Search results for: randomization inference and sun

Authors: Arian Hosseini, Mahmudul Hasan

Abstract:

To address the increasing need for efficient and accurate content moderation, we propose an efficient and lightweight deep classification ensemble structure. Our approach is based on a combination of simple visual features, designed for high-accuracy classification of violent content with low false positives. Our ensemble architecture utilizes a set of lightweight models with narrowed-down color features, and we apply it to both images and videos. We evaluated our approach using a large dataset of explosion and blast contents and compared its performance to popular deep learning models such as ResNet-50. Our evaluation results demonstrate significant improvements in prediction accuracy, while benefiting from 7.64x faster inference and lower computation cost. While our approach is tailored to explosion detection, it can be applied to other similar content moderation and violence detection use cases as well. Based on our experiments, we propose a "think small, think many" philosophy in classification scenarios. We argue that transforming a single, large, monolithic deep model into a verification-based step model ensemble of multiple small, simple, and lightweight models with narrowed-down visual features can possibly lead to predictions with higher accuracy.

Keywords: deep classification, content moderation, ensemble learning, explosion detection, video processing

Procedia PDF Downloads 22

Authors: Kong Ngai Pei

Abstract:

The argument scheme approach to argumentation has two components. One is to identify the recurrent patterns of inferences used in everyday discourse. The second is to devise critical questions to evaluate the inferences in these patterns. Although this approach is intuitive and contains many insightful ideas, it has been noted to be not free of problems. One is that due to its disavowing the probability calculus, it cannot give the exact strength of an inference. In order to tackle this problem, thereby paving the way to a more complete normative account of argument strength, it has been proposed, the most promising way is to combine the scheme-based approach with Bayesian networks (BNs). This paper pursues this line of thought, attempting to combine three common schemes, Appeal to Ignorance, Composition, and Division, with BNs. In the first part, it is argued that most (if not all) formulations of the critical questions corresponding to these schemes in the current argumentation literature are incomplete and not very informative. To remedy these flaws, more thorough and precise formulations of these questions are provided. In the second part, how to use graphical idioms (e.g. measurement and synthesis idioms) to translate the schemes as well as their corresponding critical questions to graphical structure of BNs, and how to define probability tables of the nodes using functions of various sorts are shown. In the final part, it is argued that many misuses of these schemes, traditionally called fallacies with the same names as the schemes, can indeed be adequately accounted for by the BN models proposed in this paper.

Keywords: appeal to ignorance, argument schemes, Bayesian networks, composition, division

Procedia PDF Downloads 254

Authors: Roseline Oghogho Osaseri, Osaseri E. I.

Abstract:

Lassa fever is an epidemic hemorrhagic fever caused by the Lassa virus, an extremely virulent arena virus. This highly fatal disorder kills 10% to 50% of its victims, but those who survive its early stages usually recover and acquire immunity to secondary attacks. One of the major challenges in giving proper treatment is lack of fast and accurate diagnosis of the disease due to multiplicity of symptoms associated with the disease which could be similar to other clinical conditions and makes it difficult to diagnose early. This paper proposed an Adaptive Neuro Fuzzy Inference System (ANFIS) for the prediction of Lass Fever. In the design of the diagnostic system, four main attributes were considered as the input parameters and one output parameter for the system. The input parameters are Temperature on admission (TA), White Blood Count (WBC), Proteinuria (P) and Abdominal Pain (AP). Sixty-one percent of the datasets were used in training the system while fifty-nine used in testing. Experimental results from this study gave a reliable and accurate prediction of Lassa fever when compared with clinically confirmed cases. In this study, we have proposed Lassa fever diagnostic system to aid surgeons and medical healthcare practictionals in health care facilities who do not have ready access to Polymerase Chain Reaction (PCR) diagnosis to predict possible Lassa fever infection.

Keywords: anfis, lassa fever, medical diagnosis, soft computing

Procedia PDF Downloads 237

Authors: Swathi M. Vanniarajan

Abstract:

Reading research suggests that online contextual vocabulary comprehension while reading is an interactive and integrative process. One’s success in it depends on a variety of factors including the amount and the nature of available linguistic and nonlinguistic cues, his/her analytical and integrative skills, schema memory (content familiarity), and processing speed characterized along the continuum of controlled to automatic processing. The experiment reported here, conducted with 30 native speakers as one group and 30 nonnative speakers as another group (all graduate students), hypothesized that while working on (24) tasks which required them to comprehend an unfamiliar word in real time without backtracking, due to the differences in the nature of their respective reading processes, the nonnative subjects would be less able to construct the meanings of the unknown words by integrating the multiple but sufficient contextual cues provided in the text but the native subjects would be able to. The results indicated that there were significant inter-group as well as intra-group differences in terms of the quality of definitions given. However, when given additional time, while the nonnative speakers could significantly improve the quality of their definitions, the native speakers in general would not, suggesting that all things being equal, time is a significant factor for success in nonnative vocabulary and reading comprehension processes and that accuracy precedes automaticity in the development of nonnative reading processes also.

