Search results for: deep learning

Authors: Ani Kojoyan

Abstract:

Linguistic studies have been investigating a connection between gender and linguistic development since 1970s. Scholars claim that gender differences in first and second language learning are socially constructed. Recent studies to language learning and gender reveal that second language acquisition is also a social phenomenon directly influencing one’s gender identity. Those responsible for designing language learning-teaching materials should be encouraged to understand the importance of and address the gender sensitivity accurately in textbooks. Writing or compiling a textbook is not an easy task; it requires strong academic abilities, patience, and experience. For a long period of time Armenia has been involved in the compilation process of a number of foreign language textbooks. However, there have been very few discussions or evaluations of those textbooks which will allow specialists to theorize that practice. The present paper focuses on the analysis of gender sensitivity issues and policy aspects involved in an EFL textbook. For the research the following material has been considered – “A Basic English Grammar: Morphology”, first printed in 2011. The selection of the material is not accidental. First, the mentioned textbook has been widely used in university teaching over years. Secondly, in Armenia “A Basic English Grammar: Morphology” has considered one of the most successful English grammar textbooks in a university teaching environment and served a source-book for other authors to compile and design their textbooks. The present paper aims to find out whether an EFL textbook is gendered in the Armenian teaching environment, and whether the textbook compilers are aware of gendered messages while compiling educational materials. It also aims at investigating students’ attitude toward the gendered messages in those materials. And finally, it also aims at increasing the gender sensitivity among book compilers and educators in various educational settings. For this study qualitative and quantitative research methods of analyses have been applied, the quantitative – in terms of carrying out surveys among students (45 university students, 18-25 age group), and the qualitative one – by discourse analysis of the material and conducting in-depth and semi-structured interviews with the Armenian compilers of the textbook (interviews with 3 authors). The study is based on passive and active observations and teaching experience done in a university classroom environment in 2014-2015, 2015-2016. The findings suggest that the discussed and analyzed teaching materials (145 extracts and examples) include traditional examples of intensive use of language and role-modelling, particularly, men are mostly portrayed as active, progressive, aggressive, whereas women are often depicted as passive and weak. These modeled often serve as a ‘reliable basis’ for reinforcing the traditional roles that have been projected on female and male students. The survey results also show that such materials contribute directly to shaping learners’ social attitudes and expectations around issues of gender. The applied techniques and discussed issues can be generalized and applied to other foreign language textbook compilation processes, since those principles, regardless of a language, are mostly the same.

Keywords: EFL textbooks, gender policy, gender sensitivity, qualitative and quantitative research methods

Procedia PDF Downloads 194

Authors: Hanan K. Hassanein

Abstract:

Task-based language teaching (TBLT) and focus on form instruction (FFI) methods were proven to improve quality and quantity of immediate language production. However, studies that compare between the effectiveness of the language production when using TBLT versus FFI are very little with results that are not consistent. Moreover, teaching Arabic using TBLT is a new field with few research that has investigated its application inside classrooms. Furthermore, to the best knowledge of the researcher, there are no prior studies that compared teaching Arabic as a foreign language in a classroom setting using computer-assisted task-based language teaching (CATBLT) with computer-assisted form focused language instruction (CAFFI). Accordingly, the focus of this presentation is to display CATBLT and CAFFI tools when teaching Arabic as a foreign language as well as demonstrate an experimental study that aims to identify whether or not CATBLT is a more effective instruction method. The effectiveness will be determined through comparing CATBLT and CAFFI in terms of accuracy, lexical complexity, and fluency of language produced by students. The participants of the study are 20 students enrolled in two intermediate-level Arabic as a foreign language classes. The experiment will take place over the course of 7 days. Based on a study conducted by Abdurrahman Arslanyilmaz for teaching Turkish as a second language, an in-house computer assisted tool for the TBLT and another one for FFI will be designed for the experiment. The experimental group will be instructed using the in-house CATBLT tool and the control group will be taught through the in-house CAFFI tool. The data that will be analyzed are the dialogues produced by students in both the experimental and control groups when completing a task or communicating in conversational activities. The dialogues of both groups will be analyzed to understand the effect of the type of instruction (CATBLT or CAFFI) on accuracy, lexical complexity, and fluency. Thus, the study aims to demonstrate whether or not there is an instruction method that positively affects the language produced by students learning Arabic as a foreign language more than the other.

Keywords: computer assisted language teaching, foreign language teaching, form-focused instruction, task based language teaching

Procedia PDF Downloads 249

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 41

Authors: Da Kuang

Abstract:

Resilience is a hotly debated topic in many research fields (e.g., engineering, ecology, society, psychology). In flood management studies, we are experiencing the paradigm shift from flood resistance to flood resilience. Flood resilience refers to tolerate flooding through adaptation or transformation. It is increasingly argued that our city as a social-ecological system holds the ability to learn from experience and adapt to flood rather than simply resist it. This research aims to investigate what kinds of adaptation knowledge the frequently flooded village learned from past experience and its advantages and limitations in coping with floods. The study area – Xinnongcun village, located in the west of Wuhan city, is a linear village and continuously suffered from both flash flood and drainage flood during the past 30 years. We have a field trip to the site in June 2017 and conducted semi-structured interviews with local residents. Our research summarizes two types of adaptation knowledge that people learned from the past floods. Firstly, at the village scale, it has formed a collective urban form which could help people live during both flood and dry season. All houses and front yards were elevated about 2m higher than the road. All the front yards in the village are linked and there is no barrier. During flooding time, people walk to neighbors through houses yards and boat to outside village on the lower road. Secondly, at individual scale, local people learned tacit knowledge of preparedness and emergency response to flood. Regarding the advantages and limitations, the adaptation knowledge could effectively help people to live with flood and reduce the chances of getting injuries. However, it cannot reduce local farmers’ losses on their agricultural land. After flood, it is impossible for local people to recover to the pre-disaster state as flood emerges during June and July will result in no harvest. Therefore, we argue that learning from past flood experience could increase people’s adaptive capacity. However, once the adaptive capacity cannot reduce people’s losses, it requires a transformation to a better regime.

