Search results for: deep work

Authors: Vishnu Pratap Singh Kirar, Madhuri Saxena

Abstract:

Advancement in sequence data generation technologies is churning out voluminous omics data and posing a massive challenge to annotate the biological functional features. With so much data available, the use of machine learning methods and tools to make novel inferences has become obvious. Machine learning methods have been successfully applied to a lot of disciplines, including computational biology and bioinformatics. Researchers in computational biology are interested to develop novel machine learning frameworks to classify the huge amounts of biological data. In this proposal, it plan to employ novel machine learning approaches to aid the understanding of how apparently innocuous mutations (in intergenic DNA and at synonymous sites) cause diseases. We are also interested in discovering novel functional sites in the genome and mutations in which can affect a phenotype of interest.

Keywords: genome wide association studies (GWAS), next generation sequencing (NGS), deep learning, omics

Procedia PDF Downloads 97

Authors: Jagtar Singh, Kulwinder Singh

Abstract:

Slurry erosion is a type of erosion wherein material is removed from the target surface due to impingement of solid particles entrained in liquid medium. Slurry erosion performance of deep cryogenic treatment on impeller steel SS 316 L has been investigated. Slurry collected from an actual irrigation pump used as the abrasive media in an erosion test rig. An attempt has been made to study the effect of velocity of fluid and impingement angle by constant concentration (ppm) on the slurry erosion behavior of these cryotreated steels under different experimental conditions. The slurry erosion wear analysis of cryotreated and untreated steels was done. The slurry erosion performance of cryotreated SS 316L impeller steel has been found to superior to that of untreated steel. Metallurgical investigation, hardness as well as %age of carbide in both types of steel was also investigated.

Keywords: deep cryogenic treatment, impeller, Irrigation pumps SS316L, slurry erosion

Procedia PDF Downloads 392

Authors: Ksheeraj Sai Vepuri, Nada Attar

Abstract:

We as humans use words with accompanying visual and facial cues to communicate effectively. Classifying facial emotion using computer vision methodologies has been an active research area in the computer vision field. In this paper, we propose a simple method for facial expression recognition that enhances accuracy. We tested our method on the FER-2013 dataset that contains static images. Instead of using Histogram equalization to preprocess the dataset, we used Unsharp Mask to emphasize texture and details and sharpened the edges. We also used ImageDataGenerator from Keras library for data augmentation. Then we used Convolutional Neural Networks (CNN) model to classify the images into 7 different facial expressions, yielding an accuracy of 69.46% on the test set. Our results show that using image preprocessing such as the sharpening technique for a CNN model can improve the performance, even when the CNN model is relatively simple.

Keywords: facial expression recognittion, image preprocessing, deep learning, CNN

Procedia PDF Downloads 143

Authors: Tolulope Aremu

Abstract:

This paper is based on the idea of using deep learning methodology for optimizing production yield by tuning a few key process parameters in a manufacturing environment. The study was explicitly on how to maximize production yield and minimize operational costs by utilizing advanced neural network models, specifically Long Short-Term Memory and Convolutional Neural Networks. These models were implemented using Python-based frameworks—TensorFlow and Keras. The targets of the research are the precision molding processes in which temperature ranges between 150°C and 220°C, the pressure ranges between 5 and 15 bar, and the material flow rate ranges between 10 and 50 kg/h, which are critical parameters that have a great effect on yield. A dataset of 1 million production cycles has been considered for five continuous years, where detailed logs are present showing the exact setting of parameters and yield output. The LSTM model would model time-dependent trends in production data, while CNN analyzed the spatial correlations between parameters. Models are designed in a supervised learning manner. For the model's loss, an MSE loss function is used, optimized through the Adam optimizer. After running a total of 100 training epochs, 95% accuracy was achieved by the models recommending optimal parameter configurations. Results indicated that with the use of RSM and DOE traditional methods, there was an increase in production yield of 12%. Besides, the error margin was reduced by 8%, hence consistent quality products from the deep learning models. The monetary value was annually around $2.5 million, the cost saved from material waste, energy consumption, and equipment wear resulting from the implementation of optimized process parameters. This system was deployed in an industrial production environment with the help of a hybrid cloud system: Microsoft Azure, for data storage, and the training and deployment of their models were performed on Google Cloud AI. The functionality of real-time monitoring of the process and automatic tuning of parameters depends on cloud infrastructure. To put it into perspective, deep learning models, especially those employing LSTM and CNN, optimize the production yield by fine-tuning process parameters. Future research will consider reinforcement learning with a view to achieving further enhancement of system autonomy and scalability across various manufacturing sectors.

Keywords: production yield optimization, deep learning, tuning of process parameters, LSTM, CNN, precision manufacturing, TensorFlow, Keras, cloud infrastructure, cost saving

Procedia PDF Downloads 29

Authors: Fatemeh Saadat Ghareh Bagh, Srimanta Ray, Jerald Lalman

Abstract:

Lignin, the largest inventory of organic carbon with a high caloric energy value is a major component in woody and non-woody biomass. In pulping mills, a large amount of the lignin is burned for energy. At the same time, the phenolic structure of lignin enables it to be converted to value-added compounds.This study has focused on extracting lignin from black liquor using deep eutectic solvents (DESs). Therefore, three choline chloride (ChCl)-DESs paired with lactic acid (LA) (1:11), oxalic acid.2H₂O (OX) (1:4), and malic acid (MA) (1:3) were synthesized at 90oC and atmospheric pressure. The kinetics of lignin recovery from black liquor using DES was investigated at three moderate temperatures (338, 353, and 368 K) at time intervals from 30 to 210 min. The extracted lignin (acid soluble lignin plus Klason lignin) was characterized by Fourier transform infrared spectroscopy (FTIR). The FTIR studies included comparing the extracted lignin with a model Kraft lignin. The extracted lignin was characterized spectrophotometrically to determine the acid soluble lignin (ASL) [TAPPI UM 250] fraction and Klason lignin was determined gravimetrically using TAPPI T 222 om02. The lignin extraction reaction using DESs was modeled by first-order reaction kinetics and the activation energy of the process was determined. The ChCl:LA-DES recovered lignin was 79.7±2.1% at 368K and a DES:BL ratio of 4:1 (v/v). The quantity of lignin extracted for the control solvent, [emim][OAc], was 77.5+2.2%. The activation energy measured for the LA-DES system was 22.7 KJ mol⁻¹, while the activation energy for the OX-DES and MA-DES systems were 7.16 KJ·mol⁻¹ and 8.66 KJ·mol⁻¹ when the total lignin recovery was 75.4 ±0.9% and 62.4 ±1.4, % respectively.

Keywords: black liquor, deep eutectic solvents, kinetics, lignin

Procedia PDF Downloads 147

Authors: Anjalika Fernando, Banuka Athuraliya

Abstract:

This paper presents a comprehensive approach for converting Sinhala Sign Language (SSL) into grammatically correct sentences using Natural Language Processing (NLP) techniques in real-time. While previous studies have explored various aspects of SSL translation, the research gap lies in the absence of grammar checking for SSL. This work aims to bridge this gap by proposing a two-stage methodology that leverages deep learning models to detect signs and translate them into coherent sentences, ensuring grammatical accuracy. The first stage of the approach involves the utilization of a Long Short-Term Memory (LSTM) deep learning model to recognize and interpret SSL signs. By training the LSTM model on a dataset of SSL gestures, it learns to accurately classify and translate these signs into textual representations. The LSTM model achieves a commendable accuracy rate of 94%, demonstrating its effectiveness in accurately recognizing and translating SSL gestures. Building upon the successful recognition and translation of SSL signs, the second stage of the methodology focuses on improving the grammatical correctness of the translated sentences. The project employs a Neural Machine Translation (NMT) architecture, consisting of an encoder and decoder with LSTM components, to enhance the syntactical structure of the generated sentences. By training the NMT model on a parallel corpus of Sinhala wrong sentences and their corresponding grammatically correct translations, it learns to generate coherent and grammatically accurate sentences. The NMT model achieves an impressive accuracy rate of 98%, affirming its capability to produce linguistically sound translations. The proposed approach offers significant contributions to the field of SSL translation and grammar correction. Addressing the critical issue of grammar checking, it enhances the usability and reliability of SSL translation systems, facilitating effective communication between hearing-impaired and non-sign language users. Furthermore, the integration of deep learning techniques, such as LSTM and NMT, ensures the accuracy and robustness of the translation process. This research holds great potential for practical applications, including educational platforms, accessibility tools, and communication aids for the hearing-impaired. Furthermore, it lays the foundation for future advancements in SSL translation systems, fostering inclusive and equal opportunities for the deaf community. Future work includes expanding the existing datasets to further improve the accuracy and generalization of the SSL translation system. Additionally, the development of a dedicated mobile application would enhance the accessibility and convenience of SSL translation on handheld devices. Furthermore, efforts will be made to enhance the current application for educational purposes, enabling individuals to learn and practice SSL more effectively. Another area of future exploration involves enabling two-way communication, allowing seamless interaction between sign-language users and non-sign-language users.In conclusion, this paper presents a novel approach for converting Sinhala Sign Language gestures into grammatically correct sentences using NLP techniques in real time. The two-stage methodology, comprising an LSTM model for sign detection and translation and an NMT model for grammar correction, achieves high accuracy rates of 94% and 98%, respectively. By addressing the lack of grammar checking in existing SSL translation research, this work contributes significantly to the development of more accurate and reliable SSL translation systems, thereby fostering effective communication and inclusivity for the hearing-impaired community

Keywords: Sinhala sign language, sign Language, NLP, LSTM, NMT

Procedia PDF Downloads 104

Authors: Tonislav Ivanov, Oleksii Nedashkivskyi, Denis Babeshko, Vadim Pinskiy, Matthew Putman

Abstract:

Training a machine learning model for object detection that generalizes well is known to benefit from a training dataset with diverse examples. However, training datasets usually contain many repeats of common examples of a class and lack rarely seen examples. This is due to the process commonly used during human annotation where a person would proceed sequentially through a list of images labeling a sufficiently high total number of examples. Instead, the method presented involves an active process where, after the initial labeling of several images is completed, the next subset of images for labeling is selected by an algorithm. This process of algorithmic image selection and manual labeling continues in an iterative fashion. The algorithm used for the image selection is a deep learning algorithm, based on the U-shaped architecture, which quantifies the presence of unseen data in each image in order to find images that contain the most novel examples. Moreover, the location of the unseen data in each image is highlighted, aiding the labeler in spotting these examples. Experiments performed using semiconductor wafer data show that labeling a subset of the data, curated by this algorithm, resulted in a model with a better performance than a model produced from sequentially labeling the same amount of data. Also, similar performance is achieved compared to a model trained on exhaustive labeling of the whole dataset. Overall, the proposed approach results in a dataset that has a diverse set of examples per class as well as more balanced classes, which proves beneficial when training a deep learning model.