Keywords: reading, second language processing, vocabulary comprehension

Procedia PDF Downloads 141

Authors: Johnson C. Y. Pang, Bo Pengb, Kara K. L. Reevesc, Allan C. L. Fud

Abstract:

Objective: Dry needling (DN) has long been used as a treatment method for various musculoskeletal pain conditions. However, the evidence level of the studies was low due to the limitations of the methodology. Lack of randomization and inappropriate blinding are potentially the main sources of bias. A method that can differentiate clinical results due to the targeted experimental procedure from its placebo effect is needed to enhance the validity of the trial. Therefore, this study aimed to validate the method as a placebo ultrasound(US)-guided DN for patients with knee osteoarthritis (KOA). Design: This is a randomized controlled trial (RCT). Ninety subjects (25 males and 65 females) aged between 51 and 80 (61.26±5.57) with radiological KOA were recruited and randomly assigned into three groups with a computer program. Group 1 (G1) received real US-guided DN, Group 2 (G2) received placebo US-guided DN, and Group 3 (G3) was the control group. Both G1 and G2 subjects received the same procedure of US-guided DN, except the US monitor was turned off in G2, blinding the G2 subjects to the incorporation of faux US guidance. This arrangement created the placebo effect intended to permit comparison of their results to those who received actual US-guided DN. Outcome measures, including the visual analog scale (VAS) and Knee injury and Osteoarthritis Outcome Score (KOOS) subscales of pain, symptoms and quality of life (QOL), were analyzed by repeated-measures analysis of covariance (ANCOVA) for time effects and group effects. The data regarding the perception of receiving real US-guided DN or placebo US-guided DN were analyzed by the chi-squared test. The missing data were analyzed with the intention-to-treat (ITT) approach if more than 5% of the data were missing. Results: The placebo US-guided DN (G2) subjects had the same perceptions as the use of real US guidance in the advancement of DN (p<0.128). G1 had significantly higher pain reduction (VAS and KOOS-pain) than G2 and G3 at 8 weeks (both p<0.05) only. There was no significant difference between G2 and G3 at 8 weeks (both p>0.05). Conclusion: The method with the US monitor turned off during the application of DN is credible for blinding the participants and allowing researchers to incorporate faux US guidance. The validated placebo US-guided DN technique can aid in investigations of the effects of US-guided DN with short-term effects of pain reduction for patients with KOA. Acknowledgment: This work was supported by the Caritas Institute of Higher Education [grant number IDG200101].

Keywords: reliability, jumping, 3D motion analysis, anterior crucial ligament reconstruction

Procedia PDF Downloads 90

Authors: Ofosuhene O. Apenteng, Noor Azina Ismail

Abstract:

Over the last several years, concern has developed over how to minimize the spread of HIV/AIDS epidemic in many countries. AIDS epidemic has tremendously stimulated the development of mathematical models of infectious diseases. The transmission dynamics of HIV infection that eventually developed AIDS has taken a pivotal role of much on building mathematical models. From the initial HIV and AIDS models introduced in the 80s, various improvements have been taken into account as how to model HIV/AIDS frameworks. In this paper, we present the impact of migration on the spread of HIV/AIDS. Epidemic model is considered by a system of nonlinear differential equations to supplement the statistical method approach. The model is calibrated using HIV incidence data from Malaysia between 1986 and 2011. Bayesian inference based on Markov Chain Monte Carlo is used to validate the model by fitting it to the data and to estimate the unknown parameters for the model. The results suggest that the migrants stay for a long time contributes to the spread of HIV. The model also indicates that susceptible individual becomes infected and moved to HIV compartment at a rate that is more significant than the removal rate from HIV compartment to AIDS compartment. The disease-free steady state is unstable since the basic reproduction number is 1.627309. This is a big concern and not a good indicator from the public heath point of view since the aim is to stabilize the epidemic at the disease equilibrium.

Keywords: epidemic model, HIV, MCMC, parameter estimation

Procedia PDF Downloads 568

Authors: Najrullah Khan, Athar Ali Khan

Abstract:

The Topp-Leone distribution was introduced by Topp- Leone in 1955. In this paper, an attempt has been made to fit Topp-Leone Generalized exponential (TPGE) distribution. A real survival data set is used for illustrations. Implementation is done using R and JAGS and appropriate illustrations are made. R and JAGS codes have been provided to implement censoring mechanism using both optimization and simulation tools. The main aim of this paper is to describe and illustrate the Bayesian modelling approach to the analysis of survival data. Emphasis is placed on the modeling of data and the interpretation of the results. Crucial to this is an understanding of the nature of the incomplete or 'censored' data encountered. Analytic approximation and simulation tools are covered here, but most of the emphasis is on Markov chain based Monte Carlo method including independent Metropolis algorithm, which is currently the most popular technique. For analytic approximation, among various optimization algorithms and trust region method is found to be the best. In this paper, TPGE model is also used to analyze the lifetime data in Bayesian paradigm. Results are evaluated from the above mentioned real survival data set. The analytic approximation and simulation methods are implemented using some software packages. It is clear from our findings that simulation tools provide better results as compared to those obtained by asymptotic approximation.

Keywords: Bayesian Inference, JAGS, Laplace Approximation, LaplacesDemon, posterior, R Software, simulation

Procedia PDF Downloads 502

Authors: Jeewantha Ranawaka, Gunawardane Kapila, Wijethunaga Mudiyanselage B. G. Jayathilake

Abstract:

Introduction: Sweeping of the membranes is a fairly simple technique that may positively influence the shift from maintenance of pregnancy to the beginning of labor. Objective: To assess the effectiveness and acceptability of twice versus once-only membrane sweeping in uncomplicated primi gravid at 40 weeks of gestational age in a tertiary care hospital in Sri Lanka. Methods: A randomized controlled clinical trial was done in Ward 05 of Teaching Hospital, Kandy. The participants were primi-gravida with a singleton live fetus who was at 40 weeks of gestation with intact fetal membranes and with a Modified Bishop’s score <5. After randomization both groups received membrane sweeping at 40 weeks of gestation and the experimental group received membrane sweeping after 48 hours (40+2 days). The modified Bishop Score was assessed at 40+5 days. In two groups who did not go into natural labor at 40+5 days were managed according to the ward policy of cervical ripening and with labor induction at 40+5 days. Two different methods were used to assess discomfort and pain. Patient acceptability was assessed using recommendation to another patient and acceptance during next pregnancy. Perinatal, maternal and labour outcomes were assessed. Results: A change of the Bishops score was 67.3% (n= 31 of 46) in experimental group whereas in control group it was 57.5% (n= 38 of 66). (p = 0.21, OR-1.52, CI = 0.6 -3.34). Mean (SD) of Modified Bishop score was 6.36 (1.94) in experimental group and 6.03 (.84) in control group (p = 0.354). The probability of having the spontaneous onset of labour in experimental group was 61.6% (n=74 of 120) whereas in control group it was 45% (n= 54 of 120) (p=0.01, OR-1.966, CI = 1.17 – 3.28 NNT = 5.99). Recommending the method to another among experimental group was 75% (n= 90 of 120) whereas in control group it was 79.2% (n= 95 of 120) (p= 0.443). Accepting membrane Sweeping for subsequent pregnancy among experimental was 72.5% (n=87 of 120) whereas in control group was 72.5% (n=87 of 120) (p= 1.00) Need of formal induction of labour at 40+ 5 days in experimental group was 38.4% (n=46 of 120) whereas in control group was 61.6% (n=66 of 120) (p=0.01, OR=0.5, CI= 0.3 – 0.8, NNT=6). Neonatal outcome, labour outcome such as Cesarean -section rate, need for augmentation and maternal complications such as fever, Premature rupture of membrane, bleeding were comparable in two groups. Conclusions and Recommendations: It can be concluded that twice sweeping of membrane was effective to reduce the need of formal induction of labour and increase the chances of having spontaneous onset of labour (SOL) at 40+5 days without increasing maternal or fetal morbidity. Acceptability of twice sweeping is not different from sweeping once. Hence we recommend consideration of multiple membranes sweeping as first line for women at 40 weeks of gestation.

Keywords: acceptability, induction, labour, membrane sweeping

Procedia PDF Downloads 273

Authors: Arkadiusz Urzędowski, Dorota Wójcicka-Migasiuk, Andrzej Sachajdak, Magdalena Paśnikowska-Łukaszuk

Abstract:

Due to the impact of climate changes and inevitability to reduce greenhouse gases, the need to use low-carbon and sustainable construction has increased. In this work, it is investigated how texture of the surface building materials and radiative heat transfer phenomenon in flat multilayer can be correlated. Attempts to test the surface emissivity are taken however, the trustworthiness of measurement results remains a concern since sensor size and thickness are common problems. This paper presents an experimental method to studies surface emissivity with use self constructed thermal sensors and thermal imaging technique. The surface of building materials was modified by mechanical and chemical treatment affecting the reduction of the emissivity. For testing the shaping surface of materials and mapping its three-dimensional structure, scanning profilometry were used in a laboratory. By comparing the results of laboratory tests and performed analysis of 3D computer fluid dynamics software, it can be shown that a change in the surface coverage of materials affects the heat transport by radiation between layers. Motivated by recent advancements in variational inference, this publication evaluates the potential use a dedicated data processing approach, and properly constructed temperature sensors, the influence of the surface emissivity on the phenomenon of radiation and heat transport in the entire partition can be determined.

Keywords: heat transfer, surface roughness, surface emissivity, radiation

Procedia PDF Downloads 66

Authors: Xinyuan Y. Song, Kai Kang

Abstract:

Joint models for analyzing longitudinal and survival data are widely used to investigate the relationship between a failure time process and time-variant predictors. A common assumption in conventional joint models in the survival analysis literature is that all predictors are observable. However, this assumption may not always be supported because unobservable traits, namely, latent variables, which are indirectly observable and should be measured through multiple observed variables, are commonly encountered in the medical, behavioral, and financial research settings. In this study, a joint modeling approach to deal with this feature is proposed. The proposed model comprises three parts. The first part is a dynamic factor analysis model for characterizing latent variables through multiple observed indicators over time. The second part is a random coefficient trajectory model for describing the individual trajectories of latent variables. The third part is a proportional hazard model for examining the effects of time-invariant predictors and the longitudinal trajectories of time-variant latent risk factors on hazards of interest. A Bayesian approach coupled with a Markov chain Monte Carlo algorithm to perform statistical inference. An application of the proposed joint model to a study on the Alzheimer's disease neuroimaging Initiative is presented.

Keywords: Bayesian analysis, joint model, longitudinal data, time-to-event data

Procedia PDF Downloads 115

Authors: Endeshaw Assefa Derso, Maria Gabriella Campolo, Angela Alibrandi

Abstract:

Objectives:Early childhood malnutrition can have long-term and irreversible effects on a child's health and development. This study uses the Bayesian method with spatial variation to investigate the flexible trends of metrical covariates and to identify communities at high risk of injury. Methods: Cross-sectional data on underweight are collected from the 2016 Ethiopian Demographic and Health Survey (EDHS). The Bayesian geo-additive model is performed. Appropriate prior distributions were provided for scall parameters in the models, and the inference is entirely Bayesian, using Monte Carlo Markov chain (MCMC) stimulation. Results: The results show that metrical covariates like child age, maternal body mass index (BMI), and maternal age affect a child's underweight non-linearly. Lower and higher maternal BMI seem to have a significant impact on the child’s high underweight. There was also a significant spatial heterogeneity, and based on IDW interpolation of predictive values, the western, central, and eastern parts of the country are hotspot areas. Conclusion: Socio-demographic and community- based programs development should be considered compressively in Ethiopian policy to combat childhood underweight malnutrition.