Keywords: adaptation, flood resilience, tacit knowledge, transformation

Procedia PDF Downloads 333

Authors: A. Driouiche, S. Mohareb, A. Hadfi

Abstract:

In Morocco’s Agadir region, the reuse of treated wastewater for irrigation of green spaces has faced the problem of scaling of the pipes of these waters. This research paper aims at studying the phenomenon of scaling caused by the treated wastewater from the Mzar sewage treatment plant. These waters are used in the irrigation of golf turf for the Ocean Golf Resort. Ocean Golf, located about 10 km from the center of the city of Agadir, is one of the most important recreation centers in Morocco. The course is a Belt Collins design with 27 holes, and is quite open with deep challenging bunkers. The formation of solid deposits in the irrigation systems has led to a decrease in their lifetime and, consequently, a loss of load and performance. Thus, the sprinklers used in golf turf irrigation are plugged in the first weeks of operation. To study this phenomenon, the wastewater used for the irrigation of the golf turf was taken and analyzed at various points, and also samples of scale formed in the circuits of the passage of these waters were characterized. This characterization of the scale was performed by X-ray fluorescence spectrometry, X-ray diffraction (XRD), thermogravimetric analysis (TGA), differential thermal analysis (DTA), and scanning electron microscopy (SEM). The results of the physicochemical analysis of the waters show that they are full of bicarbonates (653 mg/L), chloride (478 mg/L), nitrate (412 mg/L), sodium (425 mg/L) and calcium (199mg/L). Their pH is slightly alkaline. The analysis of the scale reveals that it is rich in calcium and phosphorus. It is formed of calcium carbonate (CaCO₃), silica (SiO₂), calcium silicate (Ca₂SiO₄), hydroxylapatite (Ca₁₀P₆O₂₆), calcium carbonate and phosphate (Ca₁₀(PO₄) 6CO₃) and silicate calcium and magnesium (Ca₅MgSi₃O₁₂).

Keywords: Agadir, irrigation, scaling water, wastewater

Procedia PDF Downloads 118

Authors: Meng-Fei Cheng, Wei Fneg

Abstract:

New Taiwan high school standards emphasize employing scientific models and modeling practices in physics learning. However, to our knowledge. Few studies address how scientific models and modeling are approached in current science teaching, and they do not examine the views of scientific models portrayed in the textbooks. To explore the views of scientific models and modeling in textbooks, this study investigated the atomic unit in different textbook versions as an example and provided suggestions for modeling curriculum. This study adopted a quantitative analysis of qualitative data in the atomic units of four mainstream version of Taiwan high school physics textbooks. The models were further analyzed using five dimensions of the views of scientific models (nature of models, multiple models, purpose of the models, testing models, and changing models); each dimension had three levels (low, medium, high). Descriptive statistics were employed to compare the frequency of describing the five dimensions of the views of scientific models in the atomic unit to understand the emphasis of the views and to compare the frequency of the eight scientific models’ use to investigate the atomic model that was used most often in the textbooks. Descriptive statistics were further utilized to investigate the average levels of the five dimensions of the views of scientific models to examine whether the textbooks views were close to the scientific view. The average level of the five dimensions of the eight atomic models were also compared to examine whether the views of the eight atomic models were close to the scientific views. The results revealed the following three major findings from the atomic unit. (1) Among the five dimensions of the views of scientific models, the most portrayed dimension was the 'purpose of models,' and the least portrayed dimension was 'multiple models.' The most diverse view was the 'purpose of models,' and the most sophisticated scientific view was the 'nature of models.' The least sophisticated scientific view was 'multiple models.' (2) Among the eight atomic models, the most mentioned model was the atomic nucleus model, and the least mentioned model was the three states of matter. (3) Among the correlations between the five dimensions, the dimension of 'testing models' was highly related to the dimension of 'changing models.' In short, this study examined the views of scientific models based on the atomic units of physics textbooks to identify the emphasized and disregarded views in the textbooks. The findings suggest how future textbooks and curriculum can provide a thorough view of scientific models to enhance students' model-based learning.

Keywords: atomic models, textbooks, science education, scientific model

Procedia PDF Downloads 156

Authors: Evgeny E. Tereshatov, C. M. Folden

Abstract:

Ionic liquids and deep eutectic mixtures represent so-called non-conventional solvents. The former, composed of discrete ions, is a salt with a melting temperature below 100°С. The latter, consisting of hydrogen bond donors and acceptors, is a mixture of at least two compounds, resulting in a melting temperature depression in comparison with that of the individual moiety. These systems also can be water-immiscible, which makes them applicable for metal extraction. This work will cover interactions of In, Tl, Ir, and Rh in hydrochloric acid media with eutectic mixtures and Er, Ir, and At in a gas phase with chemically modified α-detectors. The purpose is to study chemical systems based on non-conventional solvents in terms of their interaction with metals. Once promising systems are found, the next step is to modify the surface of α-detectors used in the online element production at cyclotrons to get the detector chemical selectivity. Initially, the metal interactions are studied by means of the liquid-liquid extraction technique. Then appropriate molecules are chemisorbed on the surrogate surface first to understand the coating quality. Finally, a detector is covered with the same molecule, and the metal sorption on such detectors is studied in the online regime. It was found that chemical treatment of the surface can result in 99% coverage with a monolayer formation. This surface is chemically active and can adsorb metals from hydrochloric acid solutions. Similarly, a detector surface was modified and tested during cyclotron-based experiments. Thus, a procedure of detectors functionalization has been developed, and this opens an interesting opportunity of studying chemisorption of elements which do not have stable isotopes.

