased performance, profits and client satisfaction.

Keywords: construction management, general contractor, supply chain, sustainable construction

Procedia PDF Downloads 101

Authors: Zohreh Moghaddas, Morteza Yazdani, Farhad Hosseinzadeh

Abstract:

Nowadays, data envelopment analysis (DEA) models featuring network structures have gained widespread usage for evaluating the performance of production systems and activities (Decision-Making Units (DMUs)) across diverse fields. By examining the relationships between the internal stages of the network, these models offer valuable insights to managers and decision-makers regarding the performance of each stage and its impact on the overall network. To further empower system decision-makers, the inverse data envelopment analysis (IDEA) model has been introduced. This model allows the estimation of crucial information for estimating parameters while keeping the efficiency score unchanged or improved, enabling analysis of the sensitivity of system inputs or outputs according to managers' preferences. This empowers managers to apply their preferences and policies on resources, such as inputs and outputs, and analyze various aspects like production, resource allocation processes, and resource efficiency enhancement within the system. The results obtained can be instrumental in making informed decisions in the future. The top result of this study is an analysis of infeasibility and incorrect estimation that may arise in the theory and application of the inverse model of data envelopment analysis with network structures. By addressing these pitfalls, novel protocols are proposed to circumvent these shortcomings effectively. Subsequently, several theoretical and applied problems are examined and resolved through insightful case studies.

Keywords: inverse models of data envelopment analysis, series network, estimation of inputs and outputs, efficiency, resource allocation, sensitivity analysis, infeasibility

Procedia PDF Downloads 25

Authors: Akan C. Offong, E. J. Anthony, Vasilije Manovic

Abstract:

A modified steady-state numerical model is developed for the electrochemical reduction of CO₂ to formic acid. The numerical model achieves a CD (current density) (~60 mA/cm²), FE-faradaic efficiency (~98%) and conversion (~80%) for CO₂ electro-reduction to formic acid in a microfluidic cell. The model integrates charge and species transport, mass conservation, and momentum with electrochemistry. Specifically, the influences of Bi-Sn based nanoparticle catalyst (on the cathode surface) at different mole fractions and 1-ethyl-3-methyl imidazolium tetra-fluoroborate ([EMIM][BF₄]) electrolyte, on CD, FE and CO₂ conversion to formic acid is studied. The reaction is carried out at a constant concentration of electrolyte (85% v/v., [EMIM][BF₄]). Based on the mass transfer characteristics analysis (concentration contours), mole ratio 0.5:0.5 Bi-Sn catalyst displays the highest CO₂ mole consumption in the cathode gas channel. After validating with experimental data (polarisation curves) from literature, extensive simulations reveal performance measure: CD, FE and CO₂ conversion. Increasing the negative cathode potential increases the current densities for both formic acid and H₂ formations. However, H₂ formations are minimal as a result of insufficient hydrogen ions in the ionic liquid electrolyte. Moreover, the limited hydrogen ions have a negative effect on formic acid CD. As CO₂ flow rate increases, CD, FE and CO₂ conversion increases.

Keywords: carbon dioxide, electro-chemical reduction, ionic liquids, microfluidics, modelling

Procedia PDF Downloads 133

Authors: Kassahun Birhanu Tadesse, Megersa Olumana Dinka

Abstract:

Reliable future river flow information is basic for planning and management of any river systems. For data scarce river system having only a river flow records like the Waterval River, a univariate time series models are appropriate for river flow forecasting. In this study, a univariate Seasonal Autoregressive Integrated Moving Average (SARIMA) model was applied for forecasting Waterval River flow using GRETL statistical software. Mean monthly river flows from 1960 to 2016 were used for modeling. Different unit root tests and Mann-Kendall trend analysis were performed to test the stationarity of the observed flow time series. The time series was differenced to remove the seasonality. Using the correlogram of seasonally differenced time series, different SARIMA models were identified, their parameters were estimated, and diagnostic check-up of model forecasts was performed using white noise and heteroscedasticity tests. Finally, based on minimum Akaike Information (AIc) and Hannan-Quinn (HQc) criteria, SARIMA (3, 0, 2) x (3, 1, 3)12 was selected as the best model for Waterval River flow forecasting. Therefore, this model can be used to generate future river information for water resources development and management in Waterval River system. SARIMA model can also be used for forecasting other similar univariate time series with seasonality characteristics.

