Search results for: speech dataset

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 105

Authors: Pattra Pukdeekiat, Siriluk Ruangrungrote

Abstract:

This work was originated to precisely determine direct solar radiation by using atmospheric physics models since the accurate prediction of solar radiation is necessary and useful for solar energy applications including atmospheric research. The possible models and techniques for a calculation of regional direct solar radiation were challenging and compulsory for the case of unavailable instrumental measurement. The investigation was mathematically governed by six astronomical parameters i.e. declination (δ), hour angle (ω), solar time, solar zenith angle (θz), extraterrestrial radiation (Iso) and eccentricity (E0) along with two atmospheric parameters i.e. air mass (mr) and dew point temperature at Bangna meteorological station (13.67° N, 100.61° E) in Bangkok, Thailand. Analyses of five models of solar radiation determination with the assumption of clear sky were applied accompanied by three statistical tests: Mean Bias Difference (MBD), Root Mean Square Difference (RMSD) and Coefficient of determination (R2) in order to validate the accuracy of obtainable results. The calculated direct solar radiation was in a range of 491-505 Watt/m2 with relative percentage error 8.41% for winter and 532-540 Watt/m2 with relative percentage error 4.89% for summer 2014. Additionally, dataset of seven continuous days, representing both seasons were considered with the MBD, RMSD and R2 of -0.08, 0.25, 0.86 and -0.14, 0.35, 3.29, respectively, which belong to Kumar model for winter and CSR model for summer. In summary, the determination of direct solar radiation based on atmospheric models and empirical equations could advantageously provide immediate and reliable values of the solar components for any site in the region without a constraint of actual measurement.

Keywords: atmospheric physics models, astronomical parameters, atmospheric parameters, clear sky condition

Procedia PDF Downloads 403

Authors: A. M. John-Otumu, M. M. Rahman, M. C. Onuoha, E. P. Ojonugwa

Abstract:

A Convolutional Neural Network (CNN) model for predicting Lassa fever was built using Python 3.8.0 programming language, alongside Keras 2.2.4 and TensorFlow 2.6.1 libraries as the development environment in order to reduce the current high risk of Lassa fever in West Africa, particularly in Nigeria. The study was prompted by some major flaws in existing conventional laboratory equipment for diagnosing Lassa fever (RT-PCR), as well as flaws in AI-based techniques that have been used for probing and prognosis of Lassa fever based on literature. There were 15,679 blood smear microscopic image datasets collected in total. The proposed model was trained on 70% of the dataset and tested on 30% of the microscopic images in avoid overfitting. A 3x3x3 convolution filter was also used in the proposed system to extract features from microscopic images. The proposed CNN-based model had a recall value of 96%, a precision value of 93%, an F1 score of 95%, and an accuracy of 94% in predicting and accurately classifying the images into clean or infected samples. Based on empirical evidence from the results of the literature consulted, the proposed model outperformed other existing AI-based techniques evaluated. If properly deployed, the model will assist physicians, medical laboratory scientists, and patients in making accurate diagnoses for Lassa fever cases, allowing the mortality rate due to the Lassa fever virus to be reduced through sound decision-making.

Keywords: artificial intelligence, ANN, blood smear, CNN, deep learning, Lassa fever

Procedia PDF Downloads 114

Authors: Benkouiten Imene, Chabli Ouerdia, Boutoutaou Hamid, Kadri Nesrine, Bouledroua Omar

Abstract:

Algeria is actively enhancing gas productivity by augmenting the supply flow. However, this effort has led to increased internal pressure, posing a potential risk to the pipeline's integrity, particularly in the presence of corrosion defects. Sonatrach relies on a vast network of pipelines spanning 24,000 kilometers for the transportation of gas and oil. The aging of these pipelines raises the likelihood of corrosion both internally and externally, heightening the risk of ruptures. To address this issue, a comprehensive inspection is imperative, utilizing specialized scraping tools. These advanced tools furnish a detailed assessment of all pipeline defects. It is essential to recalculate the pressure parameters to safeguard the corroded pipeline's integrity while ensuring the continuity of production. In this context, Sonatrach employs symbolic pressure limit calculations, such as ASME B31G (2009) and the modified ASME B31G (2012). The aim of this study is to perform a comparative analysis of various limit pressure calculation methods documented in the literature, namely DNV RP F-101, SHELL, P-CORRC, NETTO, and CSA Z662. This comparative assessment will be based on a dataset comprising 329 burst tests published in the literature. Ultimately, we intend to introduce a novel approach grounded in the finite element method, employing ANSYS software.

