Search results for: nonprofit organizations-national data maturity index (NDI)

Authors: Elaheh Vaezpour

Abstract:

Data sensing in wireless sensor networks is done by query deceleration the network by the users. In many applications of the wireless sensor networks, many users send queries to the network simultaneously. If the queries are processed separately, the network’s energy consumption will increase significantly. Therefore, it is very important to aggregate the queries before sending them to the network. In this paper, we propose a multiple query optimization framework based on sensors physical and temporal correlation. In the proposed method, queries are merged and sent to network by considering correlation among the sensors in order to reduce the communication cost between the sensors and the base station.

Keywords: wireless sensor networks, multiple query optimization, data correlation, reducing energy consumption

Procedia PDF Downloads 334

Authors: Yoonsuh Jung

Abstract:

As DNA microarray data contain relatively small sample size compared to the number of genes, high dimensional models are often employed. In high dimensional models, the selection of tuning parameter (or, penalty parameter) is often one of the crucial parts of the modeling. Cross-validation is one of the most common methods for the tuning parameter selection, which selects a parameter value with the smallest cross-validated score. However, selecting a single value as an "optimal" value for the parameter can be very unstable due to the sampling variation since the sample sizes of microarray data are often small. Our approach is to choose multiple candidates of tuning parameter first, then average the candidates with different weights depending on their performance. The additional step of estimating the weights and averaging the candidates rarely increase the computational cost, while it can considerably improve the traditional cross-validation. We show that the selected value from the suggested methods often lead to stable parameter selection as well as improved detection of significant genetic variables compared to the tradition cross-validation via real data and simulated data sets.

Keywords: cross validation, parameter averaging, parameter selection, regularization parameter search

Procedia PDF Downloads 415

Authors: Kathrin F. Schneider, Luis Xavier Unda Galarza

Abstract:

In our modern world, few are the areas that are not permeated by digitalization: we use digital tools for work, study, entertainment, and daily life. Since technology changes rapidly, skills must adapt to the new reality, which gives a dynamic dimension to the set of skills necessary for people's academic, professional, and personal success. The concept of 21st-century digital skills, which includes skills such as collaboration, communication, digital literacy, citizenship, problem-solving, critical thinking, interpersonal skills, creativity, and productivity, have been widely discussed in the literature. Digital transformation has opened many economic opportunities for entrepreneurs for the development of their products, financing possibilities, and product distribution. One of the biggest advantages is the reduction in cost for the entrepreneur, which has opened doors not only for the entrepreneur or the entrepreneurial team but also for corporations through intrapreneurship. The development of students' general literacy level and their digital competencies is crucial for improving the effectiveness and efficiency of the learning process, as well as for students' adaptation to the constantly changing labor market. The digital economy allows a free substantial increase in the supply share of conditional and also innovative products; this is mainly achieved through 5 ways to reduce costs according to the conventional digital economy: search costs, replication, transport, tracking, and verification. Digital entrepreneurship worldwide benefits from such achievements. There is an expansion and democratization of entrepreneurship thanks to the use of digital technologies. The digital transformation that has been taking place in recent years is more challenging for developing countries, as they have fewer resources available to carry out this transformation while offering all the necessary support in terms of cybersecurity and educating their people. The degree of digitization (use of digital technology) in a country and the levels of digital literacy of its people often depend on the economic level and situation of the country. Telefónica's Digital Life Index (TIDL) scores are strongly correlated with country wealth, reflecting the greater resources that richer countries can contribute to promoting "Digital Life". According to the Digitization Index, Ecuador is in the group of "emerging countries", while Chile, Colombia, Brazil, Argentina, and Uruguay are in the group of "countries in transition". According to Herrera Espinoza et al. (2022), there are startups or digital ventures in Ecuador, especially in certain niches, but many of the ventures do not exceed six months of creation because they arise out of necessity and not out of the opportunity. However, there is a lack of relevant research, especially empirical research, to have a clearer vision. Through a self-report questionnaire, the digital skills of students will be measured in an Ecuadorian private university, according to the skills identified as the six 21st-century skills. The results will be put to the test against the variable of the intention to start a digital venture measured using the theory of planned behavior (TPB). The main hypothesis is that high digital competence is positively correlated with the intention to start digital entrepreneurship.

Keywords: new literacies, digital transformation, 21st century skills, theory of planned behavior, digital entrepreneurship

Procedia PDF Downloads 105

Authors: Mingbao Chen

Abstract:

After Rio+20, the blue economy rises all over the world, and it has become the focus field of national development. At present, the blue economy has become a new growth point in the field of global economy and the direction of the development of ‘green’ in the ocean. However, in fact, the key factors affecting the development of the blue economy have not been explored in depth, and the development policies and performance of the blue economy have not been scientifically evaluated. This cannot provide useful guidance for the development of the blue economy. Therefore, it is urgent to establish a quantitative evaluation framework to measure the performance of the blue economic development. Based on the full understanding of the connotation and elements of the blue economy, and studying the literature, this article has built an universality and operability evaluation index system, including ecological environment, social justice, sustainable growth, policy measures, and so on. And this article also established a sound evaluation framework of blue economic development performance. At the same time, this article takes China as a sample to test the framework of the adaptability, and to assess the performance of China's blue economic.

