Search results for: neural smith predictor

Authors: Jarek Krajewski, Andre Wittenborn, Martin Sauerland

Abstract:

This abstract describes a promising approach for estimating post-traumatic stress disorder (PTSD) based on prosodic speech characteristics. It illustrates the validity of this method by briefly discussing results from an Arabic refugee sample (N= 47, 32 m, 15 f). A well-established standardized self-report scale “Reaction of Adolescents to Traumatic Stress” (RATS) was used to determine the ground truth level of PTSD. The speech material was prompted by telling about autobiographical related sadness inducing experiences (sampling rate 16 kHz, 8 bit resolution). In order to investigate PTSD-induced speech changes, a self-developed set of 136 prosodic speech features was extracted from the .wav files. This set was adapted to capture traumatization related speech phenomena. An artificial neural network (ANN) machine learning model was applied to determine the PTSD level and reached a correlation of r = .37. These results indicate that our classifiers can achieve similar results to those seen in speech-based stress research.

Keywords: speech prosody, PTSD, machine learning, feature extraction

Procedia PDF Downloads 90

Authors: L. S. Chathurika

Abstract:

Early prediction of student performance is an important factor to be gained academic excellence. Whatever the study stream in secondary education, students lay the foundation for higher studies during the first year of their degree or diploma program in Sri Lanka. The information technology (IT) field has certain improvements in the education domain by selecting specialization areas to show the talents and skills of students. These specializations can be software engineering, network administration, database administration, multimedia design, etc. After completing the first-year, students attempt to select the best path by considering numerous factors. The purpose of this experiment is to predict the best study path using machine learning algorithms. Five classification algorithms: decision tree, support vector machine, artificial neural network, Naïve Bayes, and logistic regression are selected and tested. The support vector machine obtained the highest accuracy, 82.4%. Then affecting features are recognized to select the best study path.

Keywords: algorithm, classification, evaluation, features, testing, training

Procedia PDF Downloads 119

Authors: Ci Lin, Tet Yeap, Iluju Kiringa

Abstract:

This paper introduces the subspace rotation algorithm (SRA) to train the Restricted Hopfield Network (RHN) as an auto-associative memory. Subspace rotation algorithm is a gradient-free subspace tracking approach based on the singular value decomposition (SVD). In comparison with Backpropagation Through Time (BPTT) on training RHN, it is observed that SRA could always converge to the optimal solution and BPTT could not achieve the same performance when the model becomes complex, and the number of patterns is large. The AUTS case study showed that the RHN model trained by SRA could achieve a better structure of attraction basin with larger radius(in general) than the Hopfield Network(HNN) model trained by Hebbian learning rule. Through learning 10000 patterns from MNIST dataset with RHN models with different number of hidden nodes, it is observed that an several components could be adjusted to achieve a balance between recovery accuracy and noise resistance.

Keywords: hopfield neural network, restricted hopfield network, subspace rotation algorithm, hebbian learning rule

Procedia PDF Downloads 117

Authors: Madan Khadka, Dhruba Uprety, Rubina Rai

Abstract:

Background and objective: In 2011, around 273,465 women died worldwide during pregnancy, childbirth or within 42 days after childbirth. Near-miss is recognized as the predictor of the level of care and maternal death. The objective of the study was to evaluate the associated risk factors of near-miss obstetric cases and maternal death. Material and Methods A Prospective Observational Study was done from August 1, 2014, to June 30, 2015, in Department of Obstetrics and Gynecology at BPKIHS hospital, tertiary care hospital in Eastern Nepal, Dharan. Case eligible by the 5-factor scoring system and WHO near miss criteria were evaluated. Risk factors included severe hemorrhage, hypertensive disorders, and a complication of abortion, ruptured uterus, medical/surgical condition and sepsis. Results: A total of 9,727 delivery were attended during the study period from August 2014 to June 2014. There were 6307 (71.5%) vaginal delivery and 2777(28.5%) caesarean section and 181 perinatal death with a total of 9,546 live birth. A total of 162 near miss was identified, and 16 maternal death occurred during the study. Maternal near miss rate of 16.6 per 1000 live birth, Women with life-threatening conditions (WLTC) of 172, Severe maternal outcome ratio of 18.64 per 1000 live birth, Maternal near-miss mortality ratio (MNM: 1 MD) 10.1:1, Mortality index (MI) of 8.98%. Risk factors were obstetric hemorrhage 27.8%, abortion/ectopic 27.2%, eclampsia 16%, medical/surgical condition 14.8%, sepsis 13.6%, severe preeclamsia 11.1%, ruptured uterus 3.1%, and molar pregnancy 1.9%. 19.75% were prim gravidae, with mean age 25.66 yrs, and cardiovascular and coagulation dysfunction as a major life threatening condition and sepsis (25%) was the major cause of mortality. Conclusion: Hemorrhage and hypertensive disorders are the leading causes of near miss event and sepsis as a leading cause of mortality. As near miss analysis indicates the quality of health care, it is worth presenting in national indices.