Keywords: pipeline burst pressure, burst test, corrosion defect, corroded pipeline, finite element method

Procedia PDF Downloads 54

Authors: Akankunda Veronicah Karuhanga

Abstract:

Dementia is an age mental disorder that makes victims lose normal functionality that needs delicate attention. It has been technically defined as a clinical syndrome that presents a number of difficulties in speech and other cognitive functions that change someone’s behaviors and can also cause impairments in activities of daily living, not forgetting a range of neurological disorders that bring memory loss and cognitive impairment. Family members are the primary healthcare givers and therefore, the way how they handle the situation in its early stages determines future deterioration syndromes like total memory loss. Unfortunately, most family members are ignorant about this condition and in most cases, the patients are brought to our facilities when their condition was already mismanaged by family members and we thus cannot do much. For example, incontinence can be managed at early stages through potty training or toilet scheduling before resorting to 24/7 diapers which are also not good. Professional Elderly care should be understood and practiced as an extension of homes, not a dumping place for people considered “abnormal” on account of ignorance. Immediate relatives should therefore be sensitized concerning the normalcy of dementia in the context of old age so that they can be understanding and supportive of dementia patients rather than discriminating against them as present-day lepers. There is a need to skill home-based caregivers on how to handle dementia in its early stages. Unless this is done, many of our elderly homes shall be filled with patients who should have been treated and supported from their homes. This skilling of home-based caregivers is a vital intervention because until elderly care is appreciated as a human moral obligation, many transactional rehabilitation centers will crop up and this shall be one of the worst moral decadences of our times.

Keywords: dementia, family, Alzheimers, relevancy

Procedia PDF Downloads 78

Authors: Tae-Uk Yun, Pullip Chung, Kee-Young Kwahk, Hyunchul Ahn

Abstract:

Fake news is defined as the news articles that are intentionally and verifiably false, and could mislead readers. Spread of fake news may provoke anxiety, chaos, fear, or irrational decisions of the public. Thus, detecting fake news and preventing its spread has become very important issue in our society. However, due to the huge amount of fake news produced every day, it is almost impossible to identify it by a human. Under this context, researchers have tried to develop automated fake news detection using machine learning techniques over the past years. But, there have been no prior studies proposed an automated fake news detection method for Korean news to our best knowledge. In this study, we aim to detect Korean fake news using text mining and machine learning techniques. Our proposed method consists of two steps. In the first step, the news contents to be analyzed is convert to quantified values using various text mining techniques (topic modeling, TF-IDF, and so on). After that, in step 2, classifiers are trained using the values produced in step 1. As the classifiers, machine learning techniques such as logistic regression, backpropagation network, support vector machine, and deep neural network can be applied. To validate the effectiveness of the proposed method, we collected about 200 short Korean news from Seoul National University’s FactCheck. which provides with detailed analysis reports from 20 media outlets and links to source documents for each case. Using this dataset, we will identify which text features are important as well as which classifiers are effective in detecting Korean fake news.

Keywords: fake news detection, Korean news, machine learning, text mining

Procedia PDF Downloads 273

Authors: C. L. Huang, J. J. Sia

Abstract:

The interest-based social networks, also known as social bookmark sharing systems, are useful platforms for people to conveniently read and collect internet resources. These platforms also providing function of social networks, and users can share and explore internet resources from the social networks. Providing personalized internet resources to users is an important issue on these platforms. This study uses two types of relationship on the social networks—following and follower and proposes a collaborative recommender system, consisting of two main steps. First, this study calculates the relationship strength between the target user and the target user's followings and followers to find top-N similar neighbors. Second, from the top-N similar neighbors, the articles (internet resources) that may interest the target user are recommended to the target user. In this system, users can efficiently obtain recent, related and diverse internet resources (knowledge) from the interest-based social network. This study collected the experimental dataset from Diigo, which is a famous bookmark sharing system. The experimental results show that the proposed recommendation model is more accurate than two traditional baseline recommendation models but slightly lower than the cosine model in accuracy. However, in the metrics of the diversity and executing time, our proposed model outperforms the cosine model.

Keywords: recommender systems, social networks, tagging, bookmark sharing systems, collaborative recommender systems, knowledge management

Procedia PDF Downloads 168

Authors: Kolade Olawande Owoeye, George Weir

Abstract:

Terrorism and extremism activities on the internet are becoming the most significant threats to national security because of their potential dangers. In response to this challenge, law enforcement and security authorities are actively implementing comprehensive measures by countering the use of the internet for terrorism. To achieve the measures, there is need for intelligence gathering via the internet. This includes real-time monitoring of potential websites that are used for recruitment and information dissemination among other operations by extremist groups. However, with billions of active webpages, real-time monitoring of all webpages become almost impossible. To narrow down the search domain, there is a need for efficient webpage classification techniques. This research proposed a new approach tagged: SentiPosit-based method. SentiPosit-based method combines features of the Posit-based method and the Sentistrenght-based method for classification of terrorism and extremism webpages. The experiment was carried out on 7500 webpages obtained through TENE-webcrawler by International Cyber Crime Research Centre (ICCRC). The webpages were manually grouped into three classes which include the ‘pro-extremist’, ‘anti-extremist’ and ‘neutral’ with 2500 webpages in each category. A supervised learning algorithm is then applied on the classified dataset in order to build the model. Results obtained was compared with existing classification method using the prediction accuracy and runtime. It was observed that our proposed hybrid approach produced a better classification accuracy compared to existing approaches within a reasonable runtime.