Keywords: bayesX, Ethiopia, malnutrition, MCMC, semi-parametric bayesian analysis, spatial distribution, P- splines

Procedia PDF Downloads 51

Authors: Sadam Elayah

Abstract:

Background: One of the most critical competencies in advanced dentistry is alveolar ridge preservation after exodontia. The aim of this clinical trial was to assess the impact of autologous concentrated growth factor (CGF) as a socket-filling material and its ridge preservation properties following the lower third molar extraction. Materials and Methods: A total of 60 sides of 30 participants who had completely symmetrical bilateral impacted lower third molars were enrolled. The short-term outcome variables were wound healing, swelling and pain, clinically assessed at different time intervals (1st, 3rd & 7th days). While the long-term outcome variables were bone height & width, bone density and socket surface area in the coronal section. Cone beam computed tomography images were obtained immediately after surgery and three months after surgery as a temporal measure. Randomization was achieved by opaque, sealed envelopes. Follow-up data were compared to baseline using Paired & Unpaired t-tests. Results: The wound healing index was significantly better in the test sides (P =0.001). Regarding the facial swelling, the test sides had significantly fewer values than the control sides, particularly on the 1st (1.01±.57 vs 1.55 ±.56) and 3rd days (1.42±0.8 vs 2.63±1.2) postoperatively. Nonetheless, the swelling disappeared within the 7th day on both sides. The pain scores of the visual analog scale were not a statistically significant difference between both sides on the 1st day; meanwhile, the pain scores were significantly lower on the test sides compared with the control sides, especially on the 3rd (P=0.001) and 7th days (P˂0.001) postoperatively. Regarding long-term outcomes, CGF sites had higher values in height and width when compared to Control sites (Buccal wall 32.9±3.5 vs 29.4±4.3 mm, Lingual wall 25.4±3.5 vs 23.1±4 mm, and Alveolar bone width 21.07±1.55vs19.53±1.90 mm) respectively. Bone density showed significantly higher values in CGF sites than in control sites (Coronal half 200±127.3 vs -84.1±121.3, Apical half 406.5±103 vs 64.2±158.6) respectively. There was a significant difference between both sites in reducing periodontal pockets. Conclusion: CGF application following surgical extraction provides an easy, low-cost, and efficient option for alveolar ridge preservation. Thus, dentists may encourage using CGF during dental extractions, particularly when alveolar ridge preservation is required.

Keywords: platelet, extraction, impacted teeth, alveolar ridge, regeneration, CGF

Procedia PDF Downloads 45

Authors: Sonam Kelzang

Abstract:

Background: Penetrating injuries as a result of traditional sports (Archery and Khuru) are commonly seen in Bhutan. To our knowledge, there is no study carried out looking into the data of penetrating injuries due to traditional sports. Aim: This is a retrospective analysis of cases of penetrating injuries as a result of traditional sports admitted to JDWNRH over the last 3 years to draw an inference on the pattern of injury and associated morbidity and mortality. Method: Data on penetrating injuries related to traditional sports (Archery and Khuru) were collected and reviewed over the period of 3 years. Assault cases were excluded. For each year we analysed age, sex, parts of the body affected, agent of injury and whether admission was required or not. Results: Out of the total 44 victims of penetrating injury by traditional sports (Archery and Khuru) between 2013 and 2015 (average of 15 cases of penetrating injuries per year). Eighty-five percent were male and 15% were female. Their age ranged from 4 yrs to 62 years. Sixty-one percent of the victims were in the working age group of 19-58 years; 30% of the victims were referred from various district hospitals; 38% of the victims needed admission; 42 % of the victims suffered injury to the head; and 54% of the injuries were caused by Khuru. Conclusion: Penetrating injuries due to traditional sports admitted to JDWNRH, Thimphu, remained same over the three years period despite safety regulations in place. Although there were no deaths during the last three years, morbidity still remains high.

Keywords: archery, Bhutan, Khuru, darts

Procedia PDF Downloads 141

Authors: Sofia M. Karadimitriou, Kostas Triantafyllopoulos, Timothy Heaton

Abstract:

Reduced dimension Dynamic Spatio-Temporal Models (DSTMs) jointly describe the spatial and temporal evolution of a function observed subject to noise. A basic state space model is adopted for the discrete temporal variation, while a continuous autoregressive structure describes the continuous spatial evolution. Application of such a DSTM relies upon the pre-selection of a suitable reduced set of basic functions and this can present a challenge in practice. In this talk, we propose an online estimation method for high dimensional spatio-temporal data based upon DSTM and we attempt to resolve this issue by allowing the basis to adapt to the observed data. Specifically, we present a wavelet decomposition in order to obtain a parsimonious approximation of the spatial continuous process. This parsimony can be achieved by placing a Laplace prior distribution on the wavelet coefficients. The aim of using the Laplace prior, is to filter wavelet coefficients with low contribution, and thus achieve the dimension reduction with significant computation savings. We then propose a Hierarchical Bayesian State Space model, for the estimation of which we offer an appropriate particle filter algorithm. The proposed methodology is illustrated using real environmental data.