Keywords: abortion, eclampsia, hemorrhage, maternal mortility, near miss

Procedia PDF Downloads 196

Authors: Armineh Nourbakhsh, Sameena Shah, Carolyn Rose

Abstract:

Layout-aware language models have been used to create multimodal representations for documents that are in image form, achieving relatively high accuracy in document understanding tasks. However, the large number of parameters in the resulting models makes building and using them prohibitive without access to high-performing processing units with large memory capacity. We propose an alternative approach that can create efficient representations without the need for a neural visual backbone. This leads to an 80% reduction in the number of parameters compared to the smallest SOTA model, widely expanding applicability. In addition, our layout embeddings are pre-trained on spatial and visual cues alone and only fused with text embeddings in downstream tasks, which can facilitate applicability to low-resource of multi-lingual domains. Despite using 2.5% of training data, we show competitive performance on two form understanding tasks: semantic labeling and link prediction.

Keywords: layout understanding, form understanding, multimodal document understanding, bias-augmented attention

Procedia PDF Downloads 148

Authors: Ayesha Sanjana Kawser Parsha

Abstract:

S-acylation is an irreversible bond in which cysteine residues are linked to fatty acids palmitate (74%) or stearate (22%), either at the COOH or NH2 terminal, via a thioester linkage. There are several experimental methods that can be used to identify the S-palmitoylation site; however, since they require a lot of time, computational methods are becoming increasingly necessary. There aren't many predictors, however, that can locate S- palmitoylation sites in Arabidopsis Thaliana with sufficient accuracy. This research is based on the importance of building a better prediction tool. To identify the type of machine learning algorithm that predicts this site more accurately for the experimental dataset, several prediction tools were examined in this research, including the GPS PALM 6.0, pCysMod, GPS LIPID 1.0, CSS PALM 4.0, and NBA PALM. These analyses were conducted by constructing the receiver operating characteristics plot and the area under the curve score. An AI-driven deep learning-based prediction tool has been developed utilizing the analysis and three sequence-based input data, such as the amino acid composition, binary encoding profile, and autocorrelation features. The model was developed using five layers, two activation functions, associated parameters, and hyperparameters. The model was built using various combinations of features, and after training and validation, it performed better when all the features were present while using the experimental dataset for 8 and 10-fold cross-validations. While testing the model with unseen and new data, such as the GPS PALM 6.0 plant and pCysMod mouse, the model performed better, and the area under the curve score was near 1. It can be demonstrated that this model outperforms the prior tools in predicting the S- palmitoylation site in the experimental data set by comparing the area under curve score of 10-fold cross-validation of the new model with the established tools' area under curve score with their respective training sets. The objective of this study is to develop a prediction tool for Arabidopsis Thaliana that is more accurate than current tools, as measured by the area under the curve score. Plant food production and immunological treatment targets can both be managed by utilizing this method to forecast S- palmitoylation sites.

Keywords: S- palmitoylation, ROC PLOT, area under the curve, cross- validation score

Procedia PDF Downloads 77

Authors: Bharatendra Rai

Abstract:

The sequence of words in text data has long-term dependencies and is known to suffer from vanishing gradient problems when developing deep learning models. Although recurrent networks such as long short-term memory networks help to overcome this problem, achieving high text classification performance is a challenging problem. Convolutional recurrent networks that combine the advantages of long short-term memory networks and convolutional neural networks can be useful for text classification performance improvements. However, arriving at suitable hyperparameter values for convolutional recurrent networks is still a challenging task where fitting a model requires significant computing resources. This paper illustrates the advantages of using convolutional recurrent networks for text classification with the help of statistically planned computer experiments for hyperparameter tuning.

Keywords: long short-term memory networks, convolutional recurrent networks, text classification, hyperparameter tuning, Tukey honest significant differences

Procedia PDF Downloads 129

Authors: Mohsen Goodarzi, George Berghorn

Abstract:

Despite the rapid growth in the number of green building and community development projects, the long-term performance of these projects has not yet been sufficiently evaluated from the users’ points of view. This is partially due to the lack of post-occupancy evaluation tools available for this type of project. In this study, a post-construction evaluation model is developed to evaluate the relationship between the perceived performance and satisfaction of residents in LEED-certified residential buildings and communities. To develop this evaluation model, a primary five-factor model was developed based on the existing models and residential satisfaction theories. Each factor of the model included several measures that were adopted from LEED certification systems such as LEED-BD+C New Construction, LEED-BD+C Multifamily Midrise, LEED-ND, as well as the UC Berkeley’s Center for the Built Environment survey tool. The model included four predictor variables (factors), including perceived building performance (8 measures), perceived infrastructure performance (9 measures), perceived neighborhood design (6 measures), and perceived economic performance (4 measures), and one dependent variable (factor), which was residential satisfaction (6 measures). An online survey was then conducted to collect the data from the residents of LEED-certified residential communities (n=192) and the validity of the model was tested through Confirmatory Factor Analysis (CFA). After modifying the CFA model, 26 measures, out of the initial 33 measures, were retained to enter into a Structural Equation Model (SEM) and to find the relationships between the perceived buildings performance, infrastructure performance, neighborhood design, economic performance and residential Satisfaction. The results of the SEM showed that the perceived building performance was the most influential factor in determining residential satisfaction in LEED-certified communities, followed by the perceived neighborhood design. On the other hand, perceived infrastructure performance and perceived economic performance did not show any significant relationship with residential satisfaction in these communities. This study can benefit green building researchers by providing a model for the evaluation of the long-term performance of these projects. It can also provide opportunities for green building practitioners to determine priorities for future residential development projects.