Keywords: Blue economy, development performance, evaluation framework, assess method

Procedia PDF Downloads 248

Authors: Ji-Hun Choi, Kyoung-Suk Cho, Seung-Un Chae

Abstract:

Consisting of organic compounds, plastic ignites easily and burns fast. In addition, a large amount of toxic gas is produced while it is burning. When plastic is heated, its volume decreases because its surface is melted. The decomposition of its molecular bond generates combustible liquid of low viscosity, which accelerates plastic combustion and spreads the flames. Radiant heat produced in the process propagates the fire to increase the risk of human and property damages. Accordingly, the purpose of this study was to identify chemical, thermal and combustion characteristics of thermoplastic plastics using the fire propagation apparatus based on experimental criteria of ISO 12136 and ASTM E 2058. By the experiment result, as the ignition time increased, the thermal response parameter (TRP) decreased and as the TRP increased, the slope decreased. In other words, the large the TRP was, the longer the time taken for heating and ignition of the material was. It was identified that the fire propagation speed dropped accordingly.

Keywords: fire propagation apparatus (FPA), ISO 12136, thermal response parameter (TRP), fire propagation index (FPI)

Procedia PDF Downloads 202

Authors: Byung Ho Jung, Dong Hoon Lim

Abstract:

Logistic regression is a statistical method for analyzing a dataset in which there are one or more independent variables that determine an outcome. Logistic regression is used extensively in numerous disciplines, including the medical and social science fields. In this paper, we address the problem of estimating parameters in the logistic regression based on MapReduce framework with RHadoop that integrates R and Hadoop environment applicable to large scale data. There exist three learning algorithms for logistic regression, namely Gradient descent method, Cost minimization method and Newton-Rhapson's method. The Newton-Rhapson's method does not require a learning rate, while gradient descent and cost minimization methods need to manually pick a learning rate. The experimental results demonstrated that our learning algorithms using RHadoop can scale well and efficiently process large data sets on commodity hardware. We also compared the performance of our Newton-Rhapson's method with gradient descent and cost minimization methods. The results showed that our newton's method appeared to be the most robust to all data tested.

Keywords: big data, logistic regression, MapReduce, RHadoop

Procedia PDF Downloads 285

Authors: Chae Hoon Park, Jong Kang Park, Jong Tae Kim

Abstract:

Electrical characteristics of capacitive touchscreen need to be accurately analyzed to result in better performance for multi-channel capacitance sensing. In this paper, we extracted the panel resistances and capacitances of the touchscreen by comparing measurement data and model data. By employing a lumped RC model for driver-to-receiver paths in touchscreen, we estimated resistance and capacitance values according to the physical lengths of channel paths which are proportional to the RC model. As a result, we obtained the model having 95.54% accuracy of the measurement data.

Keywords: electrical characteristics of capacitive touchscreen, iterative extraction, lumped RC model, physical lengths of channel paths

Procedia PDF Downloads 334

Authors: Jun Ming Moey, Zhiyaun Chen, David Nicholson

Abstract:

Bad actors are often hard to detect in data that imprints their behaviour patterns because they are comparatively rare events embedded in non-bad actor data. An unsupervised machine learning framework is applied here to detect bad actors in financial crime datasets that record millions of transactions undertaken by hundreds of actors (<0.01% bad). Specifically, the framework combines ‘shallow’ (PCA, Isolation Forest) and ‘deep’ (Autoencoder) methods to detect outlier patterns. Detection performance analysis for both the individual methods and their combination is reported.

Keywords: detection, machine learning, deep learning, unsupervised, outlier analysis, data science, fraud, financial crime

Procedia PDF Downloads 95

Authors: Jan Stodt, Christoph Reich

Abstract:

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

Keywords: audit, machine learning, assessment, metrics

Procedia PDF Downloads 271

Authors: Keerthi S. Shetty, Annappa Basava

Abstract:

In recent years, modelling of biological systems represented by biochemical reactions has become increasingly important in Systems Biology. Biological systems represented by biochemical reactions are highly stochastic in nature. Probabilistic model is often used to describe such systems. One of the main challenges in Systems biology is to combine absolute experimental data into probabilistic model. This challenge arises because (1) some molecules may be present in relatively small quantities, (2) there is a switching between individual elements present in the system, and (3) the process is inherently stochastic on the level at which observations are made. In this paper, we describe a novel idea of combining absolute experimental data into probabilistic model using tool R2. Through a case study of the Transcription Process in Prokaryotes we explain how biological systems can be written as probabilistic program to combine experimental data into the model. The model developed is then analysed in terms of intrinsic noise and exact sampling of switching times between individual elements in the system. We have mainly concentrated on inferring number of genes in ON and OFF states from experimental data.