Keywords: multidimensional Laplace prior, particle filtering, spatio-temporal modelling, wavelets

Procedia PDF Downloads 402

Authors: Ms Azam Najafkouchak, David Todem, Dorothy Pathak, Pramod Pathak, Joseph Gardiner

Abstract:

Propensity score (PS) methods have recently become the standard analysis as a tool for the causal inference in the observational studies where exposure is not randomly assigned, thus, confounding can impact the estimation of treatment effect on the outcome. For the binary outcome, the effect of treatment on the outcome can be estimated by odds ratios, relative risks, and risk differences. However, using the different PS methods may give you a different estimation of the treatment effect on the outcome. Several methods of PS analyses have been used mainly, include matching, inverse probability of weighting, stratification, and covariate adjusted on PS. Due to the dangers of discretizing continuous variables (exposure, covariates), the focus of this paper will be on how the variation in cut-points or boundaries will affect the average treatment effect (ATE) utilizing the stratification of PS method. Therefore, we are trying to avoid choosing arbitrary cut-points, instead, we continuously discretize the PS and accumulate information across all cut-points for inferences. We will use Monte Carlo simulation to evaluate ATE, focusing on two PS methods, stratification and covariate adjusted on PS. We will then show how this can be observed based on the analyses of the data from a case-control study of breast cancer, the Polish Women’s Health Study.

Keywords: average treatment effect, propensity score, stratification, covariate adjusted, monte Calro estimation, breast cancer, case_control study

Procedia PDF Downloads 80

Authors: Isha Saxena, Behzad Kazemtabrizi, Matthias C. M. Troffaes, Christopher Crabtree

Abstract:

The calculations done at the beginning of the life of a wind farm are rarely reliable, which makes it important to conduct research and study the failure and repair rates of the wind turbines under various conditions. This miscalculation happens because the current models make a simplifying assumption that the failure/repair rate remains constant over time. This means that the reliability function is exponential in nature. This research aims to create a more accurate model using sensory data and a data-driven approach. The data cleaning and data processing is done by comparing the Power Curve data of the wind turbines with SCADA data. This is then converted to times to repair and times to failure timeseries data. Several different mathematical functions are fitted to the times to failure and times to repair data of the wind turbine components using Maximum Likelihood Estimation and the Posterior expectation method for Bayesian Parameter Estimation. Initial results indicate that two parameter Weibull function and exponential function produce almost identical results. Further analysis is being done using the complex system analysis considering the failures of each electrical and mechanical component of the wind turbine. The aim of this project is to perform a more accurate reliability analysis that can be helpful for the engineers to schedule maintenance and repairs to decrease the downtime of the turbine.

Keywords: reliability, bayesian parameter inference, maximum likelihood estimation, weibull function, SCADA data

Procedia PDF Downloads 36

Authors: Krishna Mohan Bathula, Fatou Bintou Loucoubar, FNU Kaleemunnisa, Christelle Scharff, Mark Anthony De Castro

Abstract:

Voice recognition algorithms such as automatic speech recognition and text-to-speech systems with African languages can play an important role in bridging the digital divide of Artificial Intelligence in Africa, contributing to the establishment of a fully inclusive information society. This paper proposes a Deep Learning model that can classify the user responses as inputs for an interactive voice response system. A dataset with Wolof language words ‘yes’ and ‘no’ is collected as audio recordings. A two stage Data Augmentation approach is adopted for enhancing the dataset size required by the deep neural network. Data preprocessing and feature engineering with Mel-Frequency Cepstral Coefficients are implemented. Convolutional Neural Networks (CNNs) have proven to be very powerful in image classification and are promising for audio processing when sounds are transformed into spectra. For performing voice response classification, the recordings are transformed into sound frequency feature spectra and then applied image classification methodology using a deep CNN model. The inference model of this trained and reusable Wolof voice response recognition system can be integrated with many applications associated with both web and mobile platforms.

Keywords: automatic speech recognition, interactive voice response, voice response recognition, wolof word classification

Procedia PDF Downloads 87

Authors: Ming-Yang Guo, Cheng-Xian Wu, Wei-Xiang Chen, Chun-Yuan Lin, Yen-Jen Lin, Ann-Shyn Chiang

Abstract:

With more and more Drosophila Driver and Neuron images, it is an important work to find the similarity relationships among them as the functional inference. There is a general problem that how to find a Drosophila Driver image, which can cover a set of Drosophila Driver/Neuron images. In order to solve this problem, the intersection/union region for a set of images should be computed at first, then a comparison work is used to calculate the similarities between the region and other images. In this paper, three encoding schemes, namely Integer, Boolean, Decimal, are proposed to encode each image as a one-dimensional structure. Then, the intersection/union region from these images can be computed by using the compare operations, Boolean operators and lookup table method. Finally, the comparison work is done as the union region computation, and the similarity score can be calculated by the definition of Tanimoto coefficient. The above methods for the region computation are also implemented in the multi-core CPUs environment with the OpenMP. From the experimental results, in the encoding phase, the performance by the Boolean scheme is the best than that by others; in the region computation phase, the performance by Decimal is the best when the number of images is large. The speedup ratio can achieve 12 based on 16 CPUs. This work was supported by the Ministry of Science and Technology under the grant MOST 106-2221-E-182-070.

Keywords: Drosophila driver image, Drosophila neuron images, intersection/union computation, parallel processing, OpenMP

Procedia PDF Downloads 205

Authors: Aishwarya Rai

Abstract:

This study explores the transformative influence of generative AI on startups, businesses, and industries. We will explore how large businesses can benefit in the area of customer operations, where AI-powered chatbots can improve self-service and agent effectiveness, greatly increasing efficiency. In marketing and sales, generative AI could transform businesses by automating content development, data utilization, and personalization, resulting in a substantial increase in marketing and sales productivity. In software engineering-focused startups, generative AI can streamline activities, significantly impacting coding processes and work experiences. It can be extremely useful in product R&D for market analysis, virtual design, simulations, and test preparation, altering old workflows and increasing efficiency. Zooming into the retail and CPG industry, industry findings suggest a 1-2% increase in annual revenues, equating to $400 billion to $660 billion. By automating customer service, marketing, sales, and supply chain management, generative AI can streamline operations, optimizing personalized offerings and presenting itself as a disruptive force. While celebrating economic potential, we acknowledge challenges like external inference and adversarial attacks. Human involvement remains crucial for quality control and security in the era of generative AI-driven transformative innovation. This talk provides a comprehensive exploration of generative AI's pivotal role in reshaping businesses, recognizing its strategic impact on customer interactions, productivity, and operational efficiency.