Keywords: computer vision, deep learning, object detection, semiconductor

Procedia PDF Downloads 136

Authors: Tracy B.E. Omorogiuwa

Abstract:

Understanding the role of social work in improving societal well-being cannot be separated from the place of field education, which is an integral aspect of social work education. Field learning provides students with knowledge and opportunities to experience solving issues in the field and giving them a clue of the practice situation. Despite being a crucial component in social work curriculum, field education occupies a large space in learning outcome, given the issues and challenges pertaining to its purpose and significance in the society. The drive of this paper is to provide insight on the specific ways in which field education has been conceived, realized and valued in the society. Emphasis is on the significance of field instruction; the link with classroom learning; and the structure of field experience in social work education. Given documented analysis and experience, this study intends to contribute to the development of social work curriculum, by analyzing the pattern, issues and challenges fronting the social work field education in the University of Benin, Nigeria.

Keywords: challenges, curriculum, field education, social work education

Procedia PDF Downloads 297

Authors: Djamel Fawzi Hadj Sadok, Marrone Silvério Melo Dantas Pedro Henrique Dreyer, Gabriel Fonseca Reis de Souza, Daniel Bezerra, Ricardo Souza, Silvia Lins, Judith Kelner

Abstract:

A radio base station (RBS), part of the radio access network, is a particular type of equipment that supports the connection between a wide range of cellular user devices and an operator network access infrastructure. Nowadays, most of the RBS maintenance is carried out manually, resulting in a time consuming and costly task. A suitable candidate for RBS maintenance automation is repairing faulty links between devices caused by missing or unplugged connectors. A suitable candidate for RBS maintenance automation is repairing faulty links between devices caused by missing or unplugged connectors. This paper proposes and compares two deep learning solutions to identify attached RJ45 connectors on network ports. We named connector detection, the solution based on object detection, and connector classification, the one based on object classification. With the connector detection, we get an accuracy of 0:934, mean average precision 0:903. Connector classification, get a maximum accuracy of 0:981 and an AUC of 0:989. Although connector detection was outperformed in this study, this should not be viewed as an overall result as connector detection is more flexible for scenarios where there is no precise information about the environment and the possible devices. At the same time, the connector classification requires that information to be well-defined.

Keywords: radio base station, maintenance, classification, detection, deep learning, automation

Procedia PDF Downloads 200

Authors: Chandan Deep Singh, Jaimal Singh Khamba, Harleen Kaur

Abstract:

Eminence, charge, deliverance, modernism, and awareness underlie most manufacturing strategic plan today. Firms have traditionally pursued the above tasks through the implementation of advanced technologies and manufacturing practices, such as Reverse Engineering, Value Engineering, worker empowerment, etc. Recent developments in industry suggest the materialization of another route to manufacturing brilliance, that is, there is an increasing focus by industry regulators and professional bodies on the need to stimulate innovation in a broad range of manufacturing competencies. By ‘competencies’ we mean the methods, equipment and expertise that can be developed as a leading capability in one market sector or application and have real potential to be applied successfully across other sectors or applications as well. Further, competencies are the ability to apply or use a set of related knowledge, skills, and abilities to perform 'critical work functions' or tasks in a defined work setting. Competencies often serve as the basis for skill standards that specify the level of knowledge, skills, and abilities required for success in the workplace as well as potential measurement criteria for assessing competency attainment. The present research is so designed to come up to the level of the expectations of the industrialists, policy makers, designers of the competencies, specially, the manufacturing competencies upon which the whole strategic success of the industry depends.

Keywords: manufacturing competency, strategic success, manufacturing excellence, competitive strategy

Procedia PDF Downloads 570

Authors: Roshani S. Golhar, Neerajkumar S. Sathawane, Snehal Dongre

Abstract:

Background and objectives: As the, cardiovascular illnesses increasing and becoming cause of mortality worldwide, killing around lot of people each year. Arrhythmia is a type of cardiac illness characterized by a change in the linearity of the heartbeat. The goal of this study is to develop novel deep learning algorithms for successfully interpreting arrhythmia using a single second segment. Because the ECG signal indicates unique electrical heart activity across time, considerable changes between time intervals are detected. Such variances, as well as the limited number of learning data available for each arrhythmia, make standard learning methods difficult, and so impede its exaggeration. Conclusions: The proposed method was able to outperform several state-of-the-art methods. Also proposed technique is an effective and convenient approach to deep learning for heartbeat interpretation, that could be probably used in real-time healthcare monitoring systems

Keywords: electrocardiogram, ECG classification, neural networks, convolutional neural networks, portable document format

Procedia PDF Downloads 69

Authors: Yifan Fan, Xudong Luo, Pingping Lin

Abstract:

An essential task in the field of artificial intelligence is to allow computers to interact with people through natural language. Therefore, researches such as virtual assistants and dialogue systems have received widespread attention from industry and academia. The response generation plays a crucial role in dialogue systems, so to push forward the research on this topic, this paper surveys various methods for response generation. We sort out these methods into three categories. First one includes finite state machine methods, framework methods, and instance methods. The second contains full-text indexing methods, ontology methods, vast knowledge base method, and some other methods. The third covers retrieval methods and generative methods. We also discuss some hybrid methods based knowledge and deep learning. We compare their disadvantages and advantages and point out in which ways these studies can be improved further. Our discussion covers some studies published in leading conferences such as IJCAI and AAAI in recent years.