Keywords: sentiposit, classification, extremism, terrorism

Procedia PDF Downloads 274

Authors: S. Hutasavi, D. Chen

Abstract:

The built-up area is a significant proxy to measure regional economic growth and reflects the Gross Provincial Product (GPP). However, an up-to-date and reliable database of built-up areas is not always available, especially in developing countries. The cloud-based geospatial analysis platform such as Google Earth Engine (GEE) provides an opportunity with accessibility and computational power for those countries to generate the built-up data. Therefore, this study aims to extract the built-up areas in Eastern Economic Corridor (EEC), Thailand using day and nighttime satellite imagery based on GEE facilities. The normalized indices were generated from Landsat 8 surface reflectance dataset, including Normalized Difference Built-up Index (NDBI), Built-up Index (BUI), and Modified Built-up Index (MBUI). These indices were applied to identify built-up areas in EEC. The result shows that MBUI performs better than BUI and NDBI, with the highest accuracy of 0.85 and Kappa of 0.82. Moreover, the overall accuracy of classification was improved from 79% to 90%, and error of total built-up area was decreased from 29% to 0.7%, after night-time light data from the Visible and Infrared Imaging Suite (VIIRS) Day Night Band (DNB). The results suggest that MBUI with night-time light imagery is appropriate for built-up area extraction and be utilize for further study of socioeconomic impacts of regional development policy over the EEC region.

Keywords: built-up area extraction, google earth engine, adaptive thresholding method, rapid mapping

Procedia PDF Downloads 119

Authors: Shuo Mu, Guangzhi Jiang, Jinsa Chen

Abstract:

The analysis of Indel for RNA sequencing of clinical samples is easily affected by sequencing experiment errors and software selection. In order to improve the efficiency and accuracy of analysis, we developed an automatic reporting system for Indel recognition and annotation based on image snapshot of transcriptome reads alignment. This system includes sequence local-assembly and realignment, target point snapshot, and image-based recognition processes. We integrated high-confidence Indel dataset from several known databases as a training set to improve the accuracy of image processing and added a bioinformatical processing module to annotate and filter Indel artifacts. Subsequently, the system will automatically generate data, including data quality levels and images results report. Sanger sequencing verification of the reference Indel mutation of cell line NA12878 showed that the process can achieve 83% sensitivity and 96% specificity. Analysis of the collected clinical samples showed that the interpretation accuracy of the process was equivalent to that of manual inspection, and the processing efficiency showed a significant improvement. This work shows the feasibility of accurate Indel analysis of clinical next-generation sequencing (NGS) transcriptome. This result may be useful for RNA study for clinical samples with microsatellite instability in immunotherapy in the future.

Keywords: automatic reporting, indel, next-generation sequencing, NGS, transcriptome

Procedia PDF Downloads 185

Authors: Aaditya U. Jhamb

Abstract:

Due to the current energy crisis that many countries are battling, energy-efficient buildings are the subject of extensive research in the modern technological era because of growing worries about energy consumption and its effects on the environment. The paper explores 8 factors that help determine energy efficiency for a building: (relative compactness, surface area, wall area, roof area, overall height, orientation, glazing area, and glazing area distribution), with Tsanas and Xifara providing a dataset. The data set employed 768 different residential building models to anticipate heating and cooling loads with a low mean squared error. By optimizing these characteristics, machine learning algorithms may assess and properly forecast a building's heating and cooling loads, lowering energy usage while increasing the quality of people's lives. As a result, the paper studied the magnitude of the correlation between these input factors and the two output variables using various statistical methods of analysis after determining which input variable was most closely associated with the output loads. The most conclusive model was the Decision Tree Regressor, which had a mean squared error of 0.258, whilst the least definitive model was the Isotonic Regressor, which had a mean squared error of 21.68. This paper also investigated the KNN Regressor and the Linear Regression, which had to mean squared errors of 3.349 and 18.141, respectively. In conclusion, the model, given the 8 input variables, was able to predict the heating and cooling loads of a residential building accurately and precisely.

Keywords: energy efficient buildings, heating load, cooling load, machine learning models

Procedia PDF Downloads 90

Authors: Riccardo Pasquali, Keith Harlin, Mark Muller

Abstract:

The ShallowTherm project is focussed on developing and applying a methodology for extrapolating relatively sparsely sampled thermal conductivity measurements across Ireland using mapped Litho-Electrical (LE) units. The primary data used consist of electrical resistivities derived from the Geological Survey Ireland Tellus airborne electromagnetic dataset, GIS-based maps of Irish geology, and rock thermal conductivities derived from both the current Irish Ground Thermal Properties (IGTP) database and a new programme of sampling and laboratory measurement. The workflow has been developed across three case-study areas that sample a range of different calcareous, arenaceous, argillaceous, and volcanic lithologies. Statistical analysis of resistivity data from individual geological formations has been assessed and integrated with detailed lithological descriptions to define distinct LE units. Thermal conductivity measurements from core and hand samples have been acquired for every geological formation within each study area. The variability and consistency of thermal conductivity measurements within each LE unit is examined with the aim of defining a characteristic thermal conductivity (or range of thermal conductivities) for each LE unit. Mapping of LE units, coupled with characteristic thermal conductivities, provides a method of defining thermal conductivity properties at a regional scale and facilitating the design of ground source heat pump closed-loop collectors.