Keywords: systems biology, probabilistic model, inference, biology, model

Procedia PDF Downloads 349

Authors: Niyongabo Elyse

Abstract:

Prefabricated building construction is a pioneering approach that combines design, production, and assembly to attain energy efficiency, environmental sustainability, and economic feasibility. Despite continuous development in the industry in China, the low technical maturity of standardized design, factory production, and construction assembly introduces uncertainties affecting prefabricated component production and on-site assembly processes. This research focuses on enhancing project management performance under uncertainty to help enterprises navigate these challenges and optimize project resources. The study introduces a perspective on how uncertain factors influence the implementation of prefabricated building construction projects. It proposes a theoretical model considering project process management ability, adaptability to uncertain environments, and collaboration ability of project participants. The impact of uncertain factors is demonstrated through case studies and quantitative analysis, revealing constraints on implementation time, cost, quality, and safety. To address uncertainties in prefabricated component production scheduling, a fuzzy model is presented, expressing processing times in interval values. The model utilizes a cooperative co-evolution evolution algorithm (CCEA) to optimize scheduling, demonstrated through a real case study showcasing reduced project duration and minimized effects of processing time disturbances. Additionally, the research addresses on-site assembly construction scheduling, considering the relationship between task processing times and assigned resources. A multi-objective model with fuzzy activity durations is proposed, employing a hybrid cooperative co-evolution evolution algorithm (HCCEA) to optimize project scheduling. Results from real case studies indicate improved project performance in terms of duration, cost, and resilience to processing time delays and resource changes. The study also introduces a multistage dynamic process control model, utilizing IoT technology for real-time monitoring during component production and construction assembly. This approach dynamically adjusts schedules when constraints arise, leading to enhanced project management performance, as demonstrated in a real prefabricated housing project. Key contributions include a fuzzy prefabricated components production scheduling model, a multi-objective multi-mode resource-constrained construction project scheduling model with fuzzy activity durations, a multi-stage dynamic process control model, and a cooperative co-evolution evolution algorithm. The integrated mathematical model addresses the complexity of prefabricated building construction project management, providing a theoretical foundation for practical decision-making in the field.

Keywords: prefabricated construction, project management performance, uncertainty, fuzzy scheduling

Procedia PDF Downloads 50

Authors: E. Vazquez-Santacruz, M. Gamboa-Zuniga

Abstract:

In this paper, we show a methodology for bodies classification in lying state using HOG descriptors and pressures sensors positioned in a matrix form (14 x 32 sensors) on the surface where bodies lie down. it will be done in real time. Our system is embedded in a care robot that can assist the elderly patient and medical staff around to get a better quality of life in and out of hospitals. Due to current technology a limited number of sensors is used, wich results in low-resolution data array, that will be used as image of 14 x 32 pixels. Our work considers the problem of human posture classification with few information (sensors), applying digital process to expand the original data of the sensors and so get more significant data for the classification, however, this is done with low-cost algorithms to ensure the real-time execution.

Keywords: real-time classification, sensors, robots, health care, elderly patients, artificial intelligence

Procedia PDF Downloads 866

Authors: Fatma Kürüm Varolgüneş, Fatih Canan

Abstract:

Mardin is one of the unique cities in Turkey with its rich cultural and historical heritage. Mardin’s traditional dwellings have been affected both by natural data such as climate and topography and by cultural data like lifestyle and belief. However, in the new settlements, housing is formed with modern approaches and unsuitable forms clashing with Mardin’s culture and environment. While the city is expanding, traditional textures are ignored. Thus, traditional settlements are losing their identity and are vanishing because of the rapid change and transformation. The main aim of this paper is to determine the physical and social data needed to define the characteristic features of Mardin’s old and new settlements. In this context, based on social and cultural data, old and new settlement formations of Mardin have been investigated from various aspects. During this research, the following methods have been utilized: observations, interviews, public surveys, literature review, as well as site examination via maps, photographs and questionnaire methodology. In conclusion, this paper focuses on how changes in the physical forms of cities affect the typology and the identity of cities, as in the case of Mardin.