Keywords: green building, residential satisfaction, perceived performance, confirmatory factor analysis, structural equation modeling

Procedia PDF Downloads 239

Authors: M. Hilani, B. Nassih

Abstract:

Phishing is one of the oldest and best known scams on the Internet. It can be defined as any type of telecommunications fraud that uses social engineering tricks to obtain confidential data from its victims. It’s a cybercrime aimed at stealing your sensitive information. Phishing is generally done via private email, so scammers impersonate large companies or other trusted entities to encourage victims to voluntarily provide information such as login credentials or, worse yet, credit card numbers. The COVID-19 theme is used by cybercriminals in multiple malicious campaigns like phishing. In this environment, messaging filtering solutions have become essential to protect devices that will now be used outside of the secure perimeter. Despite constantly updating methods to avoid these cyberattacks, the end result is currently insufficient. Many researchers are looking for optimal solutions to filter phishing emails, but we still need good results. In this work, we concentrated on solving the problem of detecting phishing emails using the different steps of NLP preprocessing, and we proposed and trained a model using one-dimensional CNN. Our study results show that our model obtained an accuracy of 99.99%, which demonstrates how well our model is working.

Keywords: phishing, e-mail, NLP preprocessing, CNN, e-mail filtering

Procedia PDF Downloads 126

Authors: S. Kawther, C. Marshall

Abstract:

Educational neuroscience has the intention of transforming research findings of the underpinning neural processes of learning to educational practices. A main criticism of the field, hitherto, is that less focus has been put on studying the in-progress practical application of these findings. Therefore, this study aims to gain a better understanding of teachers’ perceptions of the practical application and utilization of brain knowledge. This was approached by investigating the answer to 'What are the facilitators and barriers for bringing research from neuroscience to bear on education?'. Following a qualitative design, semi-structured interviews were conducted with 12 teachers who had a proficient course in educational neuroscience. Thematic analysis was performed on the transcribed data applying Braun & Clark’s steps. Findings emerged with four main themes: time, knowledge, teacher’s involvement, and system. These themes revealed that some effective brain-based practices are being engaged in by the teachers. However, the lack of guidance and challenges regarding this implementation were also found. This study discusses findings in light of the development of educational neuroscience implementation.

Keywords: brain-based, educational neuroscience, neuroeducation, neuroscience-informed

Procedia PDF Downloads 168

Authors: Hadeer M. Sayed, Hesham E. El Deeb, Shereen A. Taie

Abstract:

In the big data era, we are facing a diversity of datasets from different sources in different domains that describe a single life event. These datasets consist of multiple modalities, each of which has a different representation, distribution, scale, and density. Multimodal fusion is the concept of integrating information from multiple modalities in a joint representation with the goal of predicting an outcome through a classification task or regression task. In this paper, multimodal fusion techniques are classified into two main classes: model-agnostic techniques and model-based approaches. It provides a comprehensive study of recent research in each class and outlines the benefits and limitations of each of them. Furthermore, the audiovisual speech recognition task is expressed as a case study of multimodal data fusion approaches, and the open issues through the limitations of the current studies are presented. This paper can be considered a powerful guide for interested researchers in the field of multimodal data fusion and audiovisual speech recognition particularly.

Keywords: multimodal data, data fusion, audio-visual speech recognition, neural networks

Procedia PDF Downloads 112

Authors: Katalin Szöllősi, László Szabó

Abstract:

Introduction and purpose: Even though female sexual dysfunctions are common among women in the postpartum period, the profile of these disturbances has not been well investigated in Hungary yet. The aim of the study was to evaluate the postpartum female sexual functions in Hungary. This research sought to investigate the possible predictor factors which can influence postpartum female sexual functions. Method and sample: This was a cross-sectional study, including patients from two maternity clinics in Budapest. 113 women were recruited into our study 3 months after their childbirth. 53 had vaginal birth, 60 had a caesarian section. Data were collected from medical reports in addition by using self-developed questions and validated questionnaires in order to measure important predictors which may be responsible for postpartum sexual dysfunctions such as mode of delivery, parity, urinary incontinence and body image. Sexual functions were evaluated by the Hungarian version of the Female Sexual Function Index (FSFI). The Hungarian version of Body Image Questionnaire-Short Form14 (BSQ-SF14) was applied for assessing body image. Results: 82,3% of the participants began to have sexual intercourse within three months postpartum. 53,98% of the participants reported sexual dysfunctions (cut-off FSFI score 26,55). According to our results mode of delivery, parity, hemorrhoids, time of intercourse, resumption was not associated with female sexual dysfunctions. We found correlation at a tendential level between urinary incontinence and sexual dysfunctions (p=0,003, R=0,26). We found a negative correlation at a tendential level between the total score of BSQ-SF14 and FSFI (p=0,03, R=-0,269). Only 32,74% of women reported discussing sexual life with health care professionals. However, 67,25% of them would have had the need to be asked about their postpartum health issues. Conclusions and recommendations: The prevalence of female sexual dysfunctions were relatively high after childbirth. We found that incontinence and body image was associated with sexual dysfunctions; other risk factors remained unknown. Despite regular contact with health care professionals, women rarely get any information about postpartum sexual health issues. The high prevalence of dysfunctions indicates the need for further investigation to address other risk factors and proper counselling of women after childbirth.