Keywords: thermal conductivity, ground source heat pumps, resistivity, heat exchange, shallow geothermal, Ireland

Procedia PDF Downloads 173

Authors: Rishabh Ambavanekar

Abstract:

Expressive Aphasia (EA) is an oral disability, common among stroke victims, in which the Broca’s area of the brain is damaged, interfering with verbal communication abilities. EA currently has no technological solutions and its only current viable solutions are inefficient or only available to the affluent. This prompts the need for an affordable, innovative solution to facilitate recovery and assist in speech generation. This project proposes a novel concept: using a wearable low-cost electroencephalography (EEG) device-based brain-computer interface (BCI) to translate a user’s inner dialogue into words. A low-cost EEG device was developed and found to be 10 to 100 times less expensive than any current EEG device on the market. As part of the BCI, a machine learning (ML) model was developed and trained using the EEG data. Two stages of testing were conducted to analyze the effectiveness of the device: a proof-of-concept and a final solution test. The proof-of-concept test demonstrated an average accuracy of above 90% and the final solution test demonstrated an average accuracy of above 75%. These two successful tests were used as a basis to demonstrate the viability of BCI research in developing lower-cost verbal communication devices. Additionally, the device proved to not only enable users to verbally communicate but has the potential to also assist in accelerated recovery from the disorder.

Keywords: neurotechnology, brain-computer interface, neuroscience, human-machine interface, BCI, HMI, aphasia, verbal disability, stroke, low-cost, machine learning, ML, image recognition, EEG, signal analysis

Procedia PDF Downloads 116

Authors: Hala Alshamlan, Ghada Badr, Yousef Alohali

Abstract:

Cancer is one of the dreadful diseases, which causes considerable death rate in humans. DNA microarray-based gene expression profiling has been emerged as an efficient technique for cancer classification, as well as for diagnosis, prognosis, and treatment purposes. In recent years, a DNA microarray technique has gained more attraction in both scientific and in industrial fields. It is important to determine the informative genes that cause cancer to improve early cancer diagnosis and to give effective chemotherapy treatment. In order to gain deep insight into the cancer classification problem, it is necessary to take a closer look at the proposed gene selection methods. We believe that they should be an integral preprocessing step for cancer classification. Furthermore, finding an accurate gene selection method is a very significant issue in a cancer classification area because it reduces the dimensionality of microarray dataset and selects informative genes. In this paper, we classify and review the state-of-art gene selection methods. We proceed by evaluating the performance of each gene selection approach based on their classification accuracy and number of informative genes. In our evaluation, we will use four benchmark microarray datasets for the cancer diagnosis (leukemia, colon, lung, and prostate). In addition, we compare the performance of gene selection method to investigate the effective gene selection method that has the ability to identify a small set of marker genes, and ensure high cancer classification accuracy. To the best of our knowledge, this is the first attempt to compare gene selection approaches for cancer classification using microarray gene expression profile.

Keywords: gene selection, feature selection, cancer classification, microarray, gene expression profile

Procedia PDF Downloads 447

Authors: Neha Gupta

Abstract:

Landslide vulnerability is considered as the crucial parameter for the assessment of landslide risk. The term vulnerability defined as the damage or degree of elements at risk of different dimensions, i.e., physical, social, economic, and environmental dimensions. Himalaya region is very prone to multi-hazard such as floods, forest fires, earthquakes, and landslides. With the increases in fatalities rates, loss of infrastructure, and economy due to landslide in the Himalaya region, leads to the assessment of vulnerability. In this study, a methodology to measure the combination of vulnerability dimension, i.e., social vulnerability, physical vulnerability, and environmental vulnerability in one framework. A combined result of these vulnerabilities has rarely been carried out. But no such approach was applied in the Indian Scenario. The methodology was applied in an area of east Sikkim Himalaya, India. The physical vulnerability comprises of building footprint layer extracted from remote sensing data and Google Earth imaginary. The social vulnerability was assessed by using population density based on land use. The land use map was derived from a high-resolution satellite image, and for environment vulnerability assessment NDVI, forest, agriculture land, distance from the river were assessed from remote sensing and DEM. The classes of social vulnerability, physical vulnerability, and environment vulnerability were normalized at the scale of 0 (no loss) to 1 (loss) to get the homogenous dataset. Then the Multi-Criteria Analysis (MCA) was used to assign individual weights to each dimension and then integrate it into one frame. The final vulnerability was further classified into four classes from very low to very high.

Keywords: landslide, multi-criteria analysis, MCA, physical vulnerability, social vulnerability

Procedia PDF Downloads 298

Authors: Mimi Roy, Sriroop Chaudhuri

Abstract:

Sustainable water supply is among the basic needs for human development, especially in the rural areas of the developing nations where safe water supply and basic sanitation infrastructure is direly needed. In light of the above, we propose a simple methodology to develop a composite water sustainability index (WSI) to assess the collective performance of the existing rural water supply services (RWSS) in India over time. The WSI will be computed by summarizing the details of all the different varieties of water supply schemes presently available in India comprising of 40 liters per capita per day (lpcd), 55 lpcd, and piped water supply (PWS) per household. The WSI will be computed annually, between 2010 and 2016, to elucidate changes in holistic RWSS performances. Results will be integrated within a robust geospatial framework to identify the ‘hotspots’ (states/districts) which have persistent issues over adequate RWSS coverage and warrant spatially-optimized policy reforms in future to address sustainable human development. Dataset will be obtained from the National Rural Drinking Water Program (NRDWP), operating under the aegis of the Ministry of Drinking Water and Sanitation (MoDWS), at state/district/block levels to offer the authorities a cross-sectional view of RWSS at different levels of administrative hierarchy. Due to simplistic design, complemented by spatio-temporal cartograms, similar approaches can also be adopted in other parts of the world where RWSS need a thorough appraisal.