Keywords: urban and local identity, historical city center, traditional settlements, Mardin

Procedia PDF Downloads 328

Authors: Mina Salamatmanesh, Elizabeth Fitzpatrick, Tim Ramsay, Josee Lagacé, Lindsey Sikora, JoAnne Whittingham

Abstract:

Introduction: Hearing loss (HL) is one of the most common disorders that presents at birth and in early childhood. Universal newborn hearing screening (UNHS) has been adopted based on the assumption that with early identification of HL, children will have access to optimal amplification and intervention at younger ages, therefore, taking advantage of the brain’s maximal plasticity. One particular challenge for parents in the early years is achieving consistent hearing aid (HA) use which is critical to the child’s development and constitutes the first step in the rehabilitation process. This study examined the consistency of hearing aid use in young children based on data logging information documented during audiology sessions in the first three years after hearing aid fitting. Methodology: The first 100 children who were diagnosed with bilateral HL before 72 months of age since 2003 to 2015 in a pediatric audiology clinic and who had at least two hearing aid follow-up sessions with available data logging information were included in the study. Data from each audiology session (age of child at the session, average hours of use per day (for each ear) in the first three years after HA fitting) were collected. Clinical characteristics (degree of hearing loss, age of HA fitting) were also documented to further understanding of factors that impact HA use. Results: Preliminary analysis of the results of the first 20 children shows that all of them (100%) have at least one data logging session recorded in the clinical audiology system (Noah). Of the 20 children, 17(85%) have three data logging events recorded in the first three years after HA fitting. Based on the statistical analysis of the first 20 cases, the median hours of use in the first follow-up session after the hearing aid fitting in the right ear is 3.9 hours with an interquartile range (IQR) of 10.2h. For the left ear the median is 4.4 and the IQR is 9.7h. In the first session 47% of the children use their hearing aids ≤5 hours, 12% use them between 5 to 10 hours and 22% use them ≥10 hours a day. However, these children showed increased use by the third follow-up session with a median (IQR) of 9.1 hours for the right ear and 2.5, and of 8.2 hours for left ear (IQR) IQR is 5.6 By the third follow-up session, 14% of children used hearing aids ≤5 hours, while 38% of children used them ≥10 hours. Based on the primary results, factors like age and level of HL significantly impact the hours of use. Conclusion: The use of data logging information to assess the actual hours of HA provides an opportunity to examine the: a) challenges of families of young children with HAs, b) factors that impact use in very young children. Data logging when used collaboratively with parents, can be a powerful tool to identify problems and to encourage and assist families in maximizing their child’s hearing potential.

Keywords: hearing loss, hearing aid, data logging, hours of use

Procedia PDF Downloads 230

Authors: Maxence Queyrel, Edi Prifti, Alexandre Templier, Jean-Daniel Zucker

Abstract:

Analysis of the human microbiome using metagenomic sequencing data has demonstrated high ability in discriminating various human diseases. Raw metagenomic sequencing data require multiple complex and computationally heavy bioinformatics steps prior to data analysis. Such data contain millions of short sequences read from the fragmented DNA sequences and stored as fastq files. Conventional processing pipelines consist in multiple steps including quality control, filtering, alignment of sequences against genomic catalogs (genes, species, taxonomic levels, functional pathways, etc.). These pipelines are complex to use, time consuming and rely on a large number of parameters that often provide variability and impact the estimation of the microbiome elements. Training Deep Neural Networks directly from raw sequencing data is a promising approach to bypass some of the challenges associated with mainstream bioinformatics pipelines. Most of these methods use the concept of word and sentence embeddings that create a meaningful and numerical representation of DNA sequences, while extracting features and reducing the dimensionality of the data. In this paper we present an end-to-end approach that classifies patients into disease groups directly from raw metagenomic reads: metagenome2vec. This approach is composed of four steps (i) generating a vocabulary of k-mers and learning their numerical embeddings; (ii) learning DNA sequence (read) embeddings; (iii) identifying the genome from which the sequence is most likely to come and (iv) training a multiple instance learning classifier which predicts the phenotype based on the vector representation of the raw data. An attention mechanism is applied in the network so that the model can be interpreted, assigning a weight to the influence of the prediction for each genome. Using two public real-life data-sets as well a simulated one, we demonstrated that this original approach reaches high performance, comparable with the state-of-the-art methods applied directly on processed data though mainstream bioinformatics workflows. These results are encouraging for this proof of concept work. We believe that with further dedication, the DNN models have the potential to surpass mainstream bioinformatics workflows in disease classification tasks.

Keywords: deep learning, disease prediction, end-to-end machine learning, metagenomics, multiple instance learning, precision medicine

Procedia PDF Downloads 125

Authors: Si Chen, Quanhong Jiang

Abstract:

In the increasingly mature posture recognition technology, human movement information is widely used in sports rehabilitation, human-computer interaction, medical health, human posture assessment, and other fields today; this project uses the most original ideas; it is proposed to use the collection equipment for the collection of myoelectric data, reflect the muscle posture change on a degree of freedom through data processing, carry out data-muscle three-dimensional model joint adjustment, and realize basic pose recognition. Based on this, bionic aids or medical rehabilitation equipment can be further developed with the help of robotic arms and cutting-edge technology, which has a bright future and unlimited development space.