Keywords: deep learning, generative, knowledge, response generation, retrieval

Procedia PDF Downloads 134

Authors: Mohamed Ahmed Khalil

Abstract:

The number of deep well anode ground beds (GBs) have been retrieved due to unoperated anode chains. New identical magnetite anode chains (MAC) have been installed at Raslanuf complex impressed current Cathodic protection (ICCP) system, distributed at different plants (Utility, ethylene and polyethylene). All problems associated with retrieving and installation of MACs have been discussed, rectified and presented. All GB-associated severely corroded wellhead casings were well maintained and/or replaced by new fabricated and modified ones. The main cause of the wellhead casing's severe internal corrosion was discussed and the conducted remedy action to overcome future corrosion problems is presented. All GB-connected anode junction boxes (AJBs) and shunts were closely inspected, maintained and necessary replacement and/or modifications were carried out on shunts. All damaged GB concrete foundations (CF) have been inspected and completely replaced. All GB-associated Transformer-Rectifiers Units (TRU) were subjected to thorough inspection and necessary maintenance was performed on each individual TRU. After completion of all MACs and TRU maintenance activities, each cathodic protection station (CPS) has been re-operated, alternative current (AC), direct current (DC), voltage and structure to soil potential (S/P) measurements have been conducted, recorded and all obtained test results are presented. DC current outputs have been adjusted and DC current outputs of each MAC have been recorded for each GB AJB.

Keywords: magnetite anodes, deep well, ground beds, cathodic protection, transformer rectifier, impressed current, junction boxes

Procedia PDF Downloads 119

Authors: Mehmet Halkaci, Ekrem Öztürk, Mevlüt Türköz, H. Selçuk Halkacı

Abstract:

Hydromechanical deep drawing (HMD) process is an advanced manufacturing process used to form deep parts with only one forming step. In this process, sheet metal blank can be drawn deeper by means of fluid pressure acting on sheet surface in the opposite direction of punch movement. High limiting drawing ratio, good surface quality, less springback characteristic and high dimensional accuracy are some of the advantages of this process. The performance of the HMD process is affected by various process parameters such as fluid pressure, blank holder force, punch-die radius, pre-bulging pressure and height, punch diameter, friction between sheet-die and sheet-punch. The fluid pressure and bank older force are the main loading parameters and affect the formability of HMD process significantly. The punch diameter also influences the limiting drawing ratio (the ratio of initial sheet diameter to punch diameter) of the sheet metal blank. In this research, optimal loading (fluid pressure and blank holder force) profiles were determined for AA 5754-O sheet material through fuzzy control algorithm developed in previous study using LS-DYNA finite element analysis (FEA) software. In the preceding study, the fuzzy control algorithm was developed utilizing geometrical criteria such as thinning and wrinkling. In order to obtain the final desired part with the developed algorithm in terms of the punch diameter requested, the effect of punch diameter, which is the one of the process parameters, on loading profiles was investigated separately using blank thickness of 1 mm. Thus, the practicality of the previously developed fuzzy control algorithm with different punch diameters was clarified. Also, thickness distributions of the sheet metal blank along a curvilinear distance were compared for the FEA in which different punch diameters were used. Consequently, it was found that the use of different punch diameters did not affect the optimal loading profiles too much.

Keywords: Finite Element Analysis (FEA), fuzzy control, hydromechanical deep drawing, optimal loading profiles, punch diameter

Procedia PDF Downloads 431

Authors: Lina Vyas, Stuti Rawat

Abstract:

The COVID-19 outbreak has forced many employees to extensively adopt remote work or, widely known as work from home (WFH) arrangements. During the last two years, both employers and employees have had the opportunity to be increasingly aware of the benefits and drawbacks of WFH. Likewise, it gained the attention of academics from various schools of thought who have been interested in the future of work practices and work-life balance. Additionally, employees might also have various demands regarding their work practices after the pandemic. This study explores the potential of hybrid ways of working in the post-pandemic period by comparing first-timers who (sometimes or always) worked from home during the pandemic with those who did not, in terms of the aspects of work-life balance, work-life interference, job performance and willingness to work from home after the pandemic. The quantitative research approach was adopted. Data were collected via an online questionnaire from employees working from home in Hong Kong during the pandemic. There were one thousand three hundred and twenty-eight responses, but only 1,235 respondents experienced working from home during the pandemic. The findings reveal that 72.2% never had or hardly experienced work from home prior to the pandemic. There were statistically significant differences between first-timers and non-first-timers in work-life balance and work-life interference. The study also found that first-timers who were always working from home during the pandemic would prefer having longer WFH after the pandemic than those who were sometimes working from home. These results would serve as a basis for policy development, enabling policymakers to design appropriate HR policies and amend them to meet the current context of actual employee needs.