Keywords: body image, postpartum, sexual dysfunction, urinary incontinence

Procedia PDF Downloads 111

Authors: Navab Karimi, Tohid Alizadeh

Abstract:

An intelligent vision-based system was designed to measure the quality and purity of raisins. A machine vision setup was utilized to capture the images of bulk raisins in ranges of 5-50% mixed pure-impure berries. The textural features of bulk raisins were extracted using Grey-level Histograms, Co-occurrence Matrix, and Local Binary Pattern (a total of 108 features). Genetic Algorithm and neural network regression were used for selecting and ranking the best features (21 features). As a result, the GLCM features set was found to have the highest accuracy (92.4%) among the other sets. Followingly, multiple feature combinations of the previous stage were fed into the second regression (linear regression) to increase accuracy, wherein a combination of 16 features was found to be the optimum. Finally, a Support Vector Machine (SVM) classifier was used to differentiate the mixtures, producing the best efficiency and accuracy of 96.2% and 97.35%, respectively.

Keywords: sun-dried organic raisin, genetic algorithm, feature extraction, ann regression, linear regression, support vector machine, south azerbaijan.

Procedia PDF Downloads 73

Authors: Haoyu Zhao, Qianshu Ma, Min Xie, Yunqi Huang, Yunjia Liu, Huan Song, Hongsheng Gui, Mingli Li, Qiang Wang

Abstract:

Rationale: Bipolar disorder (BD) and major depressive disorder (MDD), as severe chronic illnesses that restrict patients’ psychosocial functioning and reduce their quality of life, are both categorized into mood disorders. Emerging evidence has suggested that the reliability of self-rated health (SRH) was wellvalidated and that the risk of various health outcomes, including mortality and health care costs, could be predicted by SRH. Compared with other lengthy multi-item patient-reported outcomes (PRO) measures, SRH was proven to have a comparable predictive ability to predict mortality and healthcare utilization. However, to our knowledge, no study has been conducted to assess the association between SRH and hospitalization among people with mental disorders. Therefore, our study aims to determine the association between SRH and subsequent all-cause hospitalizations in patients with BD and MDD. Methods: We conducted a prospective cohort study on people with BD or MDD in the UK from 2006 to 2010 using UK Biobank touchscreen questionnaire data and linked administrative health databases. The association between SRH and 2-year all-cause hospitalizations was assessed using proportional hazard regression after adjustment for sociodemographics, lifestyle behaviors, previous hospitalization use, the Elixhauser comorbidity index, and environmental factors. Results: A total of 29,966 participants were identified, experiencing 10,279 hospitalization events. Among the cohort, the average age was 55.88 (SD 8.01) years, 64.02% were female, and 3,029 (10.11%), 15,972 (53.30%), 8,313 (27.74%), and 2,652 (8.85%) reported excellent, good, fair, and poor SRH, respectively. Among patients reporting poor SRH, 54.19% had a hospitalization event within 2 years compared with 22.65% for those having excellent SRH. In the adjusted analysis, patients with good, fair, and poor SRH had 1.31 (95% CI 1.21-1.42), 1.82 (95% CI 1.68-1.98), and 2.45 (95% CI 2.22, 2.70) higher hazards of hospitalization, respectively, than those with excellent SRH. Conclusion: SRH was independently associated with subsequent all-cause hospitalizations in patients with BD or MDD. This large study facilitates rapid interpretation of SRH values and underscores the need for proactive SRH screening in this population, which might inform resource allocation and enhance high-risk population detection.

Keywords: severe mental illnesses, hospitalization, risk prediction, patient-reported outcomes

Procedia PDF Downloads 161

Authors: Reham Algheshairy, Khaled Tayeb, Christopher Smith, Rebecca Gregg, Haruna Musa

Abstract:

Background: Diabetes is a medical condition that refers to the pancreas’ inability to secrete sufficient insulin levels, a hormone responsible for controlling glucose levels in the body. Any surplus glucose in the blood stream is excreted through the urinary system. Insulin resistance in blood cells can also cause this condition despite the fact that the pancreas is producing the required amount of insulin A number of researchers claim that the prevalence of diabetes in Saudi Arabia has reached epidemic proportions, although one study did observe one positive in the rise in the awareness of diabetes, possibly indicative of Saudi Arabia’s improving healthcare system. While a number of factors can cause diabetes, the ever-increasing incidence of the disease in Saudi Arabia has been blamed primarily on low levels of physical activity and high levels of obesity. Objectives: The project has two aims. The first aim of the project is to investigate the regulatory effects of consumption of Nigella seeds and Ajwah dates on blood glucose levels in diabetic patients with type 2 diabetes. The second aim of the project is to investigate whether these dietary factors may have potentially beneficial effects in controlling the complications that associated with type 2 diabetes. Methods: This use a random-cross intervention trail of 75 Saudi male and female with type 2 diabetes in Al-Noor hospital in Makkah ( KSA) aged between 18 and 70 years were divided into 3 groups. Group 1 will consume 2g of Nigella Sativa seeds daily along with a modified diet for 12 weeks, group 2 will be given Ajwah dates daily with a modified diet for 12 weeks and group 3 will follow a modified diet for 12 weeks. Anthropometric measurements were taken at baseline, along with bloods for HbA1c, fasting blood sugar and at the end of 12 weeks. Results: This study found significant decrease in blood level (FBG & 2PPBG) and HbA1c in the groups with diet and Nigella seeds) compared to Ajwa date. However, there is no significant change were found in HbA1c, FBG and 2hrpp regarding Ajwa group. Conclusion: This study illustrated a significant improvement in some markers of glycaemia following 2 g of Ns and diet for 12 weeks. The dose of 2g/day of consumed Nigella seeds was found to be more effective in controlling BGL and HbA1c than control and Ajwa groups. This suggests that Nigella seeds and following a diet may have a potential effect (a role in controlling outcomes for type 2 diabetes and controlling the disease). Further research is needed on a large scale to determine the optimum dose and duration of Nigella and Ajwa in order to achieve the desired results.

Keywords: type 2 diabetes, Nigella seeds, Ajwa dates, fasting blood glucose, control

Procedia PDF Downloads 295

Authors: Atefeh Bagherianziarat, Dana Hamplova

Abstract:

The inverse social gradient in life satisfaction (LS) is a well-established research finding. To estimate the influence of inequality on LS, most of the studies have explored the effect of the objective aspects of inequality or individuals’ socioeconomic status (SES). However, relatively fewer studies have confirmed recently the significant effect of the subjective aspect of inequality or subjective socioeconomic status (SSS) on life satisfaction over and above SES. In other words, it is confirmed by some studies that individuals’ perception of their unequal status in society or SSS can moderate the impact of their absolute unequal status on their life satisfaction. Nevertheless, this newly confirmed moderating link has not been affirmed to work likewise in societies with different levels of social inequality and also for people who believe in the value of equality, at different levels. In this study, we compared the moderative influence of subjective inequality on the link between objective inequality and life satisfaction. In particular, we focus on differences across welfare state regimes based on Esping-Andersen's theory. Also, we explored the moderative role of believing in the value of equality on the link between objective and subjective inequality on LS in the given societies. Since our studied variables were measured at both individual and country levels, we applied a multilevel analysis to the European Social Survey data (round 9). The results showed that people in deferent regimes reported statistically meaningful different levels of life satisfaction that is explained to different extends by their household income and their perception of their income inequality. The findings of the study supported the previous findings of the moderator influence of perceived inequality on the link between objective inequality and LS. However, this link is different in various welfare state regimes. The results of the multilevel modeling showed that country-level subjective equality is a positive predictor for individuals’ life satisfaction, while the GINI coefficient that was considered as the indicator of absolute inequality has a smaller effect on life satisfaction. Also, country-level subjective equality moderates the confirmed link between individuals’ income and their life satisfaction. It can be concluded that both individual and country-level subjective inequality slightly moderate the effect of individuals’ income on their life satisfaction.

Keywords: individual values, life satisfaction, multilevel analysis, objective inequality, subjective inequality, welfare regimes status

Procedia PDF Downloads 98

Authors: Azene Zenebe

Abstract:

Deep learning is a subset of machine learning which incorporates techniques for the construction of artificial neural networks and found to be useful for modeling complex problems with large dataset. Deep learning requires a very high power computational and longer time for training. By aggregating computing power, high performance computer (HPC) has emerged as an approach to resolving advanced problems and performing data-driven research activities. Cyber threat intelligence (CIT) is actionable information or insight an organization or individual uses to understand the threats that have, will, or are currently targeting the organization. Results of review of literature will be presented along with results of experimental study that compares the performance of tree-based and function-base machine learning including deep learning algorithms using secondary dataset collected from darknet.

Keywords: deep-learning, cyber security, cyber threat modeling, tree-based machine learning, function-based machine learning, data science

Procedia PDF Downloads 154

Authors: Chad Goldsworthy, B. Rajeswari Matam

Abstract:

The use of deep learning for species identification in camera trap images has revolutionised our ability to study, conserve and monitor species in a highly efficient and unobtrusive manner, with state-of-the-art models achieving accuracies surpassing the accuracy of manual human classification. The high imbalance of camera trap datasets, however, results in poor accuracies for minority (rare or endangered) species due to their relative insignificance to the overall model accuracy. This paper investigates the use of Focal Loss, in comparison to the traditional Cross Entropy Loss function, to improve the identification of minority species in the “255 Bird Species” dataset from Kaggle. The results show that, although Focal Loss slightly decreased the accuracy of the majority species, it was able to increase the F1-score by 0.06 and improve the identification of the bottom two, five and ten (minority) species by 37.5%, 15.7% and 10.8%, respectively, as well as resulting in an improved overall accuracy of 2.96%.