Keywords: mechanism, radioisotopes, solvent extraction, gas phase sorption

Procedia PDF Downloads 100

Authors: Filip Rusak, Yulia Arzhaeva, Dadong Wang

Abstract:

Machine learning based Computer-Aided Diagnosis (CAD) is gaining much popularity in medical imaging and diagnostic radiology. However, it requires a large amount of high quality and labeled training image datasets. The training images may come from different sources and be acquired from different radiography machines produced by different manufacturers, digital or digitized copies of film radiographs, with various sizes as well as different pixel intensity distributions. In this paper, a content-aware image augmentation method is presented to deal with these variations. The results of the proposed method have been validated graphically by plotting the removed and added seams of pixels on original images. Two different chest X-ray (CXR) datasets are used in the experiments. The CXRs in the datasets defer in size, some are digital CXRs while the others are digitized from analog CXR films. With the proposed content-aware augmentation method, the Seam Carving algorithm is employed to resize CXRs and the corresponding labels in the form of image masks, followed by histogram matching used to normalize the pixel intensities of digital radiography, based on the pixel intensity values of digitized radiographs. We implemented the algorithms, resized the well-known Montgomery dataset, to the size of the most frequently used Japanese Society of Radiological Technology (JSRT) dataset and normalized our digital CXRs for testing. This work resulted in the unified off-the-shelf CXR dataset composed of radiographs included in both, Montgomery and JSRT datasets. The experimental results show that even though the amount of augmentation is large, our algorithm can preserve the important information in lung fields, local structures, and global visual effect adequately. The proposed method can be used to augment training and testing image data sets so that the trained machine learning model can be used to process CXRs from various sources, and it can be potentially used broadly in any medical imaging applications.

Keywords: computer-aided diagnosis, image augmentation, lung segmentation, medical imaging, seam carving

Procedia PDF Downloads 220

Authors: Jun Young Chong, Seung Whan Kim

Abstract:

Introduction: The use of personal protective equipment for respiratory infection control in cardiopulmonary resuscitation (CPR) is a physical burden to healthcare providers. It matters how long CPR quality according to recommended guidelines can be maintained under these circumstances. It was investigated whether chest compression time was appropriate for a 2-minute shift and how long it was maintained in accordance with the guidelines under such conditions. Methods: This prospective crossover simulation study was performed at a single center from September 2020 to October 2020. Five indicators of CPR quality were measured during the first and second sessions of the study period. All participants wore a Level D powered air-purifying respirator (PAPR), and the experiment was conducted using a Resusci Anne manikin, which can measure the quality of chest compressions. Each participant conducted two sessions. In session one, 2-minutes of chest compressions followed by a 2-minute rest was repeated twice; in session two, 1-minute of chest compressions followed by a 1-minute rest was repeated four times. Results: All 34 participants completed the study. The deep and sufficient compression rate was 65.9 ± 13.1 mm in the 1-minute shift group and 61.5 ± 30.5 mm in the 2-minute shift group. The mean depth was 52.8 ±4.3 mm in the 1-minute shift group and 51.0 ± 6.1 mm in the 2-minute shift group. In these two values, there was a statistically significant difference between the two sessions. There was no statistically significant difference in the other CPR quality values. Conclusions: It was suggested that the different standard of current 2-minute to 1-minute cycles due to a significant reduction in the quality of chest compression in cases of CPR with PAPR.

Keywords: cardiopulmonary resuscitation, chest compression, personal protective equipment, powered air-purifying respirator

Procedia PDF Downloads 113

Authors: Lana Singleton

Abstract:

Achieving a sufficient global solution to climate change seems elusive through disappointing climate agreements and lack of cooperation. However, is this reluctance of coordination deep rooted on a more individual, societal level within countries due to a fundamental lack of social and institutional trust? The risks of climate change are illustrious and widely accepted, yet responses on an individual level are also largely inadequate. This research looks to further investigate types of trust, risk perception of climate change, and their interaction to build a greater understanding of whether a social trap (Rothstein, 2005) – where an absence of trust can overwhelm an individuals’ risk perception and result in minimal action despite knowing the dangers of no action – exists and where it is more prevalent. Presence of the social trap will be analysed for willingness to act in favour of climate change mitigation as well as attitude (acceptance) of different types of renewable energy forms. Using probit models with cross-sectional survey data on four developed European countries (UK, France, Germany, and Norway), we find evidence of the social trap in the aggregated data model, which highlights the importance of social trust regarding willingness to act in favour of climate mitigation as there is a high probability of action regardless of risk perception of climate change when social trust is high. In contrast, the same is not true for renewables, as interactions were mainly insignificant, although there were interesting findings involving institutional trust, gender, and country specific results for particular renewables.

Keywords: climate risk, renewables, risk perception, social trap, trust, willingness to act

Procedia PDF Downloads 94

Authors: Supriya Manikrao Ovhal

Abstract:

In India, the summer monsoon rainfall occurs owing to large scale convection with reference to continental ITCZ. It was found that convection over tropical ocean increases with SST from 26 to 28 degree C, and when SST is above 29 degree C, it sharply decreases for warm pool areas of Indian and for monsoon areas of West Pacific Ocean. The reduction in convection can be influenced by large scale subsidence forced by nearby or remotely generated deep convection, thus it was observed that under the influence of strong large scale rising motion, convection does not decreases but increases monotonically with SST even if SST value is higher than 29.5 degree C. Since convection is related to SST gradient, that helps to generate low level moisture convergence and upward vertical motion in the atmosphere. Strong wind fields like cross equatorial low level jet stream on equator ward side of the warm pool are produced due to convection initiated by SST gradient. Areas having maximum SST have low SST gradient, and that result in feeble convection. Hence it is imperative to mention that the oceanic role (other than SST) could be prominent in influencing large Scale Atmospheric convection. Since warm oceanic surface somewhere or the other contributes to penetrate the heat radiation to the subsurface of the ocean, and as there is no studies seen related to oceanic subsurface role in large Scale Atmospheric convection, in the present study, we are concentrating on the oceanic subsurface contribution in large Scale Atmospheric convection by considering the SST gradient, mixed layer depth (MLD), thermocline, barrier layer. The present study examines the probable role of subsurface ocean parameters in influencing convection.