Keywords: pose recognition, 3D animation, electromyography, machine learning, bionics

Procedia PDF Downloads 79

Authors: Gaurav Kumar Sinha

Abstract:

In an era marked by increasing concerns about energy sustainability, this research endeavors to address the pressing challenge of energy consumption in buildings and industries. This study delves into the transformative potential of AWS services in optimizing energy efficiency. The research is founded on the recognition that effective management of energy consumption is imperative for both environmental conservation and economic viability. Buildings and industries account for a substantial portion of global energy use, making it crucial to develop advanced techniques for analysis and reduction. This study sets out to explore the integration of AWS services with big data analytics to provide innovative solutions for energy consumption analysis. Leveraging AWS's cloud computing capabilities, scalable infrastructure, and data analytics tools, the research aims to develop efficient methods for collecting, processing, and analyzing energy data from diverse sources. The core focus is on creating predictive models and real-time monitoring systems that enable proactive energy management. By harnessing AWS's machine learning and data analytics capabilities, the research seeks to identify patterns, anomalies, and optimization opportunities within energy consumption data. Furthermore, this study aims to propose actionable recommendations for reducing energy consumption in buildings and industries. By combining AWS services with metrics-driven insights, the research strives to facilitate the implementation of energy-efficient practices, ultimately leading to reduced carbon emissions and cost savings. The integration of AWS services not only enhances the analytical capabilities but also offers scalable solutions that can be customized for different building and industrial contexts. The research also recognizes the potential for AWS-powered solutions to promote sustainable practices and support environmental stewardship.

Keywords: energy consumption analysis, big data analytics, AWS services, energy efficiency

Procedia PDF Downloads 64

Authors: Navneet Kaur, Amarpreet Singh

Abstract:

In Vehicular Adhoc Networks, Data Dissemination is a challenging task. There are number of techniques, types and protocols available for disseminating the data but in order to preserve limited bandwidth and to disseminate maximum data over networks makes it more challenging. There are broadcasting, multicasting and geocasting based protocols. Multicasting based protocols are found to be best for conserving the bandwidth. One such protocol named BEAM exists that improves the performance of Vehicular Adhoc Networks by reducing the number of in-network message transactions and thereby efficiently utilizing the bandwidth during an emergency situation. But this protocol may result in multicar chain collision as there was no V2V communication. So, this paper proposes a new protocol named Enhanced Bandwidth Efficient Cluster Based Multicasting Protocol (EBECM) that will overcome the limitations of existing BEAM protocol. And Simulation results will show the improved performance of EBECM in terms of Routing overhead, throughput and PDR when compared with BEAM protocol.

Keywords: BEAM, data dissemination, emergency situation, vehicular adhoc network

Procedia PDF Downloads 348

Authors: Jean Marie Tshimula, Atsushi Togashi

Abstract:

We develop a data-science approach for providing an interactive visualization and predictive models to find insights into the projects' historical data in order for stakeholders understand some unseen opportunities in the African market that might escape them behind the online project portfolio of the African Development Bank. This machine learning-based web application identifies the market trend of the fastest growing economies across the continent as well skyrocketing sectors which have a significant impact on the future of business in Africa. Owing to this, the approach is tailored to predict where the investment needs are the most required. Moreover, we create a corpus that includes the descriptions of over more than 1,200 projects that approximately cover 14 sectors designed for some of 53 African countries. Then, we sift out this large amount of semi-structured data for extracting tiny details susceptible to contain some directions to follow. In the light of the foregoing, we have applied the combination of Latent Dirichlet Allocation and Random Forests at the level of the analysis module of our methodology to highlight the most relevant topics that investors may focus on for investing in Africa.

Keywords: machine learning, topic modeling, natural language processing, big data

Procedia PDF Downloads 168

Authors: Jude K. Safo

Abstract:

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

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

Procedia PDF Downloads 68

Authors: Annisa D. Febryandini

Abstract:

Batik is a unique cultural heritage that strongly linked with its community. As a product of current culture in Solo, Batik Solo not only has a specific design and color to represent the cultural identity, cultural values, and spirituality of the community, but also has some specific names given by its community which are not arbitrary. This qualitative research paper uses the primary data by interview method as well as the secondary data to support it. Based on the data, this paper concludes that the names consist of a word or words taken from a current name of things in Javanese language. They indicate the cultural meaning such as a specific event, a hope, and the social status of the people who use the motif. Different from the other research, this paper takes a look at the names of traditional motif of Batik Solo which analyzed linguistically to reveal the cultural meaning.

Keywords: traditional motif, Batik, solo, anthropological linguistics

Procedia PDF Downloads 277

Authors: XuKe Cao

Abstract:

This paper reviews the current state of research on low-resource languages in the field of speech recognition, focusing on the challenges faced by low-resource language speech recognition, including the scarcity of data resources, the lack of linguistic resources, and the diversity of dialects and accents. The article reviews recent progress in low-resource language speech recognition, including techniques such as data augmentation, end to-end models, transfer learning, and multi-task learning. Based on the challenges currently faced, the paper also provides an outlook on future research directions. Through these studies, it is expected that the performance of speech recognition for low resource languages can be improved, promoting the widespread application and adoption of related technologies.