Keywords: rural water supply services, piped water supply, sustainability, composite index, spatial, drinking water

Procedia PDF Downloads 292

Authors: Ikechi Gilbert

Abstract:

Oral communication is the transmission of ideas or messages through the speech process. Acquiring competence in this area which, by its volatile nature, is prone to errors and inaccuracies would require the adoption of a well-suited teaching methodology. Efficient oral communication facilitates exchange of ideas and easy accomplishment of day-to-day tasks, by means of a demonstrated mastery of oral expression and the making of fine presentations to audiences or individuals while recognizing verbal signals and body language of others and interpreting them correctly. In Anglophone states such as Nigeria, Ghana, etc., the French language, for instance, is studied as a foreign language, being used majorly in teaching learners who have their own mother tongue different from French. The same applies to Francophone states where English is studied as a foreign language by people whose official language or mother tongue is different from English. The ideal approach would be to teach these languages in these environments through a pedagogical approach that properly takes care of the oral perspective for effective understanding and application by the learners. In this article, we are examining the communicative approach as a methodology for teaching oral communication in a foreign language. This method is a direct response to the communicative needs of the learner involving the use of appropriate materials and teaching techniques that meet those needs. It is also a vivid improvement to the traditional grammatical and audio-visual adaptations. Our contribution will focus on the pedagogical component of oral communication improvement, highlighting its merits and also proposing diverse techniques including aspects of information and communication technology that would assist the second language learner communicate better orally.

Keywords: communication, competence, methodology, pedagogical component

Procedia PDF Downloads 261

Authors: Sara ElElimy, Samir Moustafa

Abstract:

Mobile network operators start to face many challenges in the digital era, especially with high demands from customers. Since mobile network operators are considered a source of big data, traditional techniques are not effective with new era of big data, Internet of things (IoT) and 5G; as a result, handling effectively different big datasets becomes a vital task for operators with the continuous growth of data and moving from long term evolution (LTE) to 5G. So, there is an urgent need for effective Big data analytics to predict future demands, traffic, and network performance to full fill the requirements of the fifth generation of mobile network technology. In this paper, we introduce data science techniques using machine learning and deep learning algorithms: the autoregressive integrated moving average (ARIMA), Bayesian-based curve fitting, and recurrent neural network (RNN) are employed for a data-driven application to mobile network operators. The main framework included in models are identification parameters of each model, estimation, prediction, and final data-driven application of this prediction from business and network performance applications. These models are applied to Telecom Italia Big Data challenge call detail records (CDRs) datasets. The performance of these models is found out using a specific well-known evaluation criteria shows that ARIMA (machine learning-based model) is more accurate as a predictive model in such a dataset than the RNN (deep learning model).

Keywords: big data analytics, machine learning, CDRs, 5G

Procedia PDF Downloads 137

Authors: Smaran Manchala

Abstract:

Homomorphic Encryption (HE) enables computations on encrypted data without requiring decryption, mitigating data vulnerability during processing. Usable Fully Homomorphic Encryption (FHE) could revolutionize secure data operations across cloud computing, AI training, and healthcare, providing both privacy and functionality, however, the computational inefficiency of schemes like Cheon-Kim-Kim-Song (CKKS) hinders their widespread practical use. This study focuses on optimizing CKKS for faster matrix operations through the implementation of vector computation parallelization and GPU acceleration. The variable effects of vector parallelization on GPUs were explored, recognizing that while parallelization typically accelerates operations, it could introduce overhead that results in slower runtimes, especially in smaller, less computationally demanding operations. To assess performance, two neural network models, MLPN and CNN—were tested on the MNIST dataset using both ARM and x86-64 architectures, with CNN chosen for its higher computational demands. Each test was repeated 1,000 times, and outliers were removed via Z-score analysis to measure the effect of vector parallelization on CKKS performance. Model accuracy was also evaluated under CKKS encryption to ensure optimizations did not compromise results. According to the results of the trail runs, applying vector parallelization had a 2.63X efficiency increase overall with a 1.83X performance increase for x86-64 over ARM architecture. Overall, these results suggest that the application of vector parallelization in tandem with GPU acceleration significantly improves the efficiency of CKKS even while accounting for vector parallelization overhead, providing impact in future zero trust operations.

Keywords: CKKS scheme, runtime efficiency, fully homomorphic encryption (FHE), GPU acceleration, vector parallelization

Procedia PDF Downloads 0

Authors: Deokho Kim, Wonwoo Lee, Jaewoong Lee, Teresa Ng, Gun-Ill Lee, Jiwon Jeong

Abstract:

This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.