Keywords: sst, d20, olr, wind

Procedia PDF Downloads 92

Authors: El Alia Boularas, H. Bekkar, H. Larachi, H. Rezk-kallah

Abstract:

Introduction: Notwithstanding cosmetics are used in life every day, these products are not all innocuous and harmless, as they may contain ingredients responsible for allergic reactions and, possibly, for other health problems. Additionally, environmental pollution should be taken into account. Thus, it is time to investigate what is ‘hidden behind beauty’. Aims: 1.To investigate prevalence of 13 chemical ingredients in cosmetics being object of concern, which the Algerians use regularly. 2.To know the profile of questioned consumers and describe their opinion on cosmetics. Methods: The survey was carried out in year 2013 over a period of 3 months, among Algerian Internet users having an e-mail address or a Facebook account.The study investigated 13 chemical agents showing health and environmental problems, selected after analysis of the recent studies published on the subject, the lists of national and international regulatory references on chemical hazards, and querying the database Skin Deep presented by the Environmental Working Group. Results: 300 people distributed all over the Algerian territory participated in the survey, providing information about 731 cosmetics; 86% aged from 20 to 39 years, with a sex ratio=0,27. A percentage of 43% of the analyzed cosmetics contained at least one of the 13 toxic ingredients. The targeted ingredient that has been most frequently reported was ‘perfume’ followed by parabens and PEG.85% of the participants declared that cosmetics ‘can contain toxic substances’, 27% asserted that they verify regularly the list of ingredients when they buy cosmetics, 61% said that they try to avoid the toxic ingredients, among whom 24 % were more vigilant on the presence of parabens, 95% were in favour of the strengthening of the Algerian laws on cosmetics. Conclusion: The results of the survey provide the indication of a widespread presence of toxic chemical ingredients in personal care products that Algerians use daily.

Keywords: Algerians consumers, cosmetics, survey, toxic ingredients

Procedia PDF Downloads 276

Authors: Zhilin Wang, Zhihao Zheng, Linxin Liu

Abstract:

This study proposes a low-code model-building approach that aims to simplify the development and deployment of artificial intelligence (AI) models. With an intuitive way to drag and drop and connect components, users can easily build complex models and integrate multiple algorithms for training. After the training is completed, the system automatically generates a callable model service API. This method not only lowers the technical threshold of AI development and improves development efficiency but also enhances the flexibility of algorithm integration and simplifies the deployment process of models. The core strength of this method lies in its ease of use and efficiency. Users do not need to have a deep programming background and can complete the design and implementation of complex models with a simple drag-and-drop operation. This feature greatly expands the scope of AI technology, allowing more non-technical people to participate in the development of AI models. At the same time, the method performs well in algorithm integration, supporting many different types of algorithms to work together, which further improves the performance and applicability of the model. In the experimental part, we performed several performance tests on the method. The results show that compared with traditional model construction methods, this method can make more efficient use, save computing resources, and greatly shorten the model training time. In addition, the system-generated model service interface has been optimized for high availability and scalability, which can adapt to the needs of different application scenarios.

Keywords: low-code, model building, artificial intelligence, algorithm integration, model deployment

Procedia PDF Downloads 27

Authors: Rohollah Asadian Kohestani, Nazanin Sedghi

Abstract:

As the number of banks and financial institutions working in Iran has been significantly increased, the attracting and retaining customers and encouraging them to continually use the modern banking services have been important and vital issues. Therefore, there would be a serious competition without a deep perception of consumers and fitness of banking services with their needs in the current economic conditions of Iran. It should be noted that concepts such as 'brand equity' is defined based on the view of consumers; however, it is also focused by shareholders, competitors and other beneficiaries of a firm in addition to bank and its consumers. This study examines the impact of brand equity on the competitive advantage in the banking industry as intensive competition between brands of different banks leads to pay more attention to the brands. This research is based on the Aaker’s model examining the impact of four dimensions of brand equity on the competitive advantage of private banks in Behshahr city. Moreover, conducting an applied research and data analysis has been carried out by a descriptive method. Data collection was done using literature review and questionnaire. A 'simple random' methodology was selected for sampling staff of banks while sampling methodology to select consumers of banks was the distribution of questionnaire between staff and consumers of five private banks including Tejarat, Mellat, Refah K., Ghavamin and, Tose’e Ta’avon banks. Results show that there is a significant relationship between brand equity and their competitive advantage. In this research, software of SPSS 16 and LISREL 8.5, as well as different methods of descriptive inferential statistics for analyzing data and test hypotheses, were employed.