Keywords: low-resource languages, speech recognition, data augmentation techniques, NLP

Procedia PDF Downloads 14

Authors: Sourav Ghosh, Antony Gomes

Abstract:

Since 1894 anti-snake venom serum (ASVS) is the only available treatment against snake envenomation, although there are many side effects and limitations. The need for a supportive treatment was felt for a long time to overcome the side effects and limitations of ASVS. Andrographolide conjugated with gold nanoparticle (A-GNP) has been found to antagonize viper venom-induced local damages. The present study was aimed to study the protective efficacy of A-GNP against Viper venom-induced inflammatory response and organ toxicity in animal model. Ethical clearance was obtained from animal experiments. Physico-chemical characterization of A-GNP was done by DLS (diameter and zeta potential), FE-SEM and XRD. Swiss albino male mice were divided into 4 groups: Gr.1-Sham control, Gr.2- Russell’s Viper venom (RVV) control, Gr.3- andrographolide treated and Gr.4- A-GNP treated. The 1/5th minimum lethal dose of RVV (500µg/kg, s.c.) was induced in animals of group 2, 3 & 4 animals, followed by treatment with andrographolide (100mg/kg, i.p.) and A-GNP (100mg/kg, i.v.) in group 3 & 4 animals, respectively. Blood was collected after 18 h, serum was prepared, and inflammatory markers (IL 1β, 6, 17a, 10, TNF α) and biochemical markers (AST, ACP, LDH, urea, creatinine) were assessed. Values were expressed as mean±SEM (n=4), one way ANOVA was done, P<0.05 was considered as statistically significant. DLS size showed the hydrodynamic diameter of A-GNP to be 230-260nm with polydispersity index of 0.103 and zeta potential was -18.32mV. XRD data confirmed the presence of crystalline gold in A-GNP, and FESEM indicated the presence of nearly spherical particle with size18-24nm.Treatment with A-GNP significantly decreased viper venom-induced proinflammatory markers (IL 1β, 6, 17, TNF α) increased anti-inflammatory markers (IL 10) and decreased organ toxicity markers (AST, ACP, LDH, urea, creatinine) in animal model. Venom neutralization efficacy of A-GNP was > andrographolide, which confirmed the increased efficacy of andrographolide after gold nanoparticle conjugation. Venom neutralization by A-GNP was due to anti-oxidant/anti-inflammatory activity of andrographolide, which showed increased efficacy after gold nanoparticle tagging. Thus, A-GNP may serve as a supportive therapy in snake-bite (against inflammatory response and organ toxicity) subject to further detail studies.

Keywords: andrographolide, gold nanoparticle, inflammatory response, organ toxicity, snake venom, snake venom neutralization, viper venom

Procedia PDF Downloads 374

Authors: Güzi̇n Ti̇rkeş, Özge Teki̇n, Kerem Kurtuluş, Y. Yekta Yurtseven, Murat Baran

Abstract:

A scanning electron microscope (SEM) is a type of electron microscope mainly used in nanoscience and nanotechnology areas. Automatic image recognition and classification are among the general areas of application concerning SEM. In line with these usages, the present paper proposes a deep learning algorithm that classifies SEM images into nine categories by means of an online application to simplify the process. The NFFA-EUROPE - 100% SEM data set, containing approximately 21,000 images, was used to train and test the algorithm at 80% and 20%, respectively. Validation was carried out using a separate data set obtained from the Middle East Technical University (METU) in Turkey. To increase the accuracy in the results, the Inception ResNet-V2 model was used in view of the Fine-Tuning approach. By using a confusion matrix, it was observed that the coated-surface category has a negative effect on the accuracy of the results since it contains other categories in the data set, thereby confusing the model when detecting category-specific patterns. For this reason, the coated-surface category was removed from the train data set, hence increasing accuracy by up to 96.5%.

Keywords: convolutional neural networks, deep learning, image classification, scanning electron microscope

Procedia PDF Downloads 125

Authors: Rutuja Desai

Abstract:

Search engine is fundamentally a framework used to search the data which is pertinent to the client via WWW. Looking close-by spot identified with the keywords is an imperative concept in developing web advances. For such kind of searching, extent pursuit or closest neighbor is utilized. In range search the forecast is made whether the objects meet to query object. Nearest neighbor is the forecast of the focuses close to the query set by the client. Here, the nearest neighbor methodology is utilized where Data recovery R+ tree is utilized rather than IR2 tree. The disadvantages of IR2 tree is: The false hit number can surpass the limit and the mark in Information Retrieval R-tree must have Voice over IP bit for each one of a kind word in W set is recouped by Data recovery R+ tree. The inquiry is fundamentally subordinate upon the key words and the geometric directions.