Keywords: heteroscedasticity, stationarity test, trend analysis, validation, white noise

Procedia PDF Downloads 194

Authors: Haiu-Lan Chin, Yu-Fen Hsiao, Hua-Ying Chuang, Wei Lee

Abstract:

An adult with intellectual disability generally has insufficient physical activity which is an important factor leading to premature weakness. Studies in recent years on frailty syndrome have accumulated substantial data about indicators of human aging, including unintentional weight loss, self-reported exhaustion, weakness, slow walking speed, and low physical activity. Of these indicators, hand-grip strength can be seen as a predictor of mortality, disability, complications, and increased length of hospital stay. Hand-grip strength in fact provides a comprehensive overview of one’s vitality. The research is about the investigation on hand-grip strength of adults with intellectual disabilities in facilities, institutions and workshops. The participants are 197 male adults (M=39.09±12.85 years old), and 114 female ones (M=35.80±8.2 years old) so far. The aim of the study is to figure out the performance of their hand-grip strength, and initiate the setting of training on hand-grip strength in their daily life which will decrease the weakening on their physical condition. Test items include weight, bone density, basal metabolic rate (BMR), static body balance except hand-grip strength. Hand-grip strength was measured by a hand dynamometer and classified as normal group ( ≧ 30 kg for male and ≧ 20 kg for female) and weak group ( < 30 kg for male, < 20 kg for female)The analysis includes descriptive statistics, and the indicators of grip strength fo the adults with intellectual disability. Though the research is still ongoing and the participants are increasing, the data indicates: (1) The correlation between hand-grip strength and degree of the intellectual disability (p ≦. 001), basal metabolic rate (p ≦ .001), and static body balance (p ≦ .01) as well. Nevertheless, there is no significant correlation between grip strength and basal metabolic rate which had been having significant correlation with hand-grip strength. (2) The difference between male and female subjects in hand-grip strength is significant, the hand-grip strength of male subjects (25.70±12.81 Kg) is much higher than female ones (16.30±8.89 Kg). Compared to the female counterparts, male participants indicate greater individual differences. And the proportion of weakness between male and female subjects is also different. (3) The regression indicates the main factors related to grip strength performance include degree of the intellectual disability, height, static body balance, training and weight sequentially. (4) There is significant difference on both hand-grip and static body balance between participants in facilities and workshops. The study supports the truth about the sex and gender differences in health. Nevertheless, the average hand-grip strength of left hand is higher than right hand in both male and female subjects. Moreover, 71.3% of male subjects and 64.2% of female subjects have better performance in their left hand-grip which is distinctive features especially in low degree of the intellectual disability.

Keywords: adult with intellectual disability, frailty syndrome, grip strength, physical condition

Procedia PDF Downloads 166

Authors: M. Govindarajan

Abstract:

One of the major developments in machine learning in the past decade is the ensemble method, which finds highly accurate classifier by combining many moderately accurate component classifiers. In this research work, new ensemble classification methods are proposed with homogeneous ensemble classifier using bagging and heterogeneous ensemble classifier using arcing and their performances are analyzed in terms of accuracy. A Classifier ensemble is designed using Radial Basis Function (RBF) and Support Vector Machine (SVM) as base classifiers. The feasibility and the benefits of the proposed approaches are demonstrated by the means of standard datasets of intrusion detection. The main originality of the proposed approach is based on three main parts: preprocessing phase, classification phase, and combining phase. A wide range of comparative experiments is conducted for standard datasets of intrusion detection. The performance of the proposed homogeneous and heterogeneous ensemble classifiers are compared to the performance of other standard homogeneous and heterogeneous ensemble methods. The standard homogeneous ensemble methods include Error correcting output codes, Dagging and heterogeneous ensemble methods include majority voting, stacking. The proposed ensemble methods provide significant improvement of accuracy compared to individual classifiers and the proposed bagged RBF and SVM performs significantly better than ECOC and Dagging and the proposed hybrid RBF-SVM performs significantly better than voting and stacking. Also heterogeneous models exhibit better results than homogeneous models for standard datasets of intrusion detection. 