Keywords: hybrid working mode, remote working, work from home, work-life balance, workplace

Procedia PDF Downloads 107

Authors: Ameer A. Jebur, William Atherton, Rafid M. Alkhaddar, Edward Loffill

Abstract:

Understanding the behaviour of the piles under the action of the independent lateral loads and the precise prediction of the capacity of piles subjected to different lateral loads are vital topics in foundation design and analysis. Moreover, the laterally loaded behaviour of deep foundations penetrated in cohesive and non-cohesive soils is basically analysed by the Winkler Model (beam on elastic foundation), in which the interaction between the pile embedded depth and contacted soil is simulated by nonlinear p–y curves. The presence of many approaches to interpret the behaviour of soil-pile interaction has resulted in numerous outputs and indicates that no general approach has yet been adopted. The current study presents the result of numerical modelling of the behaviour of steel tubular piles (25.4mm) outside diameter with various embedment depth-to-diameter ratios (L/d) embedded in a sand calibrated chamber of known relative density. The study revealed that the shear strength parameters of the sand specimens and the (L/d) ratios are the most significant factor influencing the response of the pile and its capacity while taking into consideration the complex interaction between the pile and soil. Good agreement has been achieved when comparing the application of this modelling approach with experimental physical modelling carried out by another researcher.

Keywords: deep foundations, slenderness ratio, soil-pile interaction, winkler model (beam on elastic foundation), non-cohesive soil

Procedia PDF Downloads 299

Authors: S. Jagadeesh Kumar

Abstract:

This expose presents a luculent of recent machine learning practices as employed in the modern and as pertinent to prospective automatic speech recognition schemes. The aspiration is to promote additional traverse ablution among the machine learning and automatic speech recognition factions that have transpired in the precedent. The manuscript is structured according to the chief machine learning archetypes that are furthermore trendy by now or have latency for building momentous hand-outs to automatic speech recognition expertise. The standards offered and convoluted in this article embraces adaptive and multi-task learning, active learning, Bayesian learning, discriminative learning, generative learning, supervised and unsupervised learning. These learning archetypes are aggravated and conferred in the perspective of automatic speech recognition tools and functions. This manuscript bequeaths and surveys topical advances of deep learning and learning with sparse depictions; further limelight is on their incessant significance in the evolution of automatic speech recognition.

Keywords: automatic speech recognition, deep learning methods, machine learning archetypes, Bayesian learning, supervised and unsupervised learning

Procedia PDF Downloads 447

Authors: Yijun Shao, Yan Cheng, Rashmee U. Shah, Charlene R. Weir, Bruce E. Bray, Qing Zeng-Treitler

Abstract:

Deep neural network (DNN) models are being explored in the clinical domain, following the recent success in other domains such as image recognition. For clinical adoption, outcome prediction models require explanation, but due to the multiple non-linear inner transformations, DNN models are viewed by many as a black box. In this study, we developed a deep neural network model for predicting 1-year mortality of patients who underwent major cardio vascular procedures (MCVPs), using temporal image representation of past medical history as input. The dataset was obtained from the electronic medical data warehouse administered by Veteran Affairs Information and Computing Infrastructure (VINCI). We identified 21,355 veterans who had their first MCVP in 2014. Features for prediction included demographics, diagnoses, procedures, medication orders, hospitalizations, and frailty measures extracted from clinical notes. Temporal variables were created based on the patient history data in the 2-year window prior to the index MCVP. A temporal image was created based on these variables for each individual patient. To generate the explanation for the DNN model, we defined a new concept called impact score, based on the presence/value of clinical conditions’ impact on the predicted outcome. Like (log) odds ratio reported by the logistic regression (LR) model, impact scores are continuous variables intended to shed light on the black box model. For comparison, a logistic regression model was fitted on the same dataset. In our cohort, about 6.8% of patients died within one year. The prediction of the DNN model achieved an area under the curve (AUC) of 78.5% while the LR model achieved an AUC of 74.6%. A strong but not perfect correlation was found between the aggregated impact scores and the log odds ratios (Spearman’s rho = 0.74), which helped validate our explanation.

Keywords: deep neural network, temporal data, prediction, frailty, logistic regression model

Procedia PDF Downloads 153

Authors: Qasim Munye, Juned Islam, Haseeb Qureshi, Syed Jung

Abstract:

Lung nodules are commonly identified in computed tomography (CT) scans by experienced radiologists at a relatively late stage. Early diagnosis can greatly increase survival. We propose using a pix2pix conditional generative adversarial network to generate realistic images simulating early-stage lung nodule growth. We have applied deep images prior to 2341 slices from 895 computed tomography (CT) scans from the Lung Image Database Consortium (LIDC) dataset to generate pseudo-healthy medical images. From these images, 819 were chosen to train a pix2pix network. We observed that for most of the images, the pix2pix network was able to generate images where the nodule increased in size and intensity across epochs. To evaluate the images, 400 generated images were chosen at random and shown to a medical student beside their corresponding original image. Of these 400 generated images, 384 were defined as satisfactory - meaning they resembled a nodule and were visually similar to the corresponding image. We believe that this generated dataset could be used as training data for neural networks to detect lung nodules at an early stage or to improve the accuracy of such networks. This is particularly significant as datasets containing the growth of early-stage nodules are scarce. This project shows that the combination of deep image prior and generative models could potentially open the door to creating larger datasets than currently possible and has the potential to increase the accuracy of medical classification tasks.