Keywords: brand awareness, brand loyalty, brand equity, competitive advantage

Procedia PDF Downloads 137

Authors: Tsu-Yin Wu, Jenni Hoffman, Lydia McMurrows, Sarah Lally

Abstract:

Recent events of the Flint Water Crisis and elevated lead levels in Detroit public school water have highlighted a specific public health disparity and shown the need for better education of healthcare providers on lead education. Identifying children and pregnant women with a high risk for lead poisoning and ensuring lead testing is completed is critical. The purpose of this study is to explore the impact of an educational intervention on knowledge and confidence levels among nursing students enrolled in the prelicensure Bachelor of Science in Nursing (BSN) and Registered Nurse to BSN program (R2B). The study used both quantitative and qualitative research methods to assess the impact of multi-modal pedagogy on knowledge and confidence of lead screening and prevention among prelicensure and R2B nursing students. The students received lead poisoning and prevention content in addition to completing an e-learning module developed by the Pediatric Environmental Health Specialty Units. A total of 115 students completed the pre-and post-test instrument that consisted of demographic, lead knowledge, and confidence items. Despite the increase of total knowledge, three dimensions of lead poisoning, and confidence from pre- to post-test scores for both groups, there was no statistical significance on the increase between prelicensure and R2B students. Thematic analysis of qualitative data showed five themes from participants' learning experiences: lead exposure, signs and symptoms of lead poisoning, screening and diagnosis, prevention, and policy and statewide issues. The study is limited by a small sample and participants recalling some correct answers from the pretest, thus, scoring higher on the post-test. The results contribute to the minimally existent literature examining a critical public health concern regarding lead health exposure and prevention education of nursing students. Incorporating such content area into the nursing curriculum is essential in ensuring that such public health disparities are mitigated.

Keywords: lead poisoning, emerging public health issue, community health, nursing edducation

Procedia PDF Downloads 195

Authors: Eli Maria Forti-Pazzianotto, Carolina Moraes Da Costa, Daniela Faleiros Berteli Merino, Maura Rigoldi Simões Da Rocha, Irineu Rasera-Junior

Abstract:

There is no conclusive evidence to support the use of one type or brand of incentive espirometry over others. The decision as to which equipment is best, have being based on empirical assessment of patient acceptance, ease of use, and cost. The aim was to evaluate the effects of use of two methodologies of breathing exercises, performed by flow-oriented incentive spirometry, in the reversal of diaphragmatic dysfunction in postoperative bariatric surgery. 38 morbid obese women were selected. Respiratory muscle strength was evaluated through the nasal inspiratory pressure (NIP), and the respiratory muscles endurance, through incremental test by measurement of sustained maximal inspiratory pressure (SMIP). They were randomized in 2 groups: 1- Respiron® Classic (RC) the inspirations were slow, deep and sustained for as long as possible (5 sec). 2- Respiron® Athletic1 (RA1) - the inspirations were explosive, quick and intense, raising balls by the explosive way. 6 sets of 15 repetitions with intervals of 30 to 60 seconds were performed in groups. At the end of the intervention program (second PO), the volunteers were reevaluated. The groups were homogeneous with regard to initial assessment. However on reevaluating there was a significant decline of the variable PIN (p= < 0.0001) and SMIP (p=0.0004) in RC. In the RA1 group there was a maintenance of SMIP (p=0.5076) after surgery. The use of the Respiron Athletic 1, as well as the methodology of application used, can contribute positively to preserve the inspiratory muscle endurance and improve the diaphragmatic dysfunction in postoperative period.

Keywords: bariatric surgery, incentive spirometry, respiratory muscle, physiotherapy

Procedia PDF Downloads 371

Authors: S. Teksin, S. Yayla

Abstract:

The aim of the present study was to investigate details of flow structure in downstream of a circular cylinder base mounted on a flat surface in a rectangular duct with the dimensions of 8000 x 1000 x 750 mm in deep water flow for the Reynolds number 2500, 5000 and 7500. A flexible strip was attached to behind the cylinder and compared the bare body. Also, it was analyzed that how boundary layer affects the structure of flow around the cylinder. Diameter of the cylinder was 60 mm and the length of the flexible splitter plate which had a certain modulus of elasticity was 150 mm (L/D=2.5). Time-averaged velocity vectors, vortex contours, streamwise and transverse velocity components were investigated via Particle Image Velocimetry (PIV). Velocity vectors and vortex contours were displayed through the sections in which boundary layer effect was not present. On the other hand, streamwise and transverse velocity components were monitored for both cases, i.e. with and without boundary layer effect. Experiment results showed that the vortex formation occured in a larger area for L/D=2.5 and the point where the vortex was maximum from the base of the cylinder was shifted. Streamwise and transverse velocity component contours were symmetrical with reference to the center of the cylinder for all cases. All Froud numbers based on the Reynolds numbers were quite smaller than 1. The flow characteristics of velocity component values of attached circular cylinder arrangement decreased approximately twenty five percent comparing to bare cylinder case.

Keywords: partical image velocimetry, elastic plate, cylinder, flow structure

Procedia PDF Downloads 313

Authors: Rosa Figueroa, Christopher Flores

Abstract:

Text categorization is the problem of assigning a new document to a set of predetermined categories, on the basis of a training set of free-text data that contains documents whose category membership is known. To train a classification model, it is necessary to extract characteristics in the form of tokens that facilitate the learning and classification process. In text categorization, the feature extraction process involves the use of word sequences also known as N-grams. In general, it is expected that documents belonging to the same category share similar features. The Smith-Waterman (SW) algorithm is a dynamic programming algorithm that performs a local sequence alignment in order to determine similar regions between two strings or protein sequences. This work explores the use of SW algorithm as an alternative to feature extraction in text categorization. The dataset used for this purpose, contains 2,610 annotated documents with the classes Obese/Non-Obese. This dataset was represented in a matrix form using the Bag of Word approach. The score selected to represent the occurrence of the tokens in each document was the term frequency-inverse document frequency (TF-IDF). In order to extract features for classification, four experiments were conducted: the first experiment used SW to extract features, the second one used unigrams (single word), the third one used bigrams (two word sequence) and the last experiment used a combination of unigrams and bigrams to extract features for classification. To test the effectiveness of the extracted feature set for the four experiments, a Support Vector Machine (SVM) classifier was tuned using 20% of the dataset. The remaining 80% of the dataset together with 5-Fold Cross Validation were used to evaluate and compare the performance of the four experiments of feature extraction. Results from the tuning process suggest that SW performs better than the N-gram based feature extraction. These results were confirmed by using the remaining 80% of the dataset, where SW performed the best (accuracy = 97.10%, weighted average F-measure = 97.07%). The second best was obtained by the combination of unigrams-bigrams (accuracy = 96.04, weighted average F-measure = 95.97) closely followed by the bigrams (accuracy = 94.56%, weighted average F-measure = 94.46%) and finally unigrams (accuracy = 92.96%, weighted average F-measure = 92.90%).