Keywords: convolutional neural networks, data imbalance, deep learning, focal loss, species classification, wildlife conservation

Procedia PDF Downloads 191

Authors: F. Moghtaderi, M. Burke, J. Troelsen

Abstract:

There has been a significant decline in active travel as well as the massive increase use of car-dependent travel mode in many countries during past two decades. Evidential risks for people’s physical and mental health problems are followed by this increased use of motorized travel mode. These problems range from overweight and obesity to increasing air pollution. In response to these rising concerns, local councils and other interested organizations around the world have introduced a variety of initiatives regarding reduce the dominance of cars for the daily journeys. However, the nature of these kinds of interventions, which related to the human behavior, make lots of complexities. People’s travel behavior and changing this behavior, has two different aspects. People’s attitudes and perceptions toward the sustainable and healthy modes of travel, and motorized travel modes (especially private car use) is one these two aspects. The other one related to people’s behavior change processes. There are no comprehensive model in order to guide policy interventions to increase the level of succeed of such interventions. A comprehensive theoretical framework is required in accordance to facilitate and guide the processes of data collection and analysis to achieve the best possible guidelines for policy makers. Regarding this gaps in the travel behavior change research, this paper attempted to identify and suggest a multidimensional framework in order to facilitate planning interventions. A structured mixed-method is suggested regarding the expand the scope and improve the analytic power of the result according to the complexity of human behavior. In order to recognize people’s attitudes, a theory with the focus on people’s attitudes towards a particular travel behavior was needed. The literature around the theory of planned behavior (TPB) was the most useful, and had been proven to be a good predictor of behavior change. Another aspect of the research, related to the people’s decision-making process regarding explore guidelines for the further interventions. Therefore, a theory was needed to facilitate and direct the interventions’ design. The concept of the transtheoretical model of behavior change (TTM) was used regarding reach a set of useful guidelines for the further interventions with the aim to increase active travel and sustainable modes of travel. Consequently, a combination of these two theories (TTM and TPB) had presented as an appropriate concept to identify and design implemented travel behavior change interventions.

Keywords: behavior change theories, theoretical framework, travel behavior change interventions, urban research

Procedia PDF Downloads 373

Authors: Tushar K. Routh

Abstract:

If inputs are engineered in certain manners, they can influence deep neural networks’ (DNN) performances by facilitating misclassifications, a phenomenon well-known as adversarial attacks that question networks’ vulnerability. Recent studies have unfolded the relationship between vulnerability of such networks with their complexity. In this paper, the distinctive influence of additional convolutional layers at the decision boundaries of several DNN architectures was investigated. Here, to engineer inputs from widely known image datasets like MNIST, Fashion MNIST, and Cifar 10, we have exercised One Step Spectral Attack (OSSA) and Fast Gradient Method (FGM) techniques. The aftermaths of adding layers to the robustness of the architectures have been analyzed. For reasoning, separation width from linear class partitions and local geometry (curvature) near the decision boundary have been examined. The result reveals that model complexity has significant roles in adjusting relative distances from margins, as well as the local features of decision boundaries, which impact robustness.

Keywords: DNN robustness, decision boundary, local curvature, network complexity

Procedia PDF Downloads 75

Authors: Sham Kidane, Fitsum Gaim, Ibrahim Abdella, Sirak Asmerom, Yoel Ghebrihiwot, Simon Mulugeta, Natnael Ambassager

Abstract:

The lack of annotated datasets is a bottleneck to the progress of NLP in low-resourced languages. The work presented here consists of large-scale annotated datasets and models for the named entity recognition (NER) system for the Tigrinya language. Our manually constructed corpus comprises over 340K words tagged for NER, with over 118K of the tokens also having parts-of-speech (POS) tags, annotated with 12 distinct classes of entities, represented using several types of tagging schemes. We conducted extensive experiments covering convolutional neural networks and transformer models; the highest performance achieved is 88.8% weighted F1-score. These results are especially noteworthy given the unique challenges posed by Tigrinya’s distinct grammatical structure and complex word morphologies. The system can be an essential building block for the advancement of NLP systems in Tigrinya and other related low-resourced languages and serve as a bridge for cross-referencing against higher-resourced languages.

Keywords: Tigrinya NER corpus, TiBERT, TiRoBERTa, BiLSTM-CRF

Procedia PDF Downloads 131

Authors: Boguslaw Schreyer

Abstract:

The paper investigates the application of the Kohonen’s Self-Organizing Map (SOM) to the wheeled robot starting and braking dynamic states. In securing wheeled robot stability as well as minimum starting and braking time, it is important to ensure correct torque distribution as well as proper slope of braking and driving moments. In this paper, a correct movement distribution has been formulated, securing optimum adhesion coefficient and good transversal stability of a wheeled robot. A neural tuner has been proposed to secure the above properties, although most of the attention is attached to the SOM network application. If the delay of the torque application or torque release is not negligible, it is important to change the rising and falling slopes of the torque. The road/surface condition is also paramount in robot dynamic states control. As the road conditions may randomly change in time, application of the SOM network has been suggested in order to classify the actual road conditions.