Keywords: generative AI, digital transformation, LLM, artificial intelligence, startups, businesses

Procedia PDF Downloads 39

Authors: Peter Egri, Tamas Kis

Abstract:

Scheduling problems have been studied by the algorithmic mechanism design research from the beginning. This paper is focusing on a practically important, but theoretically rather neglected field: the project scheduling problem where the jobs connected by precedence constraints compete for various nonrenewable resources, such as materials. Although the centralized problem can be solved in polynomial-time by applying the algorithm of Carlier and Rinnooy Kan from the Eighties, obtaining materials in a decentralized environment is usually far from optimal. It can be observed in practical production scheduling situations that project managers tend to cache the required materials as soon as possible in order to avoid later delays due to material shortages. This greedy practice usually leads both to excess stocks for some projects and materials, and simultaneously, to shortages for others. The aim of this study is to develop a model for the material allocation problem of a production plant, where a central decision maker—the inventory—should assign the resources arriving at different points in time to the jobs. Since the actual due dates are not known by the inventory, the mechanism design approach is applied with the projects as the self-interested agents. The goal of the mechanism is to elicit the required information and allocate the available materials such that it minimizes the maximal tardiness among the projects. It is assumed that except the due dates, the inventory is familiar with every other parameters of the problem. A further requirement is that due to practical considerations monetary transfer is not allowed. Therefore a mechanism without money is sought which excludes some widely applied solutions such as the Vickrey–Clarke–Groves scheme. In this work, a type of Serial Dictatorship Mechanism (SDM) is presented for the studied problem, including a polynomial-time algorithm for computing the material allocation. The resulted mechanism is both truthful and Pareto optimal. Thus the randomization over the possible priority orderings of the projects results in a universally truthful and Pareto optimal randomized mechanism. However, it is shown that in contrast to problems like the many-to-many matching market, not every Pareto optimal solution can be generated with an SDM. In addition, no performance guarantee can be given compared to the optimal solution, therefore this approximation characteristic is investigated with experimental study. All in all, the current work studies a practically relevant scheduling problem and presents a novel truthful material allocation mechanism which eliminates the potential benefit of the greedy behavior that negatively influences the outcome. The resulted allocation is also shown to be Pareto optimal, which is the most widely used criteria describing a necessary condition for a reasonable solution.

Keywords: material allocation, mechanism without money, polynomial-time mechanism, project scheduling

Procedia PDF Downloads 300

Authors: Chih-Chieh Yang, Chung-Neng Huang

Abstract:

In power system, how to improve the fault diagnosis technology of transmission line has always been the primary goal of power grid operators. In recent years, due to the rise of green energy, the addition of all kinds of distributed power also has an impact on the stability of the power system. Because the smart meters are with the function of data recording and bidirectional transmission, the adaptive Fuzzy Neural inference system, ANFIS, as well as the artificial intelligence that has the characteristics of learning and estimation in artificial intelligence. For transmission network, in order to avoid misjudgment of the fault type and location due to the input of these unstable power sources, combined with the above advantages of smart meter and ANFIS, a method for identifying fault types and location of faults is proposed in this study. In ANFIS training, the bus voltage and current information collected by smart meters can be trained through the ANFIS tool in MATLAB to generate fault codes to identify different types of faults and the location of faults. In addition, due to the uncertainty of distributed generation, a wind power system is added to the transmission network to verify the diagnosis correctness of the study. Simulation results show that the method proposed in this study can correctly identify the fault type and location of fault with more efficiency, and can deal with the interference caused by the addition of unstable power sources.

Keywords: ANFIS, fault diagnosis, power system, smart meter

Procedia PDF Downloads 116

Authors: Jiahui Yang, John Quigley, Lesley Walls

Abstract:

In this paper, the authors develop a stochastic model regarding the investment in supplier delivery performance development from a buyer’s perspective. The authors propose a multivariate model through a Multinomial-Dirichlet distribution within an Empirical Bayesian inference framework, representing both the epistemic and aleatory uncertainties in deliveries. A closed form solution is obtained and the lower and upper bound for both optimal investment level and expected profit under uncertainty are derived. The theoretical properties provide decision makers with useful insights regarding supplier delivery performance improvement problems where multiple delivery statuses are involved. The authors also extend the model from a single supplier investment into a supplier portfolio, using a Lagrangian method to obtain a theoretical expression for an optimal investment level and overall expected profit. The model enables a buyer to know how the marginal expected profit/investment level of each supplier changes with respect to the budget and which supplier should be invested in when additional budget is available. An application of this model is illustrated in a simulation study. Overall, the main contribution of this study is to provide an optimal investment decision making framework for supplier development, taking into account multiple delivery statuses as well as multiple projects.