Keywords: medical technology, artificial intelligence, radiology, lung cancer

Procedia PDF Downloads 66

Authors: Farzad Hosseini Hossein Abadi, Cristina Prieto Sierra, Cesar Álvarez Díaz

Abstract:

Predicting streamflow accurately in flashy catchments prone to floods is a major research and operational challenge in hydrological modeling. Recent advancements in deep learning, particularly Long Short-Term Memory (LSTM) networks, have shown to be promising in achieving accurate hydrological predictions at daily and hourly time scales. In this work, a multi-timescale LSTM (MTS-LSTM) network was applied to the context of regional hydrological predictions at an hourly time scale in flashy catchments. The case study includes 40 catchments allocated in the Basque Country, north of Spain. We explore the impact of hyperparameters on the performance of streamflow predictions given by regional deep learning models through systematic hyperparameter tuning - where optimal regional values for different catchments are identified. The results show that predictions are highly accurate, with Nash-Sutcliffe (NSE) and Kling-Gupta (KGE) metrics values as high as 0.98 and 0.97, respectively. A principal component analysis reveals that a hyperparameter related to the length of the input sequence contributes most significantly to the prediction performance. The findings suggest that input sequence lengths have a crucial impact on the model prediction performance. Moreover, employing catchment-scale analysis reveals distinct sequence lengths for individual basins, highlighting the necessity of customizing this hyperparameter based on each catchment’s characteristics. This aligns with well known “uniqueness of the place” paradigm. In prior research, tuning the length of the input sequence of LSTMs has received limited focus in the field of streamflow prediction. Initially it was set to 365 days to capture a full annual water cycle. Later, performing limited systematic hyper-tuning using grid search, revealed a modification to 270 days. However, despite the significance of this hyperparameter in hydrological predictions, usually studies have overlooked its tuning and fixed it to 365 days. This study, employing a simultaneous systematic hyperparameter tuning approach, emphasizes the critical role of input sequence length as an influential hyperparameter in configuring LSTMs for regional streamflow prediction. Proper tuning of this hyperparameter is essential for achieving accurate hourly predictions using deep learning models.

Keywords: LSTMs, streamflow, hyperparameters, hydrology

Procedia PDF Downloads 69

Authors: Ja-Keoung Koo, Kensuke Nakamura, Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Byung-Woo Hong

Abstract:

The machine learning techniques based on a convolutional neural network (CNN) have been actively developed and successfully applied to a variety of image analysis tasks including reconstruction, noise reduction, resolution enhancement, segmentation, motion estimation, object recognition. The classical visual information processing that ranges from low level tasks to high level ones has been widely developed in the deep learning framework. It is generally considered as a challenging problem to derive visual interpretation from high dimensional imagery data. A CNN is a class of feed-forward artificial neural network that usually consists of deep layers the connections of which are established by a series of non-linear operations. The CNN architecture is known to be shift invariant due to its shared weights and translation invariance characteristics. However, it is often computationally intractable to optimize the network in particular with a large number of convolution layers due to a large number of unknowns to be optimized with respect to the training set that is generally required to be large enough to effectively generalize the model under consideration. It is also necessary to limit the size of convolution kernels due to the computational expense despite of the recent development of effective parallel processing machinery, which leads to the use of the constantly small size of the convolution kernels throughout the deep CNN architecture. However, it is often desired to consider different scales in the analysis of visual features at different layers in the network. Thus, we propose a CNN model where different sizes of the convolution kernels are applied at each layer based on the random projection. We apply random filters with varying sizes and associate the filter responses with scalar weights that correspond to the standard deviation of the random filters. We are allowed to use large number of random filters with the cost of one scalar unknown for each filter. The computational cost in the back-propagation procedure does not increase with the larger size of the filters even though the additional computational cost is required in the computation of convolution in the feed-forward procedure. The use of random kernels with varying sizes allows to effectively analyze image features at multiple scales leading to a better generalization. The robustness and effectiveness of the proposed CNN based on random kernels are demonstrated by numerical experiments where the quantitative comparison of the well-known CNN architectures and our models that simply replace the convolution kernels with the random filters is performed. The experimental results indicate that our model achieves better performance with less number of unknown weights. The proposed algorithm has a high potential in the application of a variety of visual tasks based on the CNN framework. Acknowledgement—This work was supported by the MISP (Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by IITP, and NRF-2014R1A2A1A11051941, NRF2017R1A2B4006023.

Keywords: deep learning, convolutional neural network, random kernel, random projection, dimensionality reduction, object recognition

Procedia PDF Downloads 289

Authors: Khalil Bachiri, Ali Yahyaouy, Nicoleta Rogovschi

Abstract:

Improved user experience is a goal of contemporary recommender systems. Recommender systems are starting to incorporate reinforcement learning since it easily satisfies this goal of increasing a user’s reward every session. In this paper, we examine the most effective Reinforcement Learning agent tactics on the Movielens (1M) dataset, balancing precision and a variety of recommendations. The absence of variability in final predictions makes simplistic techniques, although able to optimize ranking quality criteria, worthless for consumers of the recommendation system. Utilizing the stochasticity of Leonard-Ornstein processes, our suggested strategy encourages the agent to investigate its surroundings. Research demonstrates that raising the NDCG (Discounted Cumulative Gain) and HR (HitRate) criterion without lowering the Ornstein-Uhlenbeck process drift coefficient enhances the diversity of suggestions.