Keywords: data mining, ensemble, radial basis function, support vector machine, accuracy

Procedia PDF Downloads 234

Authors: Raviporn Madarasmi, Anjalee Vacharaksa, Pravej Serichetaphongse

Abstract:

The host response in peri-implant tissue may differ from that in periodontal tissue in a healthy individual. The purpose of this study is to investigate the expression of inflammatory cytokines in peri-implant crevicular fluid (PICF) from single implant with different abutment types in comparison to healthy periodontal tissue. 19 participants with healthy implants and teeth were recruited according to inclusion and exclusion criteria. PICF and gingival crevicular fluid (GCF) was collected using sterile paper points. The expression level of inflammatory cytokines including IL-1α, IL-1β, TNF-α, IFN-γ, IL-6, and IL-8 was assessed using enzyme-linked immunosorbent assay (ELISA). Paired t test was used to compare the expression levels of inflammatory cytokines around natural teeth and peri-implant in PICF and GCF of the same individual. The Independent t-test was used to compare the expression levels of inflammatory cytokines in PICF from titanium and UCLA abutment. Expression of IL-6, TNF-α, and IFN-γ in PICF was not statistically different from GCF among titanium and UCLA abutment group. However, the level of IL-1α in the PICF from the implants with UCLA abutment was significantly higher than GCF (P=0.030). In addition, the level of IL-1β in PICF from the implants with titanium abutment was significantly higher than GCF (P=0.032). When different abutment types was compared, IL-8 expression in PICF from implants with UCLA abutment was significantly higher than titanium abutment (P=0.003).

Keywords: abutment, dental implant, gingival crevicular fluid and peri-implant crevicular fluid

Procedia PDF Downloads 171

Authors: Bita Payami-Shabestari, Dariush Eslami

Abstract:

The purpose of this paper is to develop a multi-product economic production quantity model under vendor management inventory policy and restrictions including limited warehouse space, budget, and number of orders, average shortage time and maximum permissible shortage. Since the “costs” cannot be predicted with certainty, it is assumed that data behave under uncertain environment. The problem is first formulated into the framework of a bi-objective of multi-product economic production quantity model. Then, the problem is solved with three multi-objective decision-making (MODM) methods. Then following this, three methods had been compared on information on the optimal value of the two objective functions and the central processing unit (CPU) time with the statistical analysis method and the multi-attribute decision-making (MADM). The results are compared with statistical analysis method and the MADM. The results of the study demonstrate that augmented-constraint in terms of optimal value of the two objective functions and the CPU time perform better than global criteria, and goal programming. Sensitivity analysis is done to illustrate the effect of parameter variations on the optimal solution. The contribution of this research is the use of random costs data in developing a multi-product economic production quantity model under vendor management inventory policy with several constraints.

Keywords: economic production quantity, random cost, supply chain management, vendor-managed inventory

Procedia PDF Downloads 113

Authors: Jutta Ransmayr

Abstract:

The digital age has long arrived at Austrian schools, so both society and educationalists demand that digital means should be integrated accordingly to day-to-day school routines. Therefore, the Austrian school-leaving exam (A-levels) can now be written either by hand or by using a computer. However, the choice of writing medium (pen and paper or computer) for written examination papers, which are considered 'high-stakes' exams, raises a number of questions that have not yet been adequately investigated and answered until recently, such as: What effects do the different conditions of text production in the written German A-levels have on the component of normative linguistic accuracy? How do the spelling skills of German A-level papers written with a pen differ from those that the students wrote on the computer? And how is the teacher's assessment related to this? Which practical desiderata for German didactics can be derived from this? In a trilateral pilot project of the Austrian Center for Digital Humanities (ACDH) of the Austrian Academy of Sciences and the University of Vienna in cooperation with the Austrian Ministry of Education and the Council for German Orthography, these questions were investigated. A representative Austrian learner corpus, consisting of around 530 German A-level papers from all over Austria (pen and computer written), was set up in order to subject it to a quantitative (corpus-linguistic and statistical) and qualitative investigation with regard to the spelling and punctuation performance of the high school graduates and the differences between pen- and computer-written papers and their assessments. Relevant studies are currently available mainly from the Anglophone world. These have shown that writing on the computer increases the motivation to write, has positive effects on the length of the text, and, in some cases, also on the quality of the text. Depending on the writing situation and other technical aids, better results in terms of spelling and punctuation could also be found in the computer-written texts as compared to the handwritten ones. Studies also point towards a tendency among teachers to rate handwritten texts better than computer-written texts. In this paper, the first comparable results from the German-speaking area are to be presented. Research results have shown that, on the one hand, there are significant differences between handwritten and computer-written work with regard to performance in orthography and punctuation. On the other hand, the corpus linguistic investigation and the subsequent statistical analysis made it clear that not only the teachers' assessments of the students’ spelling performance vary enormously but also the overall assessments of the exam papers – the factor of the production medium (pen and paper or computer) also seems to play a decisive role.

Keywords: exam paper assessment, pen and paper or computer, learner corpora, linguistics

Procedia PDF Downloads 155

Authors: Seyed Saeid Tabaee, Omid Bahar

Abstract:

Nowadays, using energy dissipation devices has been commonly used in structures. A high rate of energy absorption during earthquakes is the benefit of using such devices, which results in damage reduction of structural elements specifically columns. The hysteretic damping capacity of energy dissipation devices is the key point that it may adversely complicate analysis and design of such structures. This effect may be generally represented by equivalent viscous damping. The equivalent viscous damping may be obtained from the expected hysteretic behavior under the design or maximum considered displacement of a structure. In this paper, the hysteretic damping coefficient of a steel moment resisting frame (MRF), which its performance is enhanced by a buckling restrained brace (BRB) system has been evaluated. Having the foresight of damping fraction between BRB and MRF is inevitable for seismic design procedures like Direct Displacement-Based Design (DDBD) method. This paper presents an approach to calculate the damping fraction for such systems by carrying out the dynamic nonlinear time history analysis (NTHA) under harmonic loading, which is tuned to the natural frequency of the system. Two steel moment frame structures, one equipped with BRB, and the other without BRB are simultaneously studied. The extensive analysis shows that proportion of each system damping fraction may be calculated by its shear story portion. In this way, the contribution of each BRB in the floors and their general contribution in the structural performance may be clearly recognized, in advance.

Keywords: buckling restrained brace, direct displacement based design, dual systems, hysteretic damping, moment resisting frames

Procedia PDF Downloads 421

Authors: Yanhua Zhao, Chenli Rao, Dongdong Li, Chuanmin Tao

Abstract:

The Chinese Centers for Disease Control and Prevention (CCDC) issued a new laboratory-based algorithm for HIV diagnosis on April 2016, which initially screens with a combination HIV-1/HIV-2 antigen/antibody fourth-generation immunoassay (IA) followed, when reactive, an HIV-1/HIV-2 undifferentiated antibody IA in duplicate. Reactive specimens with concordant results undergo supplemental tests with western blots, or HIV-1 nucleic acid tests (NATs) and non-reactive specimens with discordant results receive HIV-1 NATs or p24 antigen tests or 2-4 weeks follow-up tests. However, little data evaluating the application of the new algorithm have been reported to date. The study was to evaluate the performance of new laboratory-based HIV diagnostic algorithm in an inpatient population of Southwest China over the initial 6 months by compared with the old algorithm. Plasma specimens collected from inpatients from May 1, 2016, to October 31, 2016, are submitted to the laboratory for screening HIV infection performed by both the new HIV testing algorithm and the old version. The sensitivity and specificity of the algorithms and the difference of the categorized numbers of plasmas were calculated. Under the new algorithm for HIV diagnosis, 170 of the total 52 749 plasma specimens were confirmed as positively HIV-infected (0.32%). The sensitivity and specificity of the new algorithm were 100% (170/170) and 100% (52 579/52 579), respectively; while 167 HIV-1 positive specimens were identified by the old algorithm with sensitivity 98.24% (167/170) and 100% (52 579/52 579), respectively. Three acute HIV-1 infections (AHIs) and two early HIV-1 infections (EHIs) were identified by the new algorithm; the former was missed by old procedure. Compared with the old version, the new algorithm produced fewer WB-indeterminate results (2 vs. 16, p = 0.001), which led to fewer follow-up tests. Therefore, the new HIV testing algorithm is more sensitive for detecting acute HIV-1 infections with maintaining the ability to verify the established HIV-1 infections and can dramatically decrease the greater number of WB-indeterminate specimens.