Keywords: comorbidities, machine learning, obesity, Smith-Waterman algorithm

Procedia PDF Downloads 296

Authors: F. Deng, R. Sanchez, A. Beltran, S. Maestre

Abstract:

The high nutritional value associated with seafood is related to the presence of essential trace elements. Moreover, seafood is considered an important source of energy, proteins, and long-chain polyunsaturated fatty acids. Generally, seafood is consumed cooked. Consequently, the nutritional value could be degraded. Seafood, such as fish, shellfish, and seaweed, could be considered as one of the main iodine sources. The deficient or excessive consumption of iodine could cause dysfunction and pathologies related to the thyroid gland. The main objective of this work is to evaluated iodine stability in hake (Merluccius) undergone different culinary techniques. The culinary process considered were: boiling, steaming, microwave cooking, baking, cooking en papillote (twisted cover with the shape of a sweet wrapper) and coating with a batter of flour and deep-frying. The determination of iodine was carried by Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Regarding sample handling strategies, liquid-liquid extraction has demonstrated to be a powerful pre-concentration and clean-up approach for trace metal analysis by ICP techniques. Extraction with tetramethylammonium hydroxide (TMAH reagent) was used as a sample preparation method in this work. Based on the results, it can be concluded that the stability of iodine was degraded with the cooking processes. The major degradation was observed for the boiling and microwave cooking processes. The content of iodine in hake decreased up to 60% and 52%, respectively. However, if the boiling cooking liquid is preserved, this loss that has been generated during cooking is reduced. Only when the fish was cooked by following the cooking en papillote process the iodine content was preserved.

Keywords: cooking process, ICP-MS, iodine, hake

Procedia PDF Downloads 140

Authors: M. Cabrera, W. Tizani, J. Ninic, F. Wang

Abstract:

Open section beam to concrete-filled tubular column structures has been increasingly utilized in construction over the past few decades due to their enhanced structural performance, as well as economic and architectural advantages. However, the use of this configuration in construction is limited due to the difficulties in connecting the structural members as there is no access to the inner part of the tube to install standard bolts. Blind-bolted systems are a relatively new approach to overcome this limitation as they only require access to one side of the tubular section to tighten the bolt. The performance of these connections in concrete-filled steel tubular sections remains uncharacterized due to the complex interactions between concrete, bolt, and steel section. Over the last years, research in structural performance has moved to a more sophisticated and efficient approach consisting of machine learning algorithms to generate metamodels. This method reduces the need for developing complex, and computationally expensive finite element models, optimizing the search for desirable design variables. Metamodels generated by a data fusion approach use numerical and experimental results by combining multiple models to capture the dependency between the simulation design variables and connection performance, learning the relations between different design parameters and predicting a given output. Fully characterizing this connection will transform high-rise and multistorey construction by means of the introduction of design guidance for moment-resisting blind-bolted connections, which is currently unavailable. This paper presents a review of the steps taken to develop metamodels generated by means of artificial neural network algorithms which predict the connection stress and stiffness based on the design parameters when using Extended Hollo-Bolt blind bolts. It also provides consideration of the failure modes and mechanisms that contribute to the deformability as well as the feasibility of achieving blind-bolted rigid connections when using the blind fastener.

Keywords: blind-bolted connections, concrete-filled tubular structures, finite element analysis, metamodeling

Procedia PDF Downloads 157

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 157

Authors: Zafer Yüce, Paşa Yayla, Alev Taşkın

Abstract:

There are heaps of analytical methods to estimate the crack growth life of a component. Soft computing methods have an increasing trend in predicting fatigue life. Their ability to build complex relationships and capability to handle huge amounts of data are motivating researchers and industry professionals to employ them for challenging problems. This study focuses on soft computing methods, especially random forest regressors and artificial neural networks with hyperparameter optimization algorithms such as grid search and random grid search, to estimate the crack growth life of an aircraft wing splice joint under variable amplitude loading. TensorFlow and Scikit-learn libraries of Python are used to build the machine learning models for this study. The material considered in this work is 7050-T7451 aluminum, which is commonly preferred as a structural element in the aerospace industry, and regarding the crack type; corner crack is used. A finite element model is built for the joint to calculate fastener loads and stresses on the structure. Since finite element model results are validated with analytical calculations, findings of the finite element model are fed to AFGROW software to calculate analytical crack growth lives. Based on Fighter Aircraft Loading Standard for Fatigue (FALSTAFF), 90 unique fatigue loading spectra are developed for various load levels, and then, these spectrums are utilized as inputs to the artificial neural network and random forest regression models for predicting crack growth life. Finally, the crack growth life predictions of the machine learning models are compared with analytical calculations. According to the findings, a good correlation is observed between analytical and predicted crack growth lives.

Keywords: aircraft, fatigue, joint, life, optimization, prediction.