Keywords: recommender systems, reinforcement learning, deep learning, DDPG, Leonard-Ornstein process

Procedia PDF Downloads 142

Authors: Migena Mana, Ahmed Khalid Syed, Abdul Malik, Nikhil Cherian

Abstract:

In recent years, advancements in deep learning enabled researchers to tackle the problem of self-driving cars. Car companies use huge datasets to train their deep learning models to make autonomous cars a reality. However, this approach has certain drawbacks in that the state space of possible actions for a car is so huge that there cannot be a dataset for every possible road scenario. To overcome this problem, the concept of reinforcement learning (RL) is being investigated in this research. Since the problem of autonomous driving can be modeled in a simulation, it lends itself naturally to the domain of reinforcement learning. The advantage of this approach is that we can model different and complex road scenarios in a simulation without having to deploy in the real world. The autonomous agent can learn to drive by finding the optimal policy. This learned model can then be easily deployed in a real-world setting. In this project, we focus on three RL algorithms: Q-learning, Deep Deterministic Policy Gradient (DDPG), and Proximal Policy Optimization (PPO). To model the environment, we have used TORCS (The Open Racing Car Simulator), which provides us with a strong foundation to test our model. The inputs to the algorithms are the sensor data provided by the simulator such as velocity, distance from side pavement, etc. The outcome of this research project is a comparative analysis of these algorithms. Based on the comparison, the PPO algorithm gives the best results. When using PPO algorithm, the reward is greater, and the acceleration, steering angle and braking are more stable compared to the other algorithms, which means that the agent learns to drive in a better and more efficient way in this case. Additionally, we have come up with a dataset taken from the training of the agent with DDPG and PPO algorithms. It contains all the steps of the agent during one full training in the form: (all input values, acceleration, steering angle, break, loss, reward). This study can serve as a base for further complex road scenarios. Furthermore, it can be enlarged in the field of computer vision, using the images to find the best policy.

Keywords: autonomous driving, DDPG (deep deterministic policy gradient), PPO (proximal policy optimization), reinforcement learning

Procedia PDF Downloads 147

Authors: Praveen Chandramenon, Andrew Gascoyne, Fideline Tchuenbou-Magaia

Abstract:

Aquaponics offers a simple conclusive solution to the food and environmental crisis of the world. This approach combines the idea of Aquaculture (growing fish) to Hydroponics (growing vegetables and plants in a soilless method). Smart Aquaponics explores the use of smart technology including artificial intelligence and IoT, to assist farmers with better decision making and online monitoring and control of the system. Identification of different growth stages of Swiss Chard plants and predicting its harvest time is found to be important in Aquaponic yield management. This paper brings out the comparative analysis of a standard Aquaponics with a Vermiponics (Aquaponics with worms), which was grown in the controlled environment, by implementing IoT and deep learning-based growth stage segregation and harvest time prediction of Swiss Chards before and after applying an optimal freshwater replenishment. Data collection, Growth stage classification and Harvest Time prediction has been performed with and without water replenishment. The paper discusses the experimental design, IoT and sensor communication with architecture, data collection process, image segmentation, various regression and classification models and error estimation used in the project. The paper concludes with the results comparison, including best models that performs growth stage segregation and harvest time prediction of the Aquaponic and Vermiponic testbed with and without freshwater replenishment.

Keywords: aquaponics, deep learning, internet of things, vermiponics

Procedia PDF Downloads 71

Authors: Witthaya Mekhum, Waleerak Sittisom

Abstract:

This research investigated the effect on occupational health safety and environment at work from metal handicraft using Rattanakosin local wisdom focusing on pollution, accidents, and injuries from work. The sample group in this study included 48 metal handicraft workers in 5 communities by using questionnaires and interview to collect data. The evaluation form TISI 18001 was used to analyze job safety analysis (JSA). The results showed that risk at work reduced after applying the developed model. Banbu Community produces alloy bowl rubbed with stone. The high risk process is melting and hitting process. Before the application, the work risk was 82.71%. After the application of the developed model, the work risk was reduced to 50.61%. Banbart Community produces monk’s food bowl. The high risk process is blow pipe welding. Before the application, the work risk was 93.59%. After the application of the developed model, the work risk was reduced to 48.14%. Bannoen Community produces circle gong. The high risk process is milling process. Before the application, the work risk was 85.18%. After the application of the developed model, the work risk was reduced to 46.91%. Teethong Community produces gold leaf. The high risk process is hitting and spreading process. Before the application, the work risk was 86.42%. After the application of the developed model, the work risk was reduced to 64.19%. Ban Changthong Community produces gold ornament. The high risk process is gold melting process. Before the application, the work risk was 67.90%. After the application of the developed model, the work risk was reduced to 37.03%. It can be concluded that with the application of the developed model, the work risk of 5 communities was reduced in the 3 main groups: (1) Work illness reduced by 16.77%; (2) Pollution from work reduced by 10.31%; (3) Accidents and injuries from work reduced by 15.62%.

Keywords: occupational health, safety, local wisdom, Rattanakosin

Procedia PDF Downloads 440

Authors: Xiaocong Liu, Huazhen Wang, Ting He, Xiaozheng Li, Weihan Zhang, Jian Chen

Abstract:

The utilization of electronic medical record (EMR) data to establish the disease diagnosis model has become an important research content of biomedical informatics. Deep learning can automatically extract features from the massive data, which brings about breakthroughs in the study of EMR data. The challenge is that deep learning lacks semantic knowledge, which leads to impracticability in medical science. This research proposes a method of incorporating lexical-semantic knowledge from abundant entities into a convolutional neural network (CNN) framework for pediatric disease diagnosis. Firstly, medical terms are vectorized into Lexical Semantic Vectors (LSV), which are concatenated with the embedded word vectors of word2vec to enrich the feature representation. Secondly, the semantic distribution of medical terms serves as Semantic Decision Guide (SDG) for the optimization of deep learning models. The study evaluate the performance of LSV-SDG-CNN model on four kinds of Chinese EMR datasets. Additionally, CNN, LSV-CNN, and SDG-CNN are designed as baseline models for comparison. The experimental results show that LSV-SDG-CNN model outperforms baseline models on four kinds of Chinese EMR datasets. The best configuration of the model yielded an F1 score of 86.20%. The results clearly demonstrate that CNN has been effectively guided and optimized by lexical-semantic knowledge, and LSV-SDG-CNN model improves the disease classification accuracy with a clear margin.