Keywords: decision making, empirical bayesian, portfolio optimization, supplier development, supply chain management

Procedia PDF Downloads 268

Authors: Jack R. McKenzie, Peter A. Appleby, Thomas House, Neil Walton

Abstract:

Cold-start is a notoriously difficult problem which can occur in recommendation systems, and arises when there is insufficient information to draw inferences for users or items. To address this challenge, a contextual bandit algorithm – the Fast Approximate Bayesian Contextual Cold Start Learning algorithm (FAB-COST) – is proposed, which is designed to provide improved accuracy compared to the traditionally used Laplace approximation in the logistic contextual bandit, while controlling both algorithmic complexity and computational cost. To this end, FAB-COST uses a combination of two moment projection variational methods: Expectation Propagation (EP), which performs well at the cold start, but becomes slow as the amount of data increases; and Assumed Density Filtering (ADF), which has slower growth of computational cost with data size but requires more data to obtain an acceptable level of accuracy. By switching from EP to ADF when the dataset becomes large, it is able to exploit their complementary strengths. The empirical justification for FAB-COST is presented, and systematically compared to other approaches on simulated data. In a benchmark against the Laplace approximation on real data consisting of over 670, 000 impressions from autotrader.co.uk, FAB-COST demonstrates at one point increase of over 16% in user clicks. On the basis of these results, it is argued that FAB-COST is likely to be an attractive approach to cold-start recommendation systems in a variety of contexts.

Keywords: cold-start learning, expectation propagation, multi-armed bandits, Thompson Sampling, variational inference

Procedia PDF Downloads 83

Authors: Philippe Larvet

Abstract:

This paper presents the concept of an object-based programming language where tests (if... then... else) and control structures (while, repeat, for...) disappear and are replaced by conditions on data. According to the object paradigm, by using this concept, data are still embedded inside objects, as variable-value couples, but object methods are expressed into the form of logical propositions (‘conditions on data’ or COD).For instance : variable1 = value1 AND variable2 > value2 => variable3 = value3. Implementing this approach, a central inference engine turns and examines objects one after another, collecting all CODs of each object. CODs are considered as rules in a rule-based system: the left part of each proposition (left side of the ‘=>‘ sign) is the premise and the right part is the conclusion. So, premises are evaluated and conclusions are fired. Conclusions modify the variable-value couples of the object and the engine goes to examine the next object. The paper develops the principles of writing CODs instead of complex algorithms. Through samples, the paper also presents several hints for implementing a simple mechanism able to process this ‘COD language’. The proposed approach can be used within the context of simulation, process control, industrial systems validation, etc. By writing simple and rigorous conditions on data, instead of using classical and long-to-learn languages, engineers and specialists can easily simulate and validate the functioning of complex systems.

Keywords: conditions on data, logical proposition, programming without code, object-oriented programming, system simulation, system validation

Procedia PDF Downloads 197

Authors: Punitha Periyasamy, Mahalingam Sudalaimuthu, Sachikanta Nanda, Arasu Sundaram

Abstract:

India holds 17.5% of the world’s population but has only 2% of the total geographical area of the world where 27.35% of the area is categorized as wasteland due to lack of or less groundwater. So there is a demand for excessive groundwater for agricultural and non agricultural activities to balance its growth rate. With this in mind, an attempt is made to find the groundwater potential zone in Gomukhi river sub basin of Vellar River basin, TamilNadu, India covering an area of 1146.6 Sq.Km consists of 9 blocks from Peddanaickanpalayam to Villupuram fall in the sub basin. The thematic maps such as Geology, Geomorphology, Lineament, Landuse, and Landcover and Drainage are prepared for the study area using IRS P6 data. The collateral data includes rainfall, water level, soil map are collected for analysis and inference. The digital elevation model (DEM) is generated using Shuttle Radar Topographic Mission (SRTM) and the slope of the study area is obtained. ArcGIS 10.1 acts as a powerful spatial analysis tool to find out the ground water potential zones in the study area by means of weighted overlay analysis. Each individual parameter of the thematic maps are ranked and weighted in accordance with their influence to increase the water level in the ground. The potential zones in the study area are classified viz., Very Good, Good, Moderate, Poor with its aerial extent of 15.67, 381.06, 575.38, 174.49 Sq.Km respectively.

Keywords: ArcGIS, DEM, groundwater, recharge, weighted overlay

Procedia PDF Downloads 418

Authors: Piotr J. Balwierz, Mikhail Pachkov, Phil Arnold, Andreas J. Gruber, Mihaela Zavolan, Erik van Nimwegen

Abstract:

Understanding the key players and interactions in the regulatory networks that control gene expression and chromatin state across different cell types and tissues in metazoans remains one of the central challenges in systems biology. Our laboratory has pioneered a number of methods for automatically inferring core gene regulatory networks directly from high-throughput data by modeling gene expression (RNA-seq) and chromatin state (ChIP-seq) measurements in terms of genome-wide computational predictions of regulatory sites for hundreds of transcription factors and micro-RNAs. These methods have now been completely automated in an integrated webserver called ISMARA that allows researchers to analyze their own data by simply uploading RNA-seq or ChIP-seq data sets and provides results in an integrated web interface as well as in downloadable flat form. For any data set, ISMARA infers the key regulators in the system, their activities across the input samples, the genes and pathways they target, and the core interactions between the regulators. We believe that by empowering experimental researchers to apply cutting-edge computational systems biology tools to their data in a completely automated manner, ISMARA can play an important role in developing our understanding of regulatory networks across metazoans.