Keywords: slip control, SOM network, torque distribution, wheeled Robot

Procedia PDF Downloads 127

Authors: S. M. Gupta, Vanita Aggarwal, Som Nath Sachdeva

Abstract:

The work reported through this paper is an experimental work conducted on High Performance Concrete (HPC) with super plasticizer with the aim to develop some models suitable for prediction of compressive strength of HPC mixes. In this study, the effect of varying proportions of fly ash (0% to 50% at 10% increment) on compressive strength of high performance concrete has been evaluated. The mix designs studied were M30, M40 and M50 to compare the effect of fly ash addition on the properties of these concrete mixes. In all eighteen concrete mixes have been designed, three as conventional concretes for three grades under discussion and fifteen as HPC with fly ash with varying percentages of fly ash. The concrete mix designing has been done in accordance with Indian standard recommended guidelines i.e. IS: 10262. All the concrete mixes have been studied in terms of compressive strength at 7 days, 28 days, 90 days and 365 days. All the materials used have been kept same throughout the study to get a perfect comparison of values of results. The models for compressive strength prediction have been developed using Linear Regression method (LR), Artificial Neural Network (ANN) and Leave One Out Validation (LOOV) methods.

Keywords: high performance concrete, fly ash, concrete mixes, compressive strength, strength prediction models, linear regression, ANN

Procedia PDF Downloads 445

Authors: R. Biondi, G. Dys, G. Ferone, T. Renard, M. Zysman

Abstract:

This paper describes an automated implementable system for impulsive signals detection and recognition. The system uses a Digital Signal Processing device for the detection and identification process. Here the system analyses the signals in real time in order to produce a particular response if needed. The system analyses the signals in real time in order to produce a specific output if needed. Detection is achieved through normalizing the inputs and comparing the read signals to a dynamic threshold and thus avoiding detections linked to loud or fluctuating environing noise. Identification is done through neuronal network algorithms. As a setup our system can receive signals to “learn” certain patterns. Through “learning” the system can recognize signals faster, inducing flexibility to new patterns similar to those known. Sound is captured through a simple jack input, and could be changed for an enhanced recording surface such as a wide-area recorder. Furthermore a communication module can be added to the apparatus to send alerts to another interface if needed.

Keywords: sound detection, impulsive signal, background noise, neural network

Procedia PDF Downloads 320

Authors: Theresa Veach, Erin Crisp

Abstract:

It has been shown that traditional students flourish as a function of several factors including level of academic challenge, student/faculty interactions, active/collaborative learning, enriching educational experiences, and supportive campus environment. With the increase in demand for remote or online courses, factors that result in academic flourishing in the virtual classroom have become more crucial to understand than ever before. This study seeks to give insight into those factors that impact student learning, overall student wellbeing, and flourishing among college students enrolled in an online program. 4160 unique students participated in the completion of End of Course Survey (EOC) before final grades were released. Quantitative results from the survey are used by program directors as a measure of student satisfaction with both the curriculum and the faculty. In addition, students also submitted narrative comments in an open comment field. No prompts were given for the comment field on the survey. The purpose of this analysis was to report on the qualitative data available with the goal of gaining insight into what matters to students. Survey results from July 1st, 2016 to December 1st, 2016 were compiled into spreadsheet data sets. The analysis approach used involved both key word and phrase searches and reading results to identify patterns in responses and to tally the frequency of those patterns. In total, just over 25,000 comments were included in the analysis. Preliminary results indicate that it is the professor-student relationship, frequency of feedback and overall engagement of both instructors and students that are indicators of flourishing in college programs offered in an online format. This qualitative study supports the notion that college students flourish with regard to 1) education, 2) overall student well-being and 3) program satisfaction when overall engagement of both the instructor and the student is high. Ways to increase engagement in the online college environment were also explored. These include 1) increasing student participation by providing more project-based assignments, 2) interacting with students in meaningful ways that are both high in frequency and in personal content, and 3) allowing students to apply newly acquired knowledge in ways that are meaningful to current life circumstances and future goals.

Keywords: college, engagement, flourishing, online

Procedia PDF Downloads 271

Authors: Kavyadevi N., Kaviya G., Gowsalya P., Janani M., Mohanraj S.

Abstract:

Hevea brasiliensis, also known as the rubber tree, is one of the foremost assets of crops in the world. One of the most significant advantages of the Rubber Plant in terms of air oxygenation is its capacity to reduce the likelihood of an individual developing respiratory allergies like asthma. To construct such a system that can properly identify crop diseases and pests and then create a database of insecticides for each pest and disease, we must first give treatment for the illness that has been detected. We shall primarily examine three major leaf diseases since they are economically deficient in this article, which is Bird's eye spot, algal spot and powdery mildew. And the recommended work focuses on disease identification on rubber tree leaves. It will be accomplished by employing one of the superior algorithms. Input, Preprocessing, Image Segmentation, Extraction Feature, and Classification will be followed by the processing technique. We will use time-consuming procedures that they use to detect the sickness. As a consequence, the main ailments, underlying causes, and signs and symptoms of diseases that harm the rubber tree are covered in this study.