Keywords: algorithm, diagnosis, HIV, laboratory

Procedia PDF Downloads 384

Authors: Kedar Hardikar, Joe Varghese

Abstract:

Conductive adhesives have found wide-ranging applications in electronics industry ranging from fixing a defective conductor on printed circuit board (PCB) attaching an electronic component in an assembly to protecting electronics components by the formation of “Faraday Cage.” The reliability requirements for the conductive adhesive vary widely depending on the application and expected product lifetime. While the conductive adhesive is required to maintain the structural integrity, the electrical performance of the associated sub-assembly can be affected by the degradation of conductive adhesive. The degradation of the adhesive is dependent upon the highly varied use case. The conventional approach to assess the reliability of the sub-assembly involves subjecting it to the standard environmental test conditions such as high-temperature high humidity, thermal cycling, high-temperature exposure to name a few. In order to enable projection of test data and observed failures to predict field performance, systematic development of an acceleration factor between the test conditions and field conditions is crucial. Common acceleration factor models such as Arrhenius model are based on rate kinetics and typically rely on an assumption of linear degradation in time for a given condition and test duration. The application of interest in this work involves conductive adhesive used in an electronic circuit of a capacitive sensor. The degradation of conductive adhesive in high temperature and humidity environment is quantified by the capacitance values. Under such conditions, the use of established models such as Hallberg-Peck model or Eyring Model to predict time to failure in the field typically relies on linear degradation rate. In this particular case, it is seen that the degradation is nonlinear in time and exhibits a square root t dependence. It is also shown that for the mechanism of interest, the presence of moisture is essential, and the dominant mechanism driving the degradation is the diffusion of moisture. In this work, a framework is developed to incorporate nonlinear degradation of the conductive adhesive for the development of an acceleration factor. This method can be extended to applications where nonlinearity in degradation rate can be adequately characterized in tests. It is shown that depending on the expected product lifetime, the use of conventional linear degradation approach can overestimate or underestimate the field performance. This work provides guidelines for suitability of linear degradation approximation for such varied applications

Keywords: conductive adhesives, nonlinear degradation, physics of failure, acceleration factor model.

Procedia PDF Downloads 119

Authors: Alanoud Moraya Aldalan, Abdulaziz Almaleh

Abstract:

A crucial component of maintaining a customer-oriented business as in the telecom industry is understanding the reasons and factors that lead to customer churn. Competition between telecom companies has greatly increased in recent years. It has become more important to understand customers’ needs in this strong market of telecom industries, especially for those who are looking to turn over their service providers. So, predictive churn is now a mandatory requirement for retaining those customers. Machine learning can be utilized to accomplish this. Churn Prediction has become a very important topic in terms of machine learning classification in the telecommunications industry. Understanding the factors of customer churn and how they behave is very important to building an effective churn prediction model. This paper aims to predict churn and identify factors of customers’ churn based on their past service usage history. Aiming at this objective, the study makes use of feature selection, normalization, and feature engineering. Then, this study compared the performance of four different machine learning algorithms on the Orange dataset: Logistic Regression, Random Forest, Decision Tree, and Gradient Boosting. Evaluation of the performance was conducted by using the F1 score and ROC-AUC. Comparing the results of this study with existing models has proven to produce better results. The results showed the Gradients Boosting with feature selection technique outperformed in this study by achieving a 99% F1-score and 99% AUC, and all other experiments achieved good results as well.

Keywords: machine learning, gradient boosting, logistic regression, churn, random forest, decision tree, ROC, AUC, F1-score

Procedia PDF Downloads 122