Keywords: visual search, deep learning, convolutional neural network, machine learning

Procedia PDF Downloads 213

Authors: Tsetsegmaa A., Bayarsuren B., Altantsetseg Ts.

Abstract:

To make the best decision on reducing harmful emissions from buses, we need to have a clear understanding of the current state of their actual emissions. The emissions from city buses running on high sulfur fuel, particularly particulate matter (PM) and nitrogen oxides (NOx) from the exhaust gases of conventional diesel engines, have been studied and measured with and without diesel particulate filter (DPF) in Ulaanbaatar city. The study was conducted by using the PEMS (Portable Emissions Measurement System) and gravimetric method in real traffic conditions. The obtained data were used to determine the actual emission rates and to evaluate the effectiveness of the selected particulate filters. Actual road and daily PM emissions from city buses were determined during the warm and cold seasons. A bus with an average daily mileage of 242 km was found to emit 166.155 g of PM into the city's atmosphere on average per day, with 141.3 g in summer and 175.8 g in winter. The actual PM of the city bus is 0.6866 g/km. The concentration of NOx in the exhaust gas averages 1410.94 ppm. The use of DPF reduced the exhaust gas opacity of 24 buses by an average of 97% and filtered a total of 340.4 kg of soot from these buses over a period of six months. Retrofitting an old conventional diesel engine with cassette-type silicon carbide (SiC) DPF, despite the laboriousness of cleaning, can significantly reduce particulate matter emissions. Innovation: First comprehensive road PM and NOx emission dataset and actual road emissions from public buses have been identified. PM and NOx mathematical model equations have been estimated as a function of the bus technical speed and engine revolution with and without DPF.

Keywords: conventional diesel, silicon carbide, real-time onboard measurements, particulate matter, diesel retrofit, fuel sulphur

Procedia PDF Downloads 157

Authors: Kusmayadi Ita Nursita, Sundoro Ali

Abstract:

Cleft lip and palate are one of the most common congenital abnormalities in the face. It could happen to anyone, but mostly affect Asian population including Indonesia. Factors that influence the occurrence of cleft lip and palate vary from genetic to environmental factors. Children with cleft lip and palate will often have various problems such as airway disorders, eating disorders, speech and language developmental disorders, hearing disorders and psycho-social disorders, one of which is caused by appearance disorders. During his life, the child will experience multidisciplinary surgery and non-surgical treatment and can be accompanied by a psychological and financial burden on himself and his family. In Indonesia, there are no detailed scientific data on the prevalence and characteristic of cleft lip and palate patients. It was mainly caused by the absence of a national level organization, differences in geographical location, and the absence of national guidelines. This study aimed to describe the characteristic and prevalence of cleft lip and palate patients in Bandung Cleft Lip and Palate Center from 1 January 2016 to 31 December 2017. A total of 560 patients were included in the study. The highest percentage of cases are left unilateral cleft lip and palate with higher number of female patient and labioplasty as the most often surgical procedure to be conducted in Bandung Cleft Lip and Palate Center. In order to improve quality of life in patients with cleft lip and palate, early recognition and early treatment based on actual comprehensive data should be conducted. The data from Bandung Cleft Lip and Palate Center as one of the largest center of cleft lip and palate in West Java Indonesia hopefully could provide a big step of further comprehensive data collection in Indonesia and for the better overall management of cleft lip and palate in the future.

Keywords: cleft lip, cleft palate, characteristic, prevalence

Procedia PDF Downloads 131

Authors: Mohamed Lotfy, Ghada Soliman

Abstract:

Recognizing Arabic dot-matrix digits is a challenging problem due to the unique characteristics of dot-matrix fonts, such as irregular dot spacing and varying dot sizes. This paper presents an approach for recognizing Arabic digits printed in dot matrix format. The proposed model is based on Convolutional Neural Networks (CNN) that take the dot matrix as input and generate embeddings that are rounded to generate binary representations of the digits. The binary embeddings are then used to perform Optical Character Recognition (OCR) on the digit images. To overcome the challenge of the limited availability of dotted Arabic expiration date images, we developed a True Type Font (TTF) for generating synthetic images of Arabic dot-matrix characters. The model was trained on a synthetic dataset of 3287 images and 658 synthetic images for testing, representing realistic expiration dates from 2019 to 2027 in the format of yyyy/mm/dd. Our model achieved an accuracy of 98.94% on the expiry date recognition with Arabic dot matrix format using fewer parameters and less computational resources than traditional CNN-based models. By investigating and presenting our findings comprehensively, we aim to contribute substantially to the field of OCR and pave the way for advancements in Arabic dot-matrix character recognition. Our proposed approach is not limited to Arabic dot matrix digit recognition but can also be extended to text recognition tasks, such as text classification and sentiment analysis.