Keywords: gene expression analysis, high-throughput sequencing analysis, transcription factor activity, transcription regulation

Procedia PDF Downloads 33

Authors: Robert O. Abolade, Olumide O. Ajayi, Zacheaus K. Adeyemo, Solomon A. Adeniran

Abstract:

Link adaptation is an important strategy for achieving robust wireless multimedia communications based on quality of service (QoS) demand. Scheme switching in multiple-input multiple-output (MIMO) systems is an aspect of link adaptation, and it involves selecting among different MIMO transmission schemes or modes so as to adapt to the varying radio channel conditions for the purpose of achieving QoS delivery. However, finding the most appropriate switching method in MIMO links is still a challenge as existing methods are either computationally complex or not always accurate. This paper presents an intelligent switching method for the MIMO system consisting of two schemes - transmit diversity (TD) and spatial multiplexing (SM) - using fuzzy logic technique. In this method, two channel quality indicators (CQI) namely average received signal-to-noise ratio (RSNR) and received signal strength indicator (RSSI) are measured and are passed as inputs to the fuzzy logic system which then gives a decision – an inference. The switching decision of the fuzzy logic system is fed back to the transmitter to switch between the TD and SM schemes. Simulation results show that the proposed fuzzy logic – based switching technique outperforms conventional static switching technique in terms of bit error rate and spectral efficiency.

Keywords: channel quality indicator, fuzzy logic, link adaptation, MIMO, spatial multiplexing, transmit diversity

Procedia PDF Downloads 124

Authors: Anthony Townsend, Robyn Fasser

Abstract:

While complete objectivity is a desirable scientific position from which to conduct a care and contact evaluation (CCE), it is precisely the recognition that we are inherently incapable of operating objectively that is the foundation of ethical practice and skilled assessment. Drawing upon recent research from Daniel Kahneman (2021) on the differences between noise and bias, as well as different inherent biases collectively termed “The Elephant in the Brain” by Kevin Simler and Robin Hanson (2019) from Oxford University, this presentation addresses both the various ways in which our judgments, perceptions and even procedures can be distorted and contaminated while conducting a CCE, but also considers the value of second order cybernetics and the psychodynamic concept of ‘ambivalence’ as a conceptual basis to inform our assessment methodologies to limit such errors or at least better identify them. Both a conceptual framework for ambivalence, our higher-order capacity to allow for the convergence and consideration of multiple emotional experiences and cognitive perceptions to inform our reasoning, and a practical methodology for assessment relying on data triangulation, Bayesian inference and hypothesis testing is presented as a means of promoting ethical practice for health care professionals conducting CCEs. An emphasis on widening awareness and perspective, limiting ‘splitting’, is demonstrated both in how this form of emotional processing plays out in alienating dynamics in families as well as the assessment thereof. In addressing this concept, this presentation aims to illuminate the value of ambivalence as foundational to ethical practice for assessors.

Keywords: ambivalence, forensic, psychology, noise, bias, ethics

Procedia PDF Downloads 64

Authors: Hounyèmè Romuald, Maxime Logez, Mama Daouda, Argillier Christine

Abstract:

In view of the very strong degradation of aquatic ecosystems, it is urgent to set up monitoring systems that are best able to report on the effects of the stresses they undergo. This is particularly true in developing countries, where specific and relevant quality standards and funding for monitoring programs are lacking. The objective of this study was to make a relevant and objective choice of physicochemical parameters informative of the main stressors occurring on African lakes and to identify their alteration thresholds. Based on statistical analyses of the relationship between several driving forces and the physicochemical parameters of the Nokoué lagoon, relevant Physico-chemical parameters were selected for its monitoring. An innovative method based on Bayesian statistical modeling was used. Eleven Physico-chemical parameters were selected for their response to at least one stressor and their threshold quality standards were also established: Total Phosphorus (<4.5mg/L), Orthophosphates (<0.2mg/L), Nitrates (<0.5 mg/L), TKN (<1.85 mg/L), Dry Organic Matter (<5 mg/L), Dissolved Oxygen (>4 mg/L), BOD (<11.6 mg/L), Salinity (7.6 .), Water Temperature (<28.7 °C), pH (>6.2), and Transparency (>0.9 m). According to the System for the Evaluation of Coastal Water Quality, these thresholds correspond to” good to medium” suitability classes, except for total phosphorus. One of the original features of this study is the use of the bounds of the credibility interval of the fixed-effect coefficients as local weathering standards for the characterization of the Physico-chemical status of this anthropized African ecosystem.

Keywords: driving forces, alteration thresholds, acadjas, monitoring, modeling, human activities

Procedia PDF Downloads 66

Authors: M. N. Khajavi, S. Nasiri, E. Farokhi, M. R. Bavir

Abstract:

Alcohol fuels are renewable, have low pollution and have high octane number; therefore, they are important as fuel in internal combustion engines. Percentage detection of these alcoholic fuels with gasoline is a complicated, time consuming, and expensive process. Nowadays, these processes are done in equipped laboratories, based on international standards. The aim of this research is to determine percentage detection of different fuels based on vibration analysis of engine block signals. By doing, so considerable saving in time and cost can be achieved. Five different fuels consisted of pure gasoline (G) as base fuel and combination of this fuel with different percent of ethanol and methanol are prepared. For example, volumetric combination of pure gasoline with 10 percent ethanol is called E10. By this convention, we made M10 (10% methanol plus 90% pure gasoline), E30 (30% ethanol plus 70% pure gasoline), and M30 (30% Methanol plus 70% pure gasoline) were prepared. To simulate real working condition for this experiment, the vehicle was mounted on a chassis dynamometer and run under 1900 rpm and 30 KW load. To measure the engine block vibration, a three axis accelerometer was mounted between cylinder 2 and 3. After acquisition of vibration signal, eight time feature of these signals were used as inputs to an Adaptive Neuro Fuzzy Inference System (ANFIS). The designed ANFIS was trained for classifying these five different fuels. The results show suitable classification ability of the designed ANFIS network with 96.3 percent of correct classification.

Keywords: internal combustion engine, vibration signal, fuel composition, classification, ANFIS

Procedia PDF Downloads 374