Procedia PDF Downloads 175

Authors: Ahmet Kurnaz, Eyup Yurt, Ümit Çiftci

Abstract:

In this study, too, an attempt was made to reveal the place and effects of information technologies on the lives and education of gifted children based on the views of gifted. To this end, the effects of information technologies on gifted are general skills, technology use, academic and social skills, and cooperative and personal skills were investigated. These skills were explored depending on whether or not gifted had their own computers, had internet connection at home, or how often they use the internet, average time period they spent at the computer, how often they played computer games and their use of social media. The study was conducted using the screening model with a quantitative approach. The sample of the study consisted of 129 gifted attending 5-12th classes in 12 provinces in different regions of Turkey. 64 of the participants were female while 65 were male. The research data were collected using the using computer of gifted and information technologies (UCIT) questionnaire which was developed by the researchers and given its final form after receiving expert view. As a result of the study, it was found that UCIT use improved foreign language speaking skills of gifted, enabled them to get to know and understand different cultures, and made use of computer and information technologies while they study. At the end of the study these result were obtained: Gifted have positive idea using computer and communication technology. There are differences whether using the internet about the ideas UCIT. But there are not differences whether having computer, inhabited city, grade level, having internet at home, daily and weekly internet usage durations, playing the computer and internet game, having Facebook and Twitter account about the UCIT. UCIT contribute to the development of gifted vocabulary, allows knowing and understand different cultures, developing foreign language speaking skills, gifted do not give up computer when they do their homework, improve their reading, listening, understanding and writing skills in a foreign language. Gifted children want to have transition to the use of tablets in education. They think UCIT facilitates doing their homework, contributes learning more information in a shorter time. They'd like to use computer-assisted instruction programs at courses. They think they will be more successful in the future if their computer skills are good. But gifted students prefer teacher instead of teaching with computers and they said that learning can be run from home without going to school.

Keywords: gifted, using computer, communication technology, information technologies

Procedia PDF Downloads 388

Authors: Huma Aly, Caramel Rodger, Saman Zafar

Abstract:

The present study explored the role of caste system in determining and understanding various personality characteristics related to different castes. It analyzed various personality characteristics of Arains, Jutts and Sheikhs caste of Pakistan. Reasons for the emphasis on within caste marriage in relation to personality characteristics were identified. In the present study a sample of 200 unmarried students were taken from different institutes of Lahore, Pakistan. 117 students were taken from Fast University and 83 from LUMS (Lahore University of Management and Sciences) on the basis of purposive and convenience sampling. 76 Arains, 59 Sheikhs and 65 Jutts were taken. Non-probability purposive sampling, quantitative research method, big five personality scale were used. Kruskal Wallis test was used as three independent groups were taken in the study. Results revealed various personality characteristics associated with different castes namely Arain, Jutts and Sheikhs. Individuals belonging to Jutts caste were reported to be high on being talkative, findings faults, doing thorough job, being depressed, reservedness, quarrelling, reliable, tensed, deep thinker, worrying a lot, imaginative, lazy, inventive, assertive, cold aloof, preserved and rude. Arains were reported to be original, helpful, careless,relaxed, curious, enthusiastic, forgiving, quiet, trusting, moody, shy, retaining anger, routinely working, planners, nervous, playing with ideas, artistic, cooperative, easily distracted and sophisticated. Lastly, Sheikhs were reported to be energetic, disorganized, stable. This study will play a significant part in changing the traditional viewpoint of majority of elders of our society who still have immense association with the caste they belong to.

Keywords: castes, personality, Arains, Jutts, Sheikhs, Pakistan

Procedia PDF Downloads 262

Authors: M. S. Maglasang, S. O. Palacio, L. P. Ogdoc

Abstract:

Business Process Outsourcing has been one of the fastest growing and emerging industry in the Philippines today. Unlike most of the contact service centers, more popularly known as "call centers", The BPO Industry’s primary outsourced service is performing audits of the global clients' logistics. As a service industry, manpower is considered as the most important yet the most expensive resource in the company. Because of this, there is a need to maximize the human resources so people are effectively and efficiently utilized. The main purpose of the study is to optimize the current manpower resources through effective distribution and assignment of different types of bills to the different skill-level of data entry operators. The assignment model parameters include the average observed time matrix gathered from through time study, which incorporates the learning curve concept. Subsequently, a simulation model was made to duplicate the arrival rate of demand which includes the different batches and types of bill per day. Next, a mathematical linear programming model was formulated. Its objective is to minimize direct labor cost per bill by allocating the different types of bills to the different skill-levels of operators. Finally, a hypothesis test was done to validate the model, comparing the actual and simulated results. The analysis of results revealed that the there’s low utilization of effective capacity because of its failure to determine the product-mix, skill-mix, and simulated demand as model parameters. Moreover, failure to consider the effects of learning curve leads to overestimation of labor needs. From 107 current number of operators, the proposed model gives a result of 79 operators. This results to an increase of utilization of effective capacity to 14.94%. It is recommended that the excess 28 operators would be reallocated to the other areas of the department. Finally, a manpower capacity planning model is also recommended in support to management’s decisions on what to do when the current capacity would reach its limit with the expected increasing demand.

Keywords: optimization modelling, linear programming, simulation, time and motion study, capacity planning

Procedia PDF Downloads 518

Authors: Aaidha Hammad

Abstract:

The age a child to be exposed to a second language is a controversial issue in communities such as the Maldives where English is taught as a second language. It has been observed that different stakeholders have different viewpoints towards the issue. Some believe that the earlier children are exposed to a second language, the better they learn, while others disagree with the notion. Hence, this case study investigates whether children learn a second language better when they are exposed at an earlier age or not. The spoken and written data collected confirm that earlier exposure helps in mastering the sound pattern and speaking fluency with more native-like accent, while a later age is better for learning more abstract and concrete aspects such as grammar and syntactic rules.