Keywords: information retrieval, nearest neighbor search, keyword search, R+ tree

Procedia PDF Downloads 291

Authors: Abdisalam Hassan Muse, Samuel Mwalili, Oscar Ngesa

Abstract:

A generalized log-logistic distribution with variable shapes of the hazard rate was introduced and studied, extending the log-logistic distribution by adding an extra parameter to the classical distribution, leading to greater flexibility in analysing and modeling various data types. The proposed distribution has a large number of well-known lifetime special sub-models such as; Weibull, log-logistic, exponential, and Burr XII distributions. Its basic mathematical and statistical properties were derived. The method of maximum likelihood was adopted for estimating the unknown parameters of the proposed distribution, and a Monte Carlo simulation study is carried out to assess the behavior of the estimators. The importance of this distribution is that its tendency to model both monotone (increasing and decreasing) and non-monotone (unimodal and bathtub shape) or reversed “bathtub” shape hazard rate functions which are quite common in survival and reliability data analysis. Furthermore, the flexibility and usefulness of the proposed distribution are illustrated in a real-life data set and compared to its sub-models; Weibull, log-logistic, and BurrXII distributions and other parametric survival distributions with 3-parmaeters; like the exponentiated Weibull distribution, the 3-parameter lognormal distribution, the 3- parameter gamma distribution, the 3-parameter Weibull distribution, and the 3-parameter log-logistic (also known as shifted log-logistic) distribution. The proposed distribution provided a better fit than all of the competitive distributions based on the goodness-of-fit tests, the log-likelihood, and information criterion values. Finally, Bayesian analysis and performance of Gibbs sampling for the data set are also carried out.

Keywords: hazard rate function, log-logistic distribution, maximum likelihood estimation, generalized log-logistic distribution, survival data, Monte Carlo simulation

Procedia PDF Downloads 202

Authors: Bedilu Tafesse

Abstract:

The diversity and ecological analysis of vascular epiphytes was studied in Gera Forest in southwestern Ethiopia at altitudes between 1600 and 2400 m.a.s.l. A total area of 4.5 ha was surveyed in coffee and non-coffee forest vegetation. Fifty sampling plots, each 30 m x 30 m (900 m2), were used for the purpose of data collection. A total of 59 species of vascular epiphytes were recorded, of which 34 (59%) were holo epiphytes, two (4%) were hemi epiphytes and 22 (37%) species were accidental vascular epiphytes. To study the altitudinal distribution of vascular epiphytes, altitudes were classified into higher >2000, middle 1800-2000 and lower 1600-1800 m.a.s.l. According to Shannon-Wiener Index (H/= 3.411) of alpha diversity the epiphyte community in the study area is medium. There was a statistically significant difference between host bark type and epiphyte richness as determined by one-way ANOVA p = 0.001 < 0.05. The post-hoc test shows that there is significant difference of vascular epiphytes richness between smooth bark with rough, flack and corky bark (P =0.001< 0.05), as well as rough and cork bark (p =0.43 <0.05). However, between rough and flack bark (p = 0.753 > 0.05) and between flack and corky bark (p = 0.854 > 0.05) no significant difference of epiphyte abundance was observed. Rough bark had 38%, corky 26%, flack 25%, and only 11% vascular epiphytes abundance occurred on smooth bark. The regression correlation test, (R2 = 0.773, p = 0.0001 < 0.05), showed that the number of species of vascular epiphytes and host DBH size are positively correlated. The regression correlation test (R2 = 0.28, p = 0.0001 < 0.05), showed that the number of species and host tree height positively correlated. The host tree preference of vascular epiphytes was recorded for only Vittaria volkensii species hosted on Syzygium guineense trees. The result of similarity analysis indicated that Gera Forest showed the highest vascular epiphytic similarity (0.35) with Yayu Forest and shared the least vascular epiphytic similarity (0.295) with Harenna Forest. It was concluded that horizontal stems and branches, large and rough, flack and corky bark type trees are more suitable for vascular epiphytes seedling attachments and growth. Conservation and protection of these phorophytes are important for the survival of vascular epiphytes and increase their ecological importance.

Keywords: accidental epiphytes, hemiepiphyte, holoepiphyte, phorophyte

Procedia PDF Downloads 332

Authors: Tanveer Hussain, Khan Muhammad, Amin Ullah, Mi Young Lee, Sung Wook Baik

Abstract:

The Big Data generated from distributed vision sensors installed on large scale in smart cities create hurdles in its efficient and beneficial exploration for browsing, retrieval, and indexing. This paper presents a three-folded framework for effective video summarization of such data and provide a compact and representative format of Big Video Data. In the first fold, the paper acquires input video data from the installed cameras and collect clues such as type and count of objects and clarity of the view from a chunk of pre-defined number of frames of each view. The decision of representative view selection for a particular interval is based on fuzzy inference system, acquiring a precise and human resembling decision, reinforced by the known clues as a part of the second fold. In the third fold, the paper forwards the selected view frames to the summary generation mechanism that is supported by a saliency-aware convolution neural network (CNN) model. The new trend of fuzzy rules for view selection followed by CNN architecture for saliency computation makes the multi-view video summarization (MVS) framework a suitable candidate for real-world practice in smart cities.