Keywords: image processing, python, convolution neural network (CNN), machine learning

Procedia PDF Downloads 76

Authors: Mohd Zamri Osman, Mohd Aizaini Maarof, Mohd Foad Rohani

Abstract:

Human skin detection recognized as the primary step in most of the applications such as face detection, illicit image filtering, hand recognition and video surveillance. The performance of any skin detection applications greatly relies on the two components: feature extraction and classification method. Skin color is the most vital information used for skin detection purpose. However, color feature alone sometimes could not handle images with having same color distribution with skin color. A color feature of pixel-based does not eliminate the skin-like color due to the intensity of skin and skin-like color fall under the same distribution. Hence, the statistical color analysis will be exploited such mean and standard deviation as an additional feature to increase the reliability of skin detector. In this paper, we studied the effectiveness of statistical color feature for human skin detection. Furthermore, the paper analyzed the integrated color and texture using eight classifiers with three color spaces of RGB, YCbCr, and HSV. The experimental results show that the integrating statistical feature using Random Forest classifier achieved a significant performance with an F1-score 0.969.

Keywords: color space, neural network, random forest, skin detection, statistical feature

Procedia PDF Downloads 462

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: Alireza Shams, Ali Zamanian, Atefehe Shamosi, Farnaz Ghorbani

Abstract:

Introduction: Although the application of a new strategy remains a remarkable challenge for treatment of disabilities due to neuronal defects, progress in Nanomedicine and tissue engineering, suggesting the new medical methods. One of the promising strategies for reconstruction and regeneration of nervous tissue is replacing of lost or damaged cells by specific scaffolds after Compressive, ischemic and traumatic injuries of central nervous system. Furthermore, ultrastructure, composition, and arrangement of tissue scaffolds are effective on cell grafts. We followed implantation and differentiation of mesenchyme stem cells to neural cells on Gelatin Polylactic-co-glycolic acid (PLGA) scaffolds coated with iron nanoparticles. The aim of this study was to evaluate the capability of stem cells to differentiate into motor neuron-like cells under topographical cues and morphogenic factors. Methods and Materials: Bone marrow mesenchymal stem cells (BMMSCs) was obtained by primary cell culturing of adult rat bone marrow got from femur bone by flushing method. BMMSCs were incubated with DMEM/F12 (Gibco), 15% FBS and 100 U/ml pen/strep as media. Then, BMMSCs seeded on Gel/PLGA scaffolds and tissue culture (TCP) polystyrene embedded and incorporated by Fe Nano particles (FeNPs) (Fe3o4 oxide (M w= 270.30 gr/mol.). For neuronal differentiation, 2×10 5 BMMSCs were seeded on Gel/PLGA/FeNPs scaffolds was cultured for 7 days and 0.5 µ mol. Retinoic acid, 100 µ mol. Ascorbic acid,10 ng/ml. Basic fibroblast growth factor (Sigma, USA), 250 μM Iso butyl methyl xanthine, 100 μM 2-mercaptoethanol, and 0.2 % B27 (Invitrogen, USA) added to media. Proliferation of BMMSCs was assessed by using MTT assay for cell survival. The morphology of BMMSCs and scaffolds was investigated by scanning electron microscopy analysis. Expression of neuron-specific markers was studied by immunohistochemistry method. Data were analyzed by analysis of variance, and statistical significance was determined by Turkey’s test. Results: Our results revealed that differentiation and survival of BMMSCs into motor neuron-like cells on Gel/PLGA/FeNPs as a biocompatible and biodegradable scaffolds were better than those cultured in Gel/PLGA in absence of FeNPs and TCP scaffolds. FeNPs had raised physical power but decreased capacity absorption of scaffolds. Well defined oriented pores in scaffolds due to FeNPs may activate differentiation and synchronized cells as a mechanoreceptor. Induction effects of magnetic FeNPs by One way flow of channels in scaffolds help to lead the cells and can facilitate direction of their growth processes. Discussion: Progression of biological properties of BMMSCs and the effects of FeNPs spreading under magnetic field was evaluated in this investigation. In vitro study showed that the Gel/PLGA/FeNPs scaffold provided a suitable structure for motor neuron-like cells differentiation. This could be a promising candidate for enhancing repair and regeneration in neural defects. Dynamic and static magnetic field for inducing and construction of cells can provide better results for further experimental studies.

Keywords: differentiation, mesenchymal stem cells, nano particles, neuronal defects, Scaffolds

Procedia PDF Downloads 166

Authors: Azeddine El-Hassouny, Chandrashekhar Meshram, Geraldin Nanfack

Abstract:

In recent years, learning from data with noisy labels (Label Noise) has been a major concern in supervised learning. This problem has become even more worrying in Deep Learning, where the generalization capabilities have been questioned lately. Indeed, deep learning requires a large amount of data that is generally collected by search engines, which frequently return data with unreliable labels. In this paper, we investigate the Label Noise in Deep Learning using variational inference. Our contributions are : (1) exploiting Label Noise concept where the true labels are learnt using reparameterization variational inference, while observed labels are learnt discriminatively. (2) the noise transition matrix is learnt during the training without any particular process, neither heuristic nor preliminary phases. The theoretical results shows how true label distribution can be learned by variational inference in any discriminate neural network, and the effectiveness of our approach is proved in several target datasets, such as MNIST and CIFAR32.

Keywords: label noise, deep learning, discrete latent variable, variational inference, MNIST, CIFAR32

Procedia PDF Downloads 128