Keywords: age, fluency, second language acquisition, development of language skills

Procedia PDF Downloads 422

Authors: Ezil Sam Leni, Shalen S.

Abstract:

Transport service monitoring and upkeep are essential components of smart city initiatives. The main risks to the relevant departments and authorities are the ever-increasing vehicular traffic and the conditions of the roads. In India, the economy is greatly impacted by the road transport sector. In 2021, the Ministry of Road Transport and Highways Transport, Government of India, produced a report with statistical data on traffic accidents. The data included the number of fatalities, injuries, and other pertinent criteria. This study proposes a distributed infrastructure for the monitoring, detection, and reporting of potholes to the appropriate authorities. In a distributed environment, the nodes are the edge devices, and local edge servers, and global servers. The edge devices receive the initial model to be employed from the global server. The YOLOv8 model for pothole detection is used in the edge devices. The edge devices run the pothole detection model, gather the pothole images on their path, and send the updates to the nearby edge server. The local edge server selects the clients for its aggregation process, aggregates the model updates and sends the updates to the global server. The global server collects the updates from the local edge servers, performs aggregation and derives the updated model. The updated model has the information about the potholes received from the local edge servers and notifies the updates to the local edge servers and concerned authorities for monitoring and maintenance of road conditions. The entire process is implemented in FedCV distributed environment with the implementation using the client-server model and aggregation entities. After choosing the clients for its aggregation process, the local edge server gathers the model updates and transmits them to the global server. After gathering the updates from the regional edge servers, the global server aggregates them and creates the updated model. Performance indicators and the experimentation environment are assessed, discussed, and presented. Accelerometer data may be taken into consideration for improved performance in the future development of this study, in addition to the images captured from the transportation routes.

Keywords: federated Learning, pothole detection, distributed framework, federated averaging

Procedia PDF Downloads 101

Authors: Abimbola Akintounde

Abstract:

Supporting the social and emotional needs of young adolescents has become an emergent concern for schools around the world. Yet educators remain in a dilemma regarding the optimum approach for integrating social and emotional learning (SEL) into their content area instruction. The purpose of this study was to explore the perception of secondary students regarding their schoolwide SEL interventions. Twenty-four International Baccalaureate students in a final year mathematics course at an American Public Secondary School near Washington D. C. were randomly selected for participation in this study via an online electronic survey. The participants in this study used Likert-scale items to rate the effectiveness of the socio-emotional and character development programs being implemented at their schools. Respondents also ranked their preferred mode of delivery of social and emotional support programs. About 71% of the teenagers surveyed preferred SEL support rendered via interactive team-building activities and games, 42% of the high school students in the study ranked focus group discussions as their preferred format for SEL interventions, while only 13% of the respondents in the study regarded lectures and presentations as their preferred mode of SEL delivery. About one-fourth of the study participants agreed that explicit instruction was critical to enhancing students’ wellness, 79% agreed that SEL programs should foster less teacher talk, while 88% of the students indicated that student engagement was critical to their mental health. Eighty percent of the teenagers surveyed decried that the focus of their school-wide social and emotional programs was poorly prioritized. About two-thirds of the students agreed that social justice and equity issues should be embedded in their schools’ advisory programs. More than half of the respondents agitated for strategies for managing stress and their school workload. About 54% of the respondents also clamored for SEL programs that reinforce emotion regulation and coping strategies for anxiety. Based on the findings of this study, recommendations were proffered for best practices in the design and implementation of effective learner-friendly social and emotional development interventions.

Keywords: SEL, math anxiety, student support, emotion regulation, social awareness, self awareness, self management, relationship building

Procedia PDF Downloads 73

Authors: Seyed Mostafa Jafari Raad, Hassan Hassanzadeh

Abstract:

Injection of impurities along with CO₂ into saline aquifers provides an exceptional prospect for low-cost carbon capture and storage technologies and can potentially accelerate large-scale implementation of geological storage of CO₂. We have conducted linear stability analyses and numerical simulations to investigate the effects of permitted impurities in CO₂ streams on the onset of natural convection and dynamics of subsequent convective mixing. We have shown that the rate of dissolution of an impure CO₂ stream with H₂S highly depends on the operating conditions such as temperature, pressure, and composition of impurity. Contrary to findings of previous studies, our results show that an impurity such as H₂S can potentially reduce the onset time of natural convection and can accelerate the subsequent convective mixing. However, at the later times, the rate of convective dissolution is adversely affected by the impurities. Therefore, the injection of an impure CO₂ stream can be engineered to improve the rate of dissolution of CO₂, which leads to higher storage security and efficiency. Accordingly, we have identified the most favorable CO₂ stream compositions based on the geophysical properties of target aquifers. Information related to the onset of natural convection such as the scaling relations and the most favorable operating conditions for CO₂ storage developed in this study are important in proper design, site screening, characterization and safety of geological storage. This information can be used to either identify future geological candidates for acid gas disposal or reviewing the current operating conditions of licensed injection sites.

Keywords: CO₂ storage, solubility trapping, convective dissolution, storage efficiency

Procedia PDF Downloads 204

Authors: Essam Al Daoud

Abstract:

Gradient boosting methods have been proven to be a very important strategy. Many successful machine learning solutions were developed using the XGBoost and its derivatives. The aim of this study is to investigate and compare the efficiency of three gradient methods. Home credit dataset is used in this work which contains 219 features and 356251 records. However, new features are generated and several techniques are used to rank and select the best features. The implementation indicates that the LightGBM is faster and more accurate than CatBoost and XGBoost using variant number of features and records.

Keywords: gradient boosting, XGBoost, LightGBM, CatBoost, home credit

Procedia PDF Downloads 170