Keywords: big video data analysis, fuzzy logic, multi-view video summarization, saliency detection

Procedia PDF Downloads 188

Authors: Pierre van Tonder, Dinesh Muthoo, Kim twiname

Abstract:

Many existing and newly built structures are constructed on the design basis of the engineer and the workmanship of the construction company. However, when considering larger structures where more people are exposed to the building, its structural integrity is of great importance considering the safety of its occupants (Raghu, 2013). But how can the structural integrity of a building be monitored efficiently and effectively. This is where the fourth industrial revolution step in, and with minimal human interaction, data can be collected, analysed, and stored, which could also give an indication of any inconsistencies found in the data collected, this is where the Fibre Bragg Grating (FBG) monitoring system is introduced. This paper illustrates how data can be collected and converted to develop stress – strain behaviour and to produce bending moment diagrams for the utilisation and prediction of the structure’s integrity. Embedded fibre optic sensors were used in this study– fibre Bragg grating sensors in particular. The procedure entailed making use of the shift in wavelength demodulation technique and an inscription process of the phase mask technique. The fibre optic sensors considered in this report were photosensitive and embedded in the slab and beams for data collection and analysis. Two sets of fibre cables have been inserted, one purposely to collect temperature recordings and the other to collect strain and temperature. The data was collected over a time period and analysed used to produce bending moment diagrams to make predictions of the structure’s integrity. The data indicated the fibre Bragg grating sensing system proved to be useful and can be used for structural health monitoring in any environment. From the experimental data for the slab and beams, the moments were found to be64.33 kN.m, 64.35 kN.m and 45.20 kN.m (from the experimental bending moment diagram), and as per the idealistic (Ultimate Limit State), the data of 133 kN.m and 226.2 kN.m were obtained. The difference in values gave room for an early warning system, in other words, a reserve capacity of approximately 50% to failure.

Keywords: fibre bragg grating, structural health monitoring, fibre optic sensors, beams

Procedia PDF Downloads 139

Authors: Sachin Jain, Daniel Perez-Astudillo, Dunia A. Bachour, Antonio P. Sanfilippo

Abstract:

The future of solar energy in Qatar is evolving steadily. Hence, high-quality spatial solar radiation data is of the uttermost requirement for any planning and commissioning of solar technology. Generally, two types of solar radiation data are available: satellite data and ground observations. Satellite solar radiation data is developed by the physical and statistical model. Ground data is collected by solar radiation measurement stations. The ground data is of high quality. However, they are limited to distributed point locations with the high cost of installation and maintenance for the ground stations. On the other hand, satellite solar radiation data is continuous and available throughout geographical locations, but they are relatively less accurate than ground data. To utilize the advantage of both data, a product has been developed here which provides spatial continuity and higher accuracy than any of the data alone. The popular satellite databases: National Solar radiation Data Base, NSRDB (PSM V3 model, spatial resolution: 4 km) is chosen here for merging with ground-measured solar radiation measurement in Qatar. The spatial distribution of ground solar radiation measurement stations is comprehensive in Qatar, with a network of 13 ground stations. The monthly average of the daily total Global Horizontal Irradiation (GHI) component from ground and satellite data is used for error analysis. The normalized root means square error (NRMSE) values of 3.31%, 6.53%, and 6.63% for October, November, and December 2019 were observed respectively when comparing in-situ and NSRDB data. The method is based on the Empirical Bayesian Kriging Regression Prediction model available in ArcGIS, ESRI. The workflow of the algorithm is based on the combination of regression and kriging methods. A regression model (OLS, ordinary least square) is fitted between the ground and NSBRD data points. A semi-variogram is fitted into the experimental semi-variogram obtained from the residuals. The kriging residuals obtained after fitting the semi-variogram model were added to NSRBD data predicted values obtained from the regression model to obtain the final predicted values. The NRMSE values obtained after merging are respectively 1.84%, 1.28%, and 1.81% for October, November, and December 2019. One more explanatory variable, that is the ground elevation, has been incorporated in the regression and kriging methods to reduce the error and to provide higher spatial resolution (30 m). The final GHI maps have been created after merging, and NRMSE values of 1.24%, 1.28%, and 1.28% have been observed for October, November, and December 2019, respectively. The proposed merging method has proven as a highly accurate method. An additional method is also proposed here to generate calibrated maps by using regression and kriging model and further to use the calibrated model to generate solar radiation maps from the explanatory variable only when not enough historical ground data is available for long-term analysis. The NRMSE values obtained after the comparison of the calibrated maps with ground data are 5.60% and 5.31% for November and December 2019 month respectively.

Keywords: global horizontal irradiation, GIS, empirical bayesian kriging regression prediction, NSRDB

Procedia PDF Downloads 89