Keywords: computer vision, pattern recognition, optical character recognition, deep learning

Procedia PDF Downloads 84

Authors: Nima Pirhadi, Shao Yong Bo, Xusheng Wan, Jianguo Lu, Jilei Hu

Abstract:

Although gravels and gravelly soils are assumed to be non-liquefiable because of high conductivity and small modulus; however, the occurrence of this phenomenon in some historical earthquakes, especially recently earthquakes during 2008 Wenchuan, Mw= 7.9, 2014 Cephalonia, Greece, Mw= 6.1 and 2016, Kaikoura, New Zealand, Mw = 7.8, has been promoted the essential consideration to evaluate risk assessment and hazard analysis of seismic gravelly soil liquefaction. Due to the limitation in sampling and laboratory testing of this type of soil, in situ tests and site exploration of case histories are the most accepted procedures. Of all in situ tests, dynamic penetration test (DPT), Which is well known as the Chinese dynamic penetration test, and shear wave velocity (Vs) test, have been demonstrated high performance to evaluate seismic gravelly soil liquefaction. However, the lack of a sufficient number of case histories provides an essential limitation for developing new models. This study at first investigates recent earthquakes that caused liquefaction in gravelly soils to collect new data. Then, it adds these data to the available literature’s dataset to extend them and finally develops new models to assess seismic gravelly soil liquefaction. To validate the presented models, their results are compared to extra available models. The results show the reasonable performance of the proposed models and the critical effect of gravel content (GC)% on the assessment.

Keywords: liquefaction, gravel, dynamic penetration test, shear wave velocity

Procedia PDF Downloads 198

Authors: Vineesh Amin, Ananya Agrawal

Abstract:

In a prolonged period of uncertainty and disruptions in the pre-said normal order, non-cognitive skills, especially entrepreneurial mindsets, have become a pillar that can reform the educational models to inform the economy. Dreamverse Learning Lab’s IVR-based storytelling program -Call-a-Kahaani- is an evolving experiment with an aim to kindle entrepreneurial mindsets in the remotest locations of India in an accessible and engaging manner. At the heart of this experiment is the belief that at every phase in our life’s story, we have a choice which brings us closer to achieving our true potential. This interactive program is thus designed using real-time storytelling principles to empower learners, ages 24 and below, to make choices and take decisions as they become more self-aware, practice grit, try new things through stories, guided activities, and interactions, simply over a phone call. This research paper highlights the framework behind an ongoing scalable, data-oriented, low-tech program to kindle entrepreneurial mindsets in media dark zones supported by iterative design and prototyping to reach 13700+ unique learners who made 59000+ calls for 183900+min listening duration to listen to content pieces of around 3 to 4 min, with the last monitored (March 2022) record of 34% serious listenership, within one and a half years of its inception. The paper provides an in-depth account of the technical development, content creation, learning, and assessment frameworks, as well as mobilization models which have been leveraged to build this end-to-end system.

Keywords: non-cognitive skills, entrepreneurial mindsets, speech interface, remote learning, storytelling

Procedia PDF Downloads 205

Authors: Emmanuel Abban Sagoe

Abstract:

A proper functioning of the Central Nervous System ensures that we are able to accomplish just about everything we do as human beings such as walking, breathing, running, etc. Myelinated neurons throughout the body which transmit signals at high speeds facilitate these actions. In the case of MS, the body’s immune system attacks the myelin sheath surrounding the neurons and overtime destroys the myelin sheaths. Depending upon where the destruction occurs in the brain symptoms can vary from person to person. Fatigue is, however, the biggest problem encountered by an MS sufferer. It is very often described as the bedrock upon which other symptoms of MS such challenges in balance and coordination, dizziness, slurred speech, etc. may occur. Classifying and distinguishing between perceptions based fatigue and performance based fatigability is key to identifying appropriate treatment options for patients. Objective methods for assessing motor fatigability is also key to providing clinicians and physiotherapist with critical information on the progression of the symptom. This study tested if the Fatigue Index Kliniken Schmieder assessment tool can detect fatigability as seen in MS patients when healthy subjects with no known history of neurological pathology mimic abnormal gaits. Thirty three healthy adults between ages 18-58years volunteered as subjects for the study. The subjects, strapped with RehaWatch sensors on both feet, completed 6 gait protocols of normal and mimicked fatigable gaits for 60 seconds per each gait and at 1.38889m/s treadmill speed following clear instructions given.

Keywords: attractor attributes, fatigue index Kliniken Schmieder, gait variability, movement pattern

Procedia PDF Downloads 116

Authors: Lison Carlier

Abstract:

The English language that is taught in school or used in media nowadays is defined as 'standard English,' although unstandardized Englishes, or 'parallel' Englishes, are practiced throughout the world. The existence of these 'parallel' Englishes has challenged standardization by imposing its own specific vocabulary or grammar. These non-standard languages tend to be regarded as inferior and, therefore, pose a problem regarding their translation. In the USA, 'slanguage', or slang, is a good example of a 'parallel' language. It consists of a particular set of vocabulary, used mostly in speech, and rarely in writing. Qualified as vulgar, often reduced to an urban language spoken by young people from lower classes, slanguage – or the language that is often first spoken between youths – is still the most common language used in the English-speaking world. Moreover, it appears that the prime meaning of 'informal' (as in an informal language) – a language that is spoken with persons the speaker knows – has been put aside and replaced in the general mind by the idea of vulgarity and non-appropriateness, when in fact informality is a sign of intimacy, not of vulgarity. When it comes to translating American slang, the main problem a translator encounters is the image and the cultural background usually associated with this 'parallel' language. Indeed, one will have, unwillingly, a predisposition to categorize a speaker of a 'parallel' language as being part of a particular group of people. The way one sees a speaker using it is paramount, and needs to be transposed into the target language. This paper will conduct an analysis of American slang – its use, perception and the image it gives of its speakers – and its translation into French, using the novel Is Everyone Hanging Out Without Me? (and other concerns) by way of example. In her autobiography/personal essay book, comedy writer, actress and author Mindy Kaling speaks with a very familiar English, including slang, which participates in the construction of her own voice and style, and enables a deeper connection with her readers.