Keywords: convolutional neural network, electronic medical record, feature representation, lexical semantics, semantic decision

Procedia PDF Downloads 124

Authors: Ouazoul Abdelouahd, Jemjami Nadia

Abstract:

The concept of well-being at work is one of today's significant challenges in maintaining quality of life and managing psycho-social risks at work. Indeed, work in the sports sector has evolved, and this exponential evolution, marked by increasing demands and psychological, physical, and social challenges, which sometimes exceed the resources of sports actors, influences their sense of well-being at work. Well-being and burnout as antagonists provide information on the quality of working life in sports. The Basic aim of this literature review is to analyze the scientific corpus dealing with the subject of well-being at work in the sports sector while exploring the link between sports burnout and well-being. The results reveal the richness of the conceptual approaches and the difficulties of implementing them. Prospects for future research have, therefore, been put forward.

Keywords: Well-being, quality of life, Burnout, ; psycho-social risk, Sports sector

Procedia PDF Downloads 85

Authors: S. Sharan Sundar, Nithish B. Moudhgalya, Nidhi Bhandari, Vineeth Vijayaraghavan

Abstract:

Automatic Short Answer Grading is one of the prime applications of artificial intelligence in education. Several approaches involving the utilization of selective handcrafted features, graphical matching techniques, concept identification and mapping, complex deep frameworks, sentence embeddings, etc. have been explored over the years. However, keeping in mind the real-world application of the task, these solutions present a slight overhead in terms of computations and resources in achieving high performances. In this work, a simple and effective solution making use of elemental features based on statistical, linguistic properties, and word-based similarity measures in conjunction with tree-based classifiers and regressors is proposed. The results for classification tasks show improvements ranging from 1%-30%, while the regression task shows a stark improvement of 35%. The authors attribute these improvements to the addition of multiple similarity scores to provide ensemble of scoring criteria to the models. The authors also believe the work could reinstate that classical natural language processing techniques and simple machine learning models can be used to achieve high results for short answer grading.

Keywords: artificial intelligence, intelligent systems, natural language processing, text mining

Procedia PDF Downloads 133

Authors: Revoti Prasad Bora, Saurabh Yadav, Nikita Katyal

Abstract:

In the context of educational data mining, the use case of extracting information from images containing both text and diagrams is of high importance. Hence, document analysis requires the extraction of diagrams from such images and processes the text and diagrams separately. To the author’s best knowledge, none among plenty of approaches for extracting tables, figures, etc., suffice the need for real-time processing with high accuracy as needed in multiple applications. In the education domain, diagrams can be of varied characteristics viz. line-based i.e. geometric diagrams, chemical bonds, mathematical formulas, etc. There are two broad categories of approaches that try to solve similar problems viz. traditional computer vision based approaches and deep learning approaches. The traditional computer vision based approaches mainly leverage connected components and distance transform based processing and hence perform well in very limited scenarios. The existing deep learning approaches either leverage YOLO or faster-RCNN architectures. These approaches suffer from a performance-accuracy tradeoff. This paper proposes a U-Net based architecture that formulates the diagram extraction as a segmentation problem. The proposed method provides similar accuracy with a much faster extraction time as compared to the mentioned state-of-the-art approaches. Further, the segmentation mask in this approach allows the extraction of diagrams of irregular shapes.

Keywords: computer vision, deep-learning, educational data mining, faster-RCNN, figure extraction, image segmentation, real-time document analysis, text extraction, U-Net, YOLO

Procedia PDF Downloads 137

Authors: Parthiban Srinivasan

Abstract:

With recent advances in computational power and access to enough data in biosciences, artificial intelligence methods are increasingly being used in drug discovery research. These methods are essentially a series of advanced statistics based exercises that review the past to indicate the likely future. Our goal is to develop a model that accurately predicts biological activity and toxicity parameters for novel compounds. We have compiled a robust library of over 150,000 chemical compounds with different pharmacological properties from literature and public domain databases. The compounds are stored in simplified molecular-input line-entry system (SMILES), a commonly used text encoding for organic molecules. We utilize an automated process to generate an array of numerical descriptors (features) for each molecule. Redundant and irrelevant descriptors are eliminated iteratively. Our prediction engine is based on a portfolio of machine learning algorithms. We found Random Forest algorithm to be a better choice for this analysis. We captured non-linear relationship in the data and formed a prediction model with reasonable accuracy by averaging across a large number of randomized decision trees. Our next step is to apply deep neural network (DNN) algorithm to predict the biological activity and toxicity properties. We expect the DNN algorithm to give better results and improve the accuracy of the prediction. This presentation will review all these prominent machine learning and deep learning methods, our implementation protocols and discuss these techniques for their usefulness in biomedical and health informatics.

Keywords: deep learning, drug discovery, health informatics, machine learning, toxicity prediction

Procedia PDF Downloads 356