Keywords: translation, English, slang, French

Procedia PDF Downloads 314

Authors: Abobaker Mohammed Qasem Farhan, Shangming Yang, Mohammed Al-Nehari

Abstract:

Pneumonia is a kind of lung disease that creates congestion in the chest. Such pneumonic conditions lead to loss of life of the severity of high congestion. Pneumonic lung disease is caused by viral pneumonia, bacterial pneumonia, or Covidi-19 induced pneumonia. The early prediction and classification of such lung diseases help to reduce the mortality rate. We propose the automatic Computer-Aided Diagnosis (CAD) system in this paper using the deep learning approach. The proposed CAD system takes input from raw computerized tomography (CT) scans of the patient's chest and automatically predicts disease classification. We designed the Hybrid Deep Learning Algorithm (HDLA) to improve accuracy and reduce processing requirements. The raw CT scans have pre-processed first to enhance their quality for further analysis. We then applied a hybrid model that consists of automatic feature extraction and classification. We propose the robust 2D Convolutional Neural Network (CNN) model to extract the automatic features from the pre-processed CT image. This CNN model assures feature learning with extremely effective 1D feature extraction for each input CT image. The outcome of the 2D CNN model is then normalized using the Min-Max technique. The second step of the proposed hybrid model is related to training and classification using different classifiers. The simulation outcomes using the publically available dataset prove the robustness and efficiency of the proposed model compared to state-of-art algorithms.

Keywords: CT scan, Covid-19, deep learning, image processing, lung disease classification

Procedia PDF Downloads 149

Authors: Rana Alaa El-Deen Ahmed, M. Elemam Shehab, Shereen Morsy, Nermeen Mekawie

Abstract:

With the growing popularity and acceptance of e-commerce platforms, users face an ever increasing burden in actually choosing the right product from the large number of online offers. Thus, techniques for personalization and shopping guides are needed by users. For a pleasant and successful shopping experience, users need to know easily which products to buy with high confidence. Since selling a wide variety of products has become easier due to the popularity of online stores, online retailers are able to sell more products than a physical store. The disadvantage is that the customers might not find products they need. In this research the customer will be able to find the products he is searching for, because recommender systems are used in some ecommerce web sites. Recommender system learns from the information about customers and products and provides appropriate personalized recommendations to customers to find the needed product. In this paper eleven classification algorithms are comparatively tested to find the best classifier fit for consumer online shopping attitudes and behavior in the experimented dataset. The WEKA knowledge analysis tool, which is an open source data mining workbench software used in comparing conventional classifiers to get the best classifier was used in this research. In this research by using the data mining tool (WEKA) with the experimented classifiers the results show that decision table and filtered classifier gives the highest accuracy and the lowest accuracy classification via clustering and simple cart.

Keywords: classification, data mining, machine learning, online shopping, WEKA

Procedia PDF Downloads 347

Authors: Sukhvinder Singh

Abstract:

Punjabi language is related to the languages of New Indo-Aryan group that, in turn, is related to the branch of Indo-European language family. Punjabi language covers the areas of Western part (that is in Pakistan) and Eastern part (the Punjab state, Haryana, Delhi Himachal and J&K) and abroad (particularly Canada, USA, U.K. and Arab Emirates), where it is spoken widely. Besides India and Pakistan, Punjabi is the third language spoken in Canada after English, French having more than one hundred millions speakers worldwide. It is the fourth language spoken in Canada after English, French, and Chinese. It is also being taught as second language in most of the community school of British Columbia. The total number of Punjabi speakers is more than one hundred millions including India, Pakistan and abroad. Punjabi has a long tradition of linguistic tradition. A large number of scholars have studied Punjabi at different linguistic levels. Various studies are devoted to its special phonological characteristics, especially the tone, which has now started disappearing in favour of aspiration, a rare example of a language change in progress in its reversal direction. This process of language change in progress in reversal is dealt with in this paper a change towards a loss of linguistic indigeneity. The tone being a distinctive linguistic feature of Punjabi language is getting lost due to the increasing influence of Hindi and English particularly in the speech Urban Punjabi and Punjabi settled abroad. In this paper, an attempt has been made to discuss the sociolinguistics and sociology of Punjabi language and Punjab to trace the initiation and progression of this change towards a loss of Linguistic Indigeneity.

Keywords: language change in reversal, reaspiration, detonalization, new Indo-Aryan group

Procedia PDF